pandas_EDA_1_handlingGFF.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EDA란?\n",
    "\n",
    "**Exploratory Data Analysis, 탐색적 데이터 분석**\n",
    "\n",
    "수집한 데이터가 들어왔을 때 이를 다양한 각도에서 관찰하고 만져가며 데이터를 이해하는 과정을 뜻합니다.\n",
    "\n",
    "단순하게는 데이터 각각의 열과 행이 무슨의미인지 확인해보고 <Br>내가 다뤄야할 관심사(변수들)를 테이블로 요약, 그래프화 하는 것부터,<br>\n",
    "좀더 많이 들어가면 변수들간의 관계확인, 분포확인, 통계적 가설 검정을 통해 데이터를 직관적으로 바라보는 과정입니다.\n",
    "\n",
    "이를 통해 \n",
    " * 데이터에 문제가 없는지 확인하고\n",
    " * 데이터의 전체적인 구성을 이해하고\n",
    " * 내가 현 데이터로 앞으로 어떤 분석을 할것인지 생각해보고\n",
    " * 목적변수, 설명변수를 확인하여 앞으로의 분석 모형을 구체화 시킬 수 있습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [Pandas EDA 실전 문제 - GFF포맷 데이터 만져보기]\n",
    "***\n",
    "\n",
    "Bio데이터(GFF파일)을 읽어 `pandas_basic_grammar` 에서 배웠던 점들을 몇가지 활용해보아요.<br>\n",
    "몇가지 질문을 코딩으로 풀어가면서 \"이렇게 쓸수 있구나~\" 를 확인하는 시간이 되었으면 합니다.\n",
    "\n",
    "필요한 데이터:\n",
    " * GFF파일 : `dataset/Homo_sapiens.GRCh38.97.chr.gff3.gz`\n",
    " \n",
    "GFF포맷이 가진 의미는 다음의 페이지에서 참고하세요\n",
    " * [Ensembl GFF설명](https://asia.ensembl.org/info/website/upload/gff.html)\n",
    " * [인실리코젠 인코덤](http://www.incodom.kr/GFF_and_GTF)\n",
    " \n",
    "***"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "우선 실습용 파일이 있는지 확인해봅시다.\n",
    "\n",
    "github에서 다운 받았다면 dataset 폴더에 데이터가 있을 것이구요 없다면 아래의 페이지에서 GFF3를 다운 받으면 됩니다.\n",
    "\n",
    "참고) `wget ftp://ftp.ensembl.org/pub/release-97/gff3/homo_sapiens/Homo_sapiens.GRCh38.97.chr.gff3.gz`"
   ]
  },
  {
   "attachments": {
    "%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.08.45.png": {
     "image/png": "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"
    },
    "%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.10.22.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 파일 읽어 DataFrame으로 저장하기\n",
    "우선 파일을 잘 읽어오기 위해선 내가 읽으려는 파일의 형태를 당연히 잘 파악하고 있어야겠죠,\n",
    "\n",
    "* GFF포맷은 아래 처럼 9개의 열로 구성되어있습니다.\n",
    "![%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.08.45.png](attachment:%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.08.45.png)\n",
    "\n",
    "\n",
    "* 각 9개의 열은 다음과 같은 의미네요\n",
    "![%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.10.22.png](attachment:%E1%84%89%E1%85%B3%E1%84%8F%E1%85%B3%E1%84%85%E1%85%B5%E1%86%AB%E1%84%89%E1%85%A3%E1%86%BA%202019-08-22%2010.10.22.png)\n",
    "\n",
    ".. 크흠.. 복잡해 보이는데, 우리가 알아야 하는 점은 9개의 열로 구성되어있고 각 열은 텝(\\t)으로 분리되어있다는 사실만 알면 읽어 올수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 참고\n",
    "# 압축이 풀려있는 파일을 pandas로 읽기 위한 함수\n",
    "# gunzip Homo_sapiens.GRCh38.97.chr.gff3.gz 명령으로 압축파일을 풀었다면 이 함수를 사용하면 됨.\n",
    "# 그러나 압축파일을 그대로 읽은 것보단 느림.\n",
    "import io\n",
    "import os\n",
    "\n",
    "def read_gff(path):\n",
    "\n",
    "    with open(path, 'r') as f:\n",
    "        lines = [l for l in f if not l.startswith('#')]\n",
    "    return pd.read_csv(\n",
    "        io.StringIO(''.join(lines)),\n",
    "        dtype={'start': int, 'end': int},\n",
    "        sep='\\t',\n",
    "        names=['chr', 'source', 'type',\n",
    "          'start', 'end', 'score', 'strand', 'phase', 'attrs']\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 파일을 읽어 DataFrame으로 읽어 온다.\n",
    "file = 'dataset/Homo_sapiens.GRCh38.97.chr.gff3.gz'\n",
    "\n",
    "df = pd.read_csv(\n",
    "        file, \n",
    "        compression='gzip', \n",
    "        sep='\\t', \n",
    "        names=['chr', 'source', 'type', 'start',\n",
    "               'end', 'score', 'strand', 'phase', 'attrs']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>##gff-version 3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>##sequence-region   1 1 248956422</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>##sequence-region   10 1 133797422</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>##sequence-region   11 1 135086622</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>##sequence-region   12 1 133275309</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>##sequence-region   13 1 114364328</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>##sequence-region   14 1 107043718</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>##sequence-region   15 1 101991189</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>##sequence-region   16 1 90338345</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>##sequence-region   17 1 83257441</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>##sequence-region   18 1 80373285</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>##sequence-region   19 1 58617616</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>##sequence-region   2 1 242193529</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>##sequence-region   20 1 64444167</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>##sequence-region   21 1 46709983</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>##sequence-region   22 1 50818468</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>##sequence-region   3 1 198295559</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>##sequence-region   4 1 190214555</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>##sequence-region   5 1 181538259</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>##sequence-region   6 1 170805979</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>##sequence-region   7 1 159345973</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>##sequence-region   8 1 145138636</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>##sequence-region   9 1 138394717</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>##sequence-region   MT 1 16569</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>##sequence-region   X 1 156040895</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>##sequence-region   Y 2781480 56887902</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>#!genome-build Ensembl GRCh38.p12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>#!genome-version GRCh38</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>#!genome-date 2013-12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>#!genome-build-accession NCBI:GCA_000001405.27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951268</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26586642.0</td>\n",
       "      <td>26586766.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951269</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26587398.0</td>\n",
       "      <td>26587560.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951270</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26591548.0</td>\n",
       "      <td>26591601.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951271</th>\n",
       "      <td>###</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951272</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26594851.0</td>\n",
       "      <td>26634652.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000237917;Name=PARP4P1;biotype=u...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951273</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26594851.0</td>\n",
       "      <td>26634652.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000435945;Parent=gene:ENSG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951274</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26594851.0</td>\n",
       "      <td>26595045.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951275</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26600891.0</td>\n",
       "      <td>26601022.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951276</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26611713.0</td>\n",
       "      <td>26611807.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951277</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26614154.0</td>\n",
       "      <td>26614962.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951278</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26615294.0</td>\n",
       "      <td>26615531.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951279</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26621773.0</td>\n",
       "      <td>26621895.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951280</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26622512.0</td>\n",
       "      <td>26622608.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951281</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26623586.0</td>\n",
       "      <td>26623666.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951282</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26627943.0</td>\n",
       "      <td>26628022.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951283</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26628271.0</td>\n",
       "      <td>26628437.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951284</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26630647.0</td>\n",
       "      <td>26630749.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951285</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26633345.0</td>\n",
       "      <td>26633431.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951286</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26634523.0</td>\n",
       "      <td>26634652.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951287</th>\n",
       "      <td>###</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951288</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26626520.0</td>\n",
       "      <td>26627159.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000231514;Name=CCNQP2;biotype=pr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951289</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26626520.0</td>\n",
       "      <td>26627159.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000435741;Parent=gene:ENSG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951290</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26626520.0</td>\n",
       "      <td>26627159.0</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435741;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951291</th>\n",
       "      <td>###</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951292</th>\n",
       "      <td>Y</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>26626966.0</td>\n",
       "      <td>26627137.0</td>\n",
       "      <td>0.994</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=rank %3D 1;logic_name=firstef</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951293</th>\n",
       "      <td>Y</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>26627457.0</td>\n",
       "      <td>26628186.0</td>\n",
       "      <td>0.997</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=rank %3D 1;logic_name=firstef</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951294</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951295</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000431853;Parent=gene:ENSG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951296</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000431853;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951297</th>\n",
       "      <td>###</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2951298 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    chr  source  \\\n",
       "0                                       ##gff-version 3     NaN   \n",
       "1                     ##sequence-region   1 1 248956422     NaN   \n",
       "2                    ##sequence-region   10 1 133797422     NaN   \n",
       "3                    ##sequence-region   11 1 135086622     NaN   \n",
       "4                    ##sequence-region   12 1 133275309     NaN   \n",
       "5                    ##sequence-region   13 1 114364328     NaN   \n",
       "6                    ##sequence-region   14 1 107043718     NaN   \n",
       "7                    ##sequence-region   15 1 101991189     NaN   \n",
       "8                     ##sequence-region   16 1 90338345     NaN   \n",
       "9                     ##sequence-region   17 1 83257441     NaN   \n",
       "10                    ##sequence-region   18 1 80373285     NaN   \n",
       "11                    ##sequence-region   19 1 58617616     NaN   \n",
       "12                    ##sequence-region   2 1 242193529     NaN   \n",
       "13                    ##sequence-region   20 1 64444167     NaN   \n",
       "14                    ##sequence-region   21 1 46709983     NaN   \n",
       "15                    ##sequence-region   22 1 50818468     NaN   \n",
       "16                    ##sequence-region   3 1 198295559     NaN   \n",
       "17                    ##sequence-region   4 1 190214555     NaN   \n",
       "18                    ##sequence-region   5 1 181538259     NaN   \n",
       "19                    ##sequence-region   6 1 170805979     NaN   \n",
       "20                    ##sequence-region   7 1 159345973     NaN   \n",
       "21                    ##sequence-region   8 1 145138636     NaN   \n",
       "22                    ##sequence-region   9 1 138394717     NaN   \n",
       "23                       ##sequence-region   MT 1 16569     NaN   \n",
       "24                    ##sequence-region   X 1 156040895     NaN   \n",
       "25               ##sequence-region   Y 2781480 56887902     NaN   \n",
       "26                    #!genome-build Ensembl GRCh38.p12     NaN   \n",
       "27                              #!genome-version GRCh38     NaN   \n",
       "28                                #!genome-date 2013-12     NaN   \n",
       "29       #!genome-build-accession NCBI:GCA_000001405.27     NaN   \n",
       "...                                                 ...     ...   \n",
       "2951268                                               Y  havana   \n",
       "2951269                                               Y  havana   \n",
       "2951270                                               Y  havana   \n",
       "2951271                                             ###     NaN   \n",
       "2951272                                               Y  havana   \n",
       "2951273                                               Y  havana   \n",
       "2951274                                               Y  havana   \n",
       "2951275                                               Y  havana   \n",
       "2951276                                               Y  havana   \n",
       "2951277                                               Y  havana   \n",
       "2951278                                               Y  havana   \n",
       "2951279                                               Y  havana   \n",
       "2951280                                               Y  havana   \n",
       "2951281                                               Y  havana   \n",
       "2951282                                               Y  havana   \n",
       "2951283                                               Y  havana   \n",
       "2951284                                               Y  havana   \n",
       "2951285                                               Y  havana   \n",
       "2951286                                               Y  havana   \n",
       "2951287                                             ###     NaN   \n",
       "2951288                                               Y  havana   \n",
       "2951289                                               Y  havana   \n",
       "2951290                                               Y  havana   \n",
       "2951291                                             ###     NaN   \n",
       "2951292                                               Y       .   \n",
       "2951293                                               Y       .   \n",
       "2951294                                               Y  havana   \n",
       "2951295                                               Y  havana   \n",
       "2951296                                               Y  havana   \n",
       "2951297                                             ###     NaN   \n",
       "\n",
       "                           type       start         end  score strand phase  \\\n",
       "0                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "1                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "2                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "3                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "4                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "5                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "6                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "7                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "8                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "9                           NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "10                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "11                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "12                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "13                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "14                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "15                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "16                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "17                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "18                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "19                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "20                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "21                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "22                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "23                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "24                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "25                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "26                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "27                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "28                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "29                          NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "...                         ...         ...         ...    ...    ...   ...   \n",
       "2951268                    exon  26586642.0  26586766.0      .      -     .   \n",
       "2951269                    exon  26587398.0  26587560.0      .      -     .   \n",
       "2951270                    exon  26591548.0  26591601.0      .      -     .   \n",
       "2951271                     NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "2951272              pseudogene  26594851.0  26634652.0      .      -     .   \n",
       "2951273  pseudogenic_transcript  26594851.0  26634652.0      .      -     .   \n",
       "2951274                    exon  26594851.0  26595045.0      .      -     .   \n",
       "2951275                    exon  26600891.0  26601022.0      .      -     .   \n",
       "2951276                    exon  26611713.0  26611807.0      .      -     .   \n",
       "2951277                    exon  26614154.0  26614962.0      .      -     .   \n",
       "2951278                    exon  26615294.0  26615531.0      .      -     .   \n",
       "2951279                    exon  26621773.0  26621895.0      .      -     .   \n",
       "2951280                    exon  26622512.0  26622608.0      .      -     .   \n",
       "2951281                    exon  26623586.0  26623666.0      .      -     .   \n",
       "2951282                    exon  26627943.0  26628022.0      .      -     .   \n",
       "2951283                    exon  26628271.0  26628437.0      .      -     .   \n",
       "2951284                    exon  26630647.0  26630749.0      .      -     .   \n",
       "2951285                    exon  26633345.0  26633431.0      .      -     .   \n",
       "2951286                    exon  26634523.0  26634652.0      .      -     .   \n",
       "2951287                     NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "2951288              pseudogene  26626520.0  26627159.0      .      -     .   \n",
       "2951289  pseudogenic_transcript  26626520.0  26627159.0      .      -     .   \n",
       "2951290                    exon  26626520.0  26627159.0      .      -     .   \n",
       "2951291                     NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "2951292       biological_region  26626966.0  26627137.0  0.994      -     .   \n",
       "2951293       biological_region  26627457.0  26628186.0  0.997      +     .   \n",
       "2951294              pseudogene  56855244.0  56855488.0      .      +     .   \n",
       "2951295  pseudogenic_transcript  56855244.0  56855488.0      .      +     .   \n",
       "2951296                    exon  56855244.0  56855488.0      .      +     .   \n",
       "2951297                     NaN         NaN         NaN    NaN    NaN   NaN   \n",
       "\n",
       "                                                     attrs  \n",
       "0                                                      NaN  \n",
       "1                                                      NaN  \n",
       "2                                                      NaN  \n",
       "3                                                      NaN  \n",
       "4                                                      NaN  \n",
       "5                                                      NaN  \n",
       "6                                                      NaN  \n",
       "7                                                      NaN  \n",
       "8                                                      NaN  \n",
       "9                                                      NaN  \n",
       "10                                                     NaN  \n",
       "11                                                     NaN  \n",
       "12                                                     NaN  \n",
       "13                                                     NaN  \n",
       "14                                                     NaN  \n",
       "15                                                     NaN  \n",
       "16                                                     NaN  \n",
       "17                                                     NaN  \n",
       "18                                                     NaN  \n",
       "19                                                     NaN  \n",
       "20                                                     NaN  \n",
       "21                                                     NaN  \n",
       "22                                                     NaN  \n",
       "23                                                     NaN  \n",
       "24                                                     NaN  \n",
       "25                                                     NaN  \n",
       "26                                                     NaN  \n",
       "27                                                     NaN  \n",
       "28                                                     NaN  \n",
       "29                                                     NaN  \n",
       "...                                                    ...  \n",
       "2951268  Parent=transcript:ENST00000456738;Name=ENSE000...  \n",
       "2951269  Parent=transcript:ENST00000456738;Name=ENSE000...  \n",
       "2951270  Parent=transcript:ENST00000456738;Name=ENSE000...  \n",
       "2951271                                                NaN  \n",
       "2951272  ID=gene:ENSG00000237917;Name=PARP4P1;biotype=u...  \n",
       "2951273  ID=transcript:ENST00000435945;Parent=gene:ENSG...  \n",
       "2951274  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951275  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951276  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951277  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951278  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951279  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951280  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951281  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951282  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951283  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951284  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951285  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951286  Parent=transcript:ENST00000435945;Name=ENSE000...  \n",
       "2951287                                                NaN  \n",
       "2951288  ID=gene:ENSG00000231514;Name=CCNQP2;biotype=pr...  \n",
       "2951289  ID=transcript:ENST00000435741;Parent=gene:ENSG...  \n",
       "2951290  Parent=transcript:ENST00000435741;Name=ENSE000...  \n",
       "2951291                                                NaN  \n",
       "2951292        external_name=rank %3D 1;logic_name=firstef  \n",
       "2951293        external_name=rank %3D 1;logic_name=firstef  \n",
       "2951294  ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...  \n",
       "2951295  ID=transcript:ENST00000431853;Parent=gene:ENSG...  \n",
       "2951296  Parent=transcript:ENST00000431853;Name=ENSE000...  \n",
       "2951297                                                NaN  \n",
       "\n",
       "[2951298 rows x 9 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# df2를 출력해 봅시다.\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 정제 하기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "우리가 위에서 봤던 포맷과는 조금 다릅니다.. 바로 #으로 시작하는 주석 내용이 앞에 달려있기 때문입니다.\n",
    "그럼 일전에 우리가 공부했던 filtering으로 주석을 날려보는 작업을 해보죠."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? 'chr'열에 `#`으로 시작하는지 아닌지 조건문을 걸어 boolean 결과를 봅시다.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    True\n",
       "1    True\n",
       "2    True\n",
       "3    True\n",
       "4    True\n",
       "5    True\n",
       "6    True\n",
       "7    True\n",
       "8    True\n",
       "9    True\n",
       "Name: chr, dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? 그럼 True인 행을 출력해볼까요?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>##gff-version 3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>##sequence-region   1 1 248956422</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>##sequence-region   10 1 133797422</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>##sequence-region   11 1 135086622</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>##sequence-region   12 1 133275309</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  chr source type  start  end score strand  \\\n",
       "0                     ##gff-version 3    NaN  NaN    NaN  NaN   NaN    NaN   \n",
       "1   ##sequence-region   1 1 248956422    NaN  NaN    NaN  NaN   NaN    NaN   \n",
       "2  ##sequence-region   10 1 133797422    NaN  NaN    NaN  NaN   NaN    NaN   \n",
       "3  ##sequence-region   11 1 135086622    NaN  NaN    NaN  NaN   NaN    NaN   \n",
       "4  ##sequence-region   12 1 133275309    NaN  NaN    NaN  NaN   NaN    NaN   \n",
       "\n",
       "  phase attrs  \n",
       "0   NaN   NaN  \n",
       "1   NaN   NaN  \n",
       "2   NaN   NaN  \n",
       "3   NaN   NaN  \n",
       "4   NaN   NaN  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? 그럼 반대인 행들을 출력해서 df에 다시 넣으면 되겠네요**\n",
    "\n",
    "힌트) 관계 연산자는 > = < ~ | & 이 있습니다. 이중 ~는 not을 의미합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>1</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>248956422.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:1;Alias=CM000663.2,chr1,NC_00000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10469.0</td>\n",
       "      <td>11240.0</td>\n",
       "      <td>1.3e+03</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=oe %3D 0.79;logic_name=cpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10650.0</td>\n",
       "      <td>10657.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10655.0</td>\n",
       "      <td>10657.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10678.0</td>\n",
       "      <td>10687.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   chr   source               type    start          end    score strand  \\\n",
       "31   1  Ensembl         chromosome      1.0  248956422.0        .      .   \n",
       "33   1        .  biological_region  10469.0      11240.0  1.3e+03      .   \n",
       "34   1        .  biological_region  10650.0      10657.0    0.999      +   \n",
       "35   1        .  biological_region  10655.0      10657.0    0.999      -   \n",
       "36   1        .  biological_region  10678.0      10687.0    0.999      +   \n",
       "\n",
       "   phase                                              attrs  \n",
       "31     .  ID=chromosome:1;Alias=CM000663.2,chr1,NC_00000...  \n",
       "33     .           external_name=oe %3D 0.79;logic_name=cpg  \n",
       "34     .                                 logic_name=eponine  \n",
       "35     .                                 logic_name=eponine  \n",
       "36     .                                 logic_name=eponine  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2951292</th>\n",
       "      <td>Y</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>26626966.0</td>\n",
       "      <td>26627137.0</td>\n",
       "      <td>0.994</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=rank %3D 1;logic_name=firstef</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951293</th>\n",
       "      <td>Y</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>26627457.0</td>\n",
       "      <td>26628186.0</td>\n",
       "      <td>0.997</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=rank %3D 1;logic_name=firstef</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951294</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951295</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000431853;Parent=gene:ENSG...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2951296</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>56855244.0</td>\n",
       "      <td>56855488.0</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000431853;Name=ENSE000...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        chr  source                    type       start         end  score  \\\n",
       "2951292   Y       .       biological_region  26626966.0  26627137.0  0.994   \n",
       "2951293   Y       .       biological_region  26627457.0  26628186.0  0.997   \n",
       "2951294   Y  havana              pseudogene  56855244.0  56855488.0      .   \n",
       "2951295   Y  havana  pseudogenic_transcript  56855244.0  56855488.0      .   \n",
       "2951296   Y  havana                    exon  56855244.0  56855488.0      .   \n",
       "\n",
       "        strand phase                                              attrs  \n",
       "2951292      -     .        external_name=rank %3D 1;logic_name=firstef  \n",
       "2951293      +     .        external_name=rank %3D 1;logic_name=firstef  \n",
       "2951294      +     .  ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...  \n",
       "2951295      +     .  ID=transcript:ENST00000431853;Parent=gene:ENSG...  \n",
       "2951296      +     .  Parent=transcript:ENST00000431853;Name=ENSE000...  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 맨 뒷부분 출력\n",
    "df.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "'source'에 'Ensembl'이라고 적혀있는 행들은 각 염색체의 gene전체를 설명하고 있네요.\n",
    "전 각 유전자에 관심있으니 이 행들을 모두 지워볼께요"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>1</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>248956422.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:1;Alias=CM000663.2,chr1,NC_00000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270858</th>\n",
       "      <td>10</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>133797422.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:10;Alias=CM000672.2,chr10,NC_000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>382137</th>\n",
       "      <td>11</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>135086622.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:11;Alias=CM000673.2,chr11,NC_000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>554243</th>\n",
       "      <td>12</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>133275309.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:12;Alias=CM000674.2,chr12,NC_000...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>720634</th>\n",
       "      <td>13</td>\n",
       "      <td>Ensembl</td>\n",
       "      <td>chromosome</td>\n",
       "      <td>1.0</td>\n",
       "      <td>114364328.0</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=chromosome:13;Alias=CM000675.2,chr13,NC_000...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       chr   source        type  start          end score strand phase  \\\n",
       "31       1  Ensembl  chromosome    1.0  248956422.0     .      .     .   \n",
       "270858  10  Ensembl  chromosome    1.0  133797422.0     .      .     .   \n",
       "382137  11  Ensembl  chromosome    1.0  135086622.0     .      .     .   \n",
       "554243  12  Ensembl  chromosome    1.0  133275309.0     .      .     .   \n",
       "720634  13  Ensembl  chromosome    1.0  114364328.0     .      .     .   \n",
       "\n",
       "                                                    attrs  \n",
       "31      ID=chromosome:1;Alias=CM000663.2,chr1,NC_00000...  \n",
       "270858  ID=chromosome:10;Alias=CM000672.2,chr10,NC_000...  \n",
       "382137  ID=chromosome:11;Alias=CM000673.2,chr11,NC_000...  \n",
       "554243  ID=chromosome:12;Alias=CM000674.2,chr12,NC_000...  \n",
       "720634  ID=chromosome:13;Alias=CM000675.2,chr13,NC_000...  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Ensembl 이 담겨있는 행들을 출력해봅시다.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10469.0</td>\n",
       "      <td>11240.0</td>\n",
       "      <td>1.3e+03</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=oe %3D 0.79;logic_name=cpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10650.0</td>\n",
       "      <td>10657.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10655.0</td>\n",
       "      <td>10657.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10678.0</td>\n",
       "      <td>10687.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10681.0</td>\n",
       "      <td>10688.0</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   chr source               type    start      end    score strand phase  \\\n",
       "33   1      .  biological_region  10469.0  11240.0  1.3e+03      .     .   \n",
       "34   1      .  biological_region  10650.0  10657.0    0.999      +     .   \n",
       "35   1      .  biological_region  10655.0  10657.0    0.999      -     .   \n",
       "36   1      .  biological_region  10678.0  10687.0    0.999      +     .   \n",
       "37   1      .  biological_region  10681.0  10688.0    0.999      -     .   \n",
       "\n",
       "                                       attrs  \n",
       "33  external_name=oe %3D 0.79;logic_name=cpg  \n",
       "34                        logic_name=eponine  \n",
       "35                        logic_name=eponine  \n",
       "36                        logic_name=eponine  \n",
       "37                        logic_name=eponine  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 각 염색체 전체를 담고있는게 맞네요, 지워 줍니다.\n",
    "df = df[~(df['source'] == \"Ensembl\")]\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 인덱스가 0번부터 시작하질 않네요. 고쳐봅시다.\n",
    "df = df."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pandas Dateframe에선 missing이 NaN입니다. '.'을 NaN으로 교체하고 싶네요~ (replace사용)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 탐색하기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "데이터가 만들어 졌다면, 각 열(변수)들을 summary하는 일은 필!수! 입니다. 해보죠"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "chr         object\n",
       "source      object\n",
       "type        object\n",
       "start        int64\n",
       "end          int64\n",
       "score       object\n",
       "strand      object\n",
       "phase       object\n",
       "attrs       object\n",
       "gene_len     int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "start, end는 포지션인데 소숫점이 찍혀있어서 뭔가 불편합니다. 형 변환이 필요합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "      <th>gene_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10469</td>\n",
       "      <td>11240</td>\n",
       "      <td>1.3e+03</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=oe %3D 0.79;logic_name=cpg</td>\n",
       "      <td>772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10650</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10655</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10678</td>\n",
       "      <td>10687</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10681</td>\n",
       "      <td>10688</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  chr source               type  start    end    score strand phase  \\\n",
       "0   1      .  biological_region  10469  11240  1.3e+03      .     .   \n",
       "1   1      .  biological_region  10650  10657    0.999      +     .   \n",
       "2   1      .  biological_region  10655  10657    0.999      -     .   \n",
       "3   1      .  biological_region  10678  10687    0.999      +     .   \n",
       "4   1      .  biological_region  10681  10688    0.999      -     .   \n",
       "\n",
       "                                      attrs  gene_len  \n",
       "0  external_name=oe %3D 0.79;logic_name=cpg       772  \n",
       "1                        logic_name=eponine         8  \n",
       "2                        logic_name=eponine         3  \n",
       "3                        logic_name=eponine        10  \n",
       "4                        logic_name=eponine         8  "
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['start', 'end']] = df[['start', 'end']].astype(int)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "chr         object\n",
       "source      object\n",
       "type        object\n",
       "start        int64\n",
       "end          int64\n",
       "score       object\n",
       "strand      object\n",
       "phase       object\n",
       "attrs       object\n",
       "gene_len     int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "다른 통계량 확인해보죠"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>gene_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.890659e+06</td>\n",
       "      <td>2.890659e+06</td>\n",
       "      <td>2.890659e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.463664e+07</td>\n",
       "      <td>7.464056e+07</td>\n",
       "      <td>3.925582e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5.604397e+07</td>\n",
       "      <td>5.604429e+07</td>\n",
       "      <td>2.980241e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.770000e+02</td>\n",
       "      <td>6.470000e+02</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.176504e+07</td>\n",
       "      <td>3.176786e+07</td>\n",
       "      <td>8.400000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.079471e+07</td>\n",
       "      <td>6.080053e+07</td>\n",
       "      <td>1.330000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.105063e+08</td>\n",
       "      <td>1.105082e+08</td>\n",
       "      <td>2.500000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.489450e+08</td>\n",
       "      <td>2.489452e+08</td>\n",
       "      <td>2.473537e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              start           end      gene_len\n",
       "count  2.890659e+06  2.890659e+06  2.890659e+06\n",
       "mean   7.463664e+07  7.464056e+07  3.925582e+03\n",
       "std    5.604397e+07  5.604429e+07  2.980241e+04\n",
       "min    5.770000e+02  6.470000e+02  1.000000e+00\n",
       "25%    3.176504e+07  3.176786e+07  8.400000e+01\n",
       "50%    6.079471e+07  6.080053e+07  1.330000e+02\n",
       "75%    1.105063e+08  1.105082e+08  2.500000e+02\n",
       "max    2.489450e+08  2.489452e+08  2.473537e+06"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['chr', 'source', 'type', 'score', 'strand', 'phase', 'attrs'], dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 범주형 컬럼들은 value_counts로 요약 가능했죠?\n",
    "# 우선 범주형 컬럼을 확인해 봅시다.\n",
    "df.columns[df.dtypes == 'object']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1     265354\n",
       "2     215915\n",
       "3     184514\n",
       "17    173690\n",
       "19    171825\n",
       "11    168744\n",
       "12    163335\n",
       "7     139227\n",
       "16    135778\n",
       "6     132429\n",
       "5     128988\n",
       "4     115581\n",
       "10    108945\n",
       "15    107781\n",
       "8     106386\n",
       "9     103633\n",
       "14     99475\n",
       "X      91975\n",
       "20     68741\n",
       "22     64026\n",
       "18     52996\n",
       "13     48798\n",
       "21     34495\n",
       "Y       7903\n",
       "MT       125\n",
       "Name: chr, dtype: int64"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# chr에 관하여 살펴봅시다.\n",
    "df['chr'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "+    1461267\n",
       "-    1406882\n",
       ".      22510\n",
       "Name: strand, dtype: int64"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# strand 방향에 대해서도 살펴 보죠.\n",
    "df['strand'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? 유전자의 길이를 나타내는 'gene_len' 컬럼을 새로 만들어 df에 넣어 봅시다.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "      <th>gene_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10469</td>\n",
       "      <td>11240</td>\n",
       "      <td>1.3e+03</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=oe %3D 0.79;logic_name=cpg</td>\n",
       "      <td>772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10650</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10655</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10678</td>\n",
       "      <td>10687</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10681</td>\n",
       "      <td>10688</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  chr source               type  start    end    score strand phase  \\\n",
       "0   1      .  biological_region  10469  11240  1.3e+03      .     .   \n",
       "1   1      .  biological_region  10650  10657    0.999      +     .   \n",
       "2   1      .  biological_region  10655  10657    0.999      -     .   \n",
       "3   1      .  biological_region  10678  10687    0.999      +     .   \n",
       "4   1      .  biological_region  10681  10688    0.999      -     .   \n",
       "\n",
       "                                      attrs  gene_len  \n",
       "0  external_name=oe %3D 0.79;logic_name=cpg       772  \n",
       "1                        logic_name=eponine         8  \n",
       "2                        logic_name=eponine         3  \n",
       "3                        logic_name=eponine        10  \n",
       "4                        logic_name=eponine         8  "
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "96        19765\n",
       "102       18220\n",
       "93        17881\n",
       "84        17843\n",
       "105       17738\n",
       "99        17656\n",
       "108       17543\n",
       "81        17398\n",
       "72        17237\n",
       "78        17063\n",
       "111       17040\n",
       "117       16850\n",
       "87        16785\n",
       "90        16703\n",
       "126       16478\n",
       "123       16380\n",
       "114       16350\n",
       "75        16158\n",
       "120       16137\n",
       "129       15571\n",
       "107       15424\n",
       "69        15321\n",
       "109       15256\n",
       "98        14993\n",
       "103       14847\n",
       "100       14645\n",
       "132       14595\n",
       "127       14519\n",
       "104       14502\n",
       "118       14412\n",
       "          ...  \n",
       "21412         1\n",
       "27555         1\n",
       "25506         1\n",
       "31649         1\n",
       "107418        1\n",
       "134127        1\n",
       "193520        1\n",
       "189426        1\n",
       "46792         1\n",
       "695011        1\n",
       "136173        1\n",
       "33727         1\n",
       "41915         1\n",
       "46009         1\n",
       "54197         1\n",
       "58291         1\n",
       "57029         1\n",
       "50886         1\n",
       "42698         1\n",
       "179193        1\n",
       "38604         1\n",
       "107419        1\n",
       "125842        1\n",
       "66447         1\n",
       "76682         1\n",
       "465740        1\n",
       "348933        1\n",
       "582323        1\n",
       "164862        1\n",
       "29650         1\n",
       "Name: gene_len, Length: 82494, dtype: int64"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# gene len에 value_count\n",
    "df['gene_len'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? chr 1에 대하여 200까지 유전자 길이의 frequency를 그려보기**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x13ba825f8>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2 = \n",
    "df2.hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "hist말고 line그래프를 그려봅시다.\n",
    "기본적으로 선 그래프는 x축 y축이 필요합니다.\n",
    "\n",
    "여기서 우린 x축을 gene_len, y축을 frequency로 잡으려 합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gene_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>1876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>1867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>1784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>1699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>1635</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     gene_len\n",
       "96       1876\n",
       "173      1867\n",
       "111      1784\n",
       "52       1699\n",
       "93       1635"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# x축을 gene_len, y축을 frequency로 잡기 위해 df3로 value_count에 대한 df를 생성합니다.\n",
    "df3 = \n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x19b081780>"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJztnXmcZFV5sJ9zb22993T3bMzCzMCwqTDACElA1LiCCpIIwS/GNWIixqxfIprExCVxN2bTDyNBoqAmxEhUIooLGmQZdmVxWAaYYdbu6b1rvef74y5116rq7uqu6u73+f2q773nvOec956qPu+9Z3mP0lojCIIgCH6MVisgCIIgtB9iHARBEIQIYhwEQRCECGIcBEEQhAhiHARBEIQIYhwEQRCECGIcBEEQhAhiHARBEIQIYhwEQRCECKlWK1CPoaEhvWXLllarIQiCsGS4++67j2itV88nj7Y3Dlu2bGHXrl2tVkMQBGHJoJR6ar55SLeSIAiCEEGMgyAIghBBjIMgCIIQoe3HHOIolUrs3buXfD7falWWJLlcjo0bN5JOp1utiiAIbcqSNA579+6lp6eHLVu2oJRqtTpLCq01w8PD7N27l61bt7ZaHUEQ2pQl2a2Uz+cZHBwUwzAHlFIMDg7KW5cgCDVZksYBEMMwD6TuBEGox5LsVhKEhebh/ePc9OB+lFIYSmEoMAzfuVIo5+jGKV+cofDSmoYrHx8fzS9aXjCtT9aIKc+NN5LzjsvPDJUnrGzEOAhCDL84OME//OAxVvIW67HGyG9sPGMSY/SMWmntczPRoNZOazaQt2s4vbK8OL8xr5HW0cU0FC89eQ3Hr+lp9dex6IhxWEJ0d3czOTnZajVWBBft2MBFOzagtUZrqGiN5ZxbWmM5R22513aY9sVZWmPViY/k55xXrBrxFsnl+eKT8g6mrZZXK97Lz4rRNVGvUNpQPVSs5PiKZcXrEqrPSForXCeQL1WYLJTn/FuYKpT545ef2MRf19JAjIMg1EC5T6dIN0s7UapYjEwVOTxRqH4mg8cjzjHJMKzqTLO6J2t/urMMdWer1z3V68GuzCLfXXuw5I3DX//3z3no2fGm5nnKMb28/zXPqSnzwQ9+kC996UusXr2aTZs2ceaZZ3LxxRdzxRVXcPjwYTo7O/n85z/PSSedxJvf/GZ6e3vZtWsXBw4c4GMf+xive93rAPj4xz/O1772NQqFAhdffDF//dd/3ZCOcen27NnD+eefz7nnnsttt93Ghg0b+MY3vkFHR8e860QQFhrL0hydLlYbeV8Db58XvcZ/ZKoYm0dPNmU37D1ZTl7Xy3nbfY1/T4bV3Tm7we/OkDaX7HycRWHJG4dWcNddd3HDDTdw//33UyqVOOOMMzjzzDO5/PLL+dznPsf27du54447eOc738n3v/99APbv389PfvITHnnkES688EJe97rXcfPNN7N7927uvPNOtNZceOGF3HrrrZx33nk1y09Kt3nzZnbv3s3111/P5z//eS699FJuuOEG3vCGNyxGtQhCBK014/lytLGfjDb+w1NFKlZ0kCebMljTaz/JHzvYyc4tqwJP9m7jv7onSy5ttuAulydL3jjUe8JfCP73f/+Xiy66iFwuRy6X4zWveQ35fJ7bbruNSy65xJMrFAre+Wtf+1oMw+CUU07h4MGDgN3I33zzzZx++ukATE5Osnv37oaMQ1y6zZs3s3XrVnbs2AHAmWeeyZ49e5p564IAwHSxHGnc7Ua/+nR/xDkWy1YkfcpQXuO+tjfHc4/pc57ss6zuyTmNf4bVPVm6symZPdUClrxxaBcsy6K/v5/77rsvNj6bzXrn2pkCo7Xmyiuv5B3veMesykpKt2fPnkA5pmkyMzMzq7yFlUuhXOHIZNFu1BOe7t1Gf6pYiaRXCga7Ml6jf9zqLu+JPvCk352lryONYUiD386IcZgD55xzDu94xzu48sorKZfLfPOb3+Tyyy9n69at/Pu//zuXXHIJWmseeOABTjvttMR8XvGKV/AXf/EX/OZv/ibd3d3s27ePdDrNmjVrapaflE4QwpQrFiPTwYFbf9/94Ym8dz02U4rNo68j7TXqp27sDzX4Ge98oDNDSvrxlw1iHObA85//fC688EJOPfVU1q5dy/Oe9zz6+vr48pe/zO/+7u/yoQ99iFKpxGWXXVbTOLz85S/n4Ycf5pd/+ZcBe6rql770pbrGISmdaUp/60rAsjRjM6XEgVv/E//wVDF2rUZXxvSe5rev6eZXjht0Bm2zgcZ/sDtDNiW/q5WI0nG/nDZi586dOrwT3MMPP8zJJ5/cIo1sJicn6e7uZnp6mvPOO4+rrrqKM844o6U6zYZ2qEOhitaayUI59HSf9zX0xYAhKMcM3GZSRrSB9z3Zu8ZgqDtLV1aeC5czSqm7tdY755OH/ELmyOWXX85DDz1EPp/nTW9605IyDMLiobXm6HSJg+N5Dk0UODie57BzPDRe4OBE3mvw86XowK1pKAa7qg38iet6ArNz/DN2enMycCs0j7rGQSl1NfBq4JDW+rlO2FcBd8lgPzCqtd6hlNoCPAw86sTdrrX+HSfNmcA1QAfwbeD3dbu/ttTguuuuW5B8h4eHeclLXhIJv+WWWxgcHFyQMoXZ42/03Yb/kM8A2NcFDk3kKVWiP/PeXIo1vTnW9GTZeeyq6KCtYwBWdWZk4FZoCY28OVwD/CNwrRugtf4N91wp9UlgzCf/uNZ6R0w+nwXeDtyBbRxeCdw0e5U9HZblU9Lg4GDijKdmsYRt8oLjLsQ65Hu6PzSR56DveHiidqO/tjfH2t4cZ2/t8gzA2t4ca3qzrO2xjzIfX2h36hoHrfWtzhtBBGW3zpcCv1orD6XUeqBXa327c30t8FrmaBxyuRzDw8Oyp8MccDf7yeVyrVZlUXEbfbeRPzReCHT1uE/+hycLsY1+X0eatb1Z1vTk2La6izU9Oe/aPUqjLywn5jvm8ALgoNZ6ty9sq1LqXmAc+HOt9Y+BDcBen8xeJ2xObNy4kb1793L48OG5ZrGicbcJXQ5YlmZkuljtv3ca/YOuAZgocNhp/OMGcfs7096T/bbVg/YTvvuk7xxl5a2wEpmvcXg9cL3vej+wWWs97Iwx/JdSatZLmJVSlwOXA2zevDkSn06nZYvLZY7b6Af68x0DYD/9O0/6NRp9twvn+NVDTpdOljW91Sd9afQFIZk5GwelVAr4NeBMN0xrXQAKzvndSqnHgROAfYD/UXWjExaL1voq4Cqwp7LOVUeh/bAszfBU0evaCffnu4O6SY3+qs6014Vz/Oohp6F3+/Ttp31p9AVh/sznzeGlwCNaa6+7SCm1GhjRWleUUtuA7cATWusRpdS4UuqXsAek3wj8w3wUF9oLt9EPTNUM9efXa/TdLpzta3u8Bn9tr+1rxz5mZUGWICwSjUxlvR54ETCklNoLvF9r/QXgMoJdSgDnAR9QSpUAC/gdrfWIE/dOqlNZb2IeM5WExcPf6FcHcv0zd+xj0sKsga4Ma5zunBPW9tjdO84Tvv9JXxp9QWgvluQKaWH+VCzN8FQh2LXj9Okf8s3oOTxZiHWj7Db6a0NTNd0un7W9OVZ3Z8mkxNeOICw2skJamDW7D07w1i/exbOj+dhG31Bw/Jpu1vd1cNK6Hm+qptu1s0YafUFYEYhxWGH0dab55W2DPD0yzYGxPAfG8wG3DZaG3YcmGZ0uMTqdY2ymxES+xGShQr5UIV+yKJUt1vXlZNBXEJYxYhxWGGt6cnzsdVVPsVrbHj73O4biwFjePh+b4cB4gSePTHHb48NM5KP78PZ3plnXm2N9X451fTnW9Xawvi/H2r5qWI9s1CIISxIxDiscpRT9nRn6OzOcvL43UW6qUPaMh/vGsX9shgNjBQ6Mz/DgvjGOTEb39e3KmFVj0dvBur4s6/o6WN/rGJS+HAPiP0gQ2g4xDkJDdGVTHLe6m+NWdyfKFMoVDo0XHMOR56D7FjI+w4GxPD99/AgHJ6ID3BnT3iPYftvoYF2vY0D6bB9F6/vsQW/ZSEYQFg8xDkLTyKZMNg10smmgM1GmYmmGJwvsdwzHwfFqN9b+sTwP7h3l5rE8hdC+w4aCoe6s1121vq/DMxx2l1ZOxkEEoYmIcRAWFdNQ9vqG3hynbYqX0VozOl0KjoGMVw1IrXEQdzFd9S3EZ0BkHEQQGkaMg9B2KKVY1ZVhVVfj4yDVQfTq9WzGQYa67W4rQ4GhFIay9XDPDaVQXpx7XlumGg+GUT/PgLy/HCNe3jQUfR2yd7iwMIhxEJYscxkHOeAbRN9fYxxkqfDp3ziNi09fHh52hfZCjIOwrGl0HGRspkTF0mitsTRYWmNpjfbOca598RaejEZTLFscnS4xMlVgeKrI0akiw1NFRnznR6eKTBUrDeufSxv0d2To70yzqtM+9jvHwa4MLzxhTTOqSRAiiHEQVjymoRjoyjQsf98zo1z9kyc5Ol1kdLrE0ekiY9MlJgrRMRCXlGFPGV7f38Eqt4HvSLOqK0Nfh93wr+pM0+czAqs6MzLALrQMMQ6CMEueGp7ip08MM5EvBVaXJ2EaiqHuLL0dKXpyabqzKTrSJpmUgWnY4wgVrSlWLAoli6lCGdNZ99GTS9OVMWUAXVh0xDgIwiy5aMcGLtphb2RYLFtM5EtM5MuMO8eJfInxfJnxGffaDbOvj0wWeeLIlBcety2pH0PZRqInZxuXXt+xt8MNT9GbS3ty4fBsyhADI8wKMQ6CMA/y5Qpn/c0tCzqgbWkYmykxNlMCZuaUR8Y0qsbCMRy/fe42XnySjFkI8YhxEIR5YFl6UWc6nbqxryn5TOTLTBWTx0gEQYyDIMyD/s4Mez7yqkBYuWLxpduf4okjUwxPFjk8WWB40p7BNDpdis0nYxoMdmfsT1eWwe4MQ91ZBrvs4+qeLOccP+SNRQjCQiPGQRCazES+zKe/t9vpBkqmM2NywtoeTlrXw7GDXQx1Z3jB9tWs68stkqaCkIzsBCcIC0C5Yq95ODJp75v9xOFJHjkwwb1Pj/KLQxMk/dv92hkb+NSlOxZXWWHZITvBCUIL0FozUSgzMllkeKrAkUl7odvwpO98qsDwZHURXHhcojub4tjBTufTxbED9nHHpv4W3ZUgBBHjIAjATLHCEWdcYMRp8Icn7fPhySJHnMbfNgJFipX49Q092RSD3RkGujJsGujk9M39DHRVxxE2DXRy7EAnA10ZmVoqtDV1jYNS6mrg1cAhrfVznbC/At4OHHbE3qu1/rYTdyXwNqACvFtr/R0n/JXAZwAT+Bet9UeaeyvCSkVrTb5kMVkoM1koMxU6euf5MpOFChP5EsOOO4vhSbvxnynFu7TIpQ1vYHhtb45T1vcy0J1hyGnsB5wB48HujKxoFpYVjbw5XAP8I3BtKPzTWutP+AOUUqcAlwHPAY4BvqeUOsGJ/ifgZcBe4C6l1I1a64fmobuwAnh4/zhfvesZe+plwZ5+6Z27DX+x0tB0UqWgK5Oi23m6H+zOsm2oi8Eu+9w+Bs87M/JyLaxM6v7ytda3KqW2NJjfRcBXtNYF4Eml1GPAWU7cY1rrJwCUUl9xZMU4CDW575lR/u32pxpq/A0FG1d1smWoi62D9tE+72KoJ0tn2pTtSAWhQebzWPQupdQbgV3AH2utjwIbgNt9MnudMIBnQuFnz6NsYYXw+rM28xs7N3F02l4vcGSiyOHJPIcn7HGBwxMF59w+PnN0mqdHprk1lE9XxmR1T9ZbMxA4784y5IVlyKaka0gQ5mocPgt8ENDO8ZPAW5ullFLqcuBygM2bNzcrW2GJYhjK7urpzsK62rLlisXIVJFDPoNRNSoFDk/k2X1oktseH05ch9CbSyUbEseYrO6xu55kX2thuTIn46C1PuieK6U+D3zTudwH+Dd/3OiEUSM8Lv+rgKvAXucwFx2FlUnKNLxtSOtRKFfsFcx+Q+KeO9c/f3acwxMFJmPccSsFqzoznrEY6s7EGpOh7iwDnRnp0hKWFHMyDkqp9Vrr/c7lxcDPnPMbgeuUUp/CHpDeDtwJKGC7UmortlG4DPg/81FcEOZLNmVyTH8Hx/R31JV1p7oeijMkztvJU09PcXiiEOvG2zRUwBVG1IBkWNOTZXV3jt4O2eNaaD2NTGW9HngRMKSU2gu8H3iRUmoHdrfSHuAdAFrrnyulvoY90FwGrtBaV5x83gV8B3sq69Va6583/W4EYYHoyNTfUQ7sabWThXLseIjfkOw+OMHhyUKsu+6MaTDUnbHHQbrj30Tcc9nrQVgoxH2GIDSBUsXi6LS9ErpcsT21VrRz9H3Klr39qCszMl3k4Ji9v/XB8TwHxvMcGMtzaCJfd58HsNdheIPqjtF4yzlbOX5N8r7awvJH3GcIwiJz91NHuffpoxwYy7N/PM/+0RkOjOU5OFFYVNfdLvmSxd6jtg6moUgbBmdtHRDjIMwbMQ6C0CCPHpjg1z9727zyMA1FR9qkI2PSmTFjzlN0OmEdGdM778yk6MyY5NK2bKcT35E2yaZNsimDTMogYxpkUzKDSpg/YhwEoUGOX9PNX776FPYMT7GuL0dH2qRYtiiULYpli2LF8q4L5YodVra4+SFvch8VS3suPRaTz1y2w9vaVBAaQYyDIDSIaSjeeu7WWae76tbH+eJtT7FvNH6Lz6HuDCet6yVtKsqWplSxKFU0xbJFqWIbnVLFolTWngFywxodMtxcZyBdEMLIgLQgLBKHJwo8sHeU+/eO2cdnRjnq7AyXSRmcvL6XHRv7OHVjP6dt6mPbULesjRDmRDMGpMU4CEKL0Fqz9+gM9+8d5YG9Y9z/zCg/2zfGVNH2ENudTfHcDb2ctqmf0zb2c+rGPjb0d8jUVaEuMltJWNJoZ6pnqaKrXSdO90nJCnallHyfYtlOZ2nfxwJLa7TGCYOK1mitsSz72h9fcc8DcdV0Vp147SvT0jhxmopf1nLy8k1rtWLCq/FwTH8HTw1PU6zYLshvf2KE258Y8eps61AX3/y9c+nKyr+usLDIL2yF854bHuArdz1TX7AOKUNhGqp6NA3vOmUqLAuKFYuy258+yz7zxUApMJTCUKCUwnTODaXsOEN58faxKmsYwTD3XCl7rMJ00tpHOz6TMkLhCtOA7Wu7vTBTKVsXw85nTU9OZiMJi4IYhxXODx89XF+oAcrOAq9CU3KzSRmKrmyKroxJZzZVPfemdRrk0vb0zmzKnsLpnTvh24a62NDfYTfUBoHGWykcA2CfS3eNIFQR47DCuf29L4mEBbp73CmaFYuSb7pmyTtqipUKRbf7xzeTxptVU7ZlAvmV/d1EFsWKpliuRGbpuOfDkwUOjOW99I2yrjcXe4+CINRGjIMQQSm7Kyhl2j6FWk25YvHM0RmeODzJE4eneOLIJI8cmOAXBya8wVuwu102D3SybaiLbau72La6m+dvGWih5oKwdBHjILQlI1NF3n/jz3no2TGeHpmu6WcokzLYOtjFpoFOMim7a2jf6Az7Rmf48e7mdJu1gnW9Hfz5q06W6axCSxDjILQlM6UKz4xMYyjFlsGuuvKW1jw1PLUImtmuiKcLZcZmSoE3l2bRkTbp60izfW0ZS2sMxDgIi48YB6Et2dDfwX9dcU6r1fC48j8f5KePH+HodClxB7lmMVOqcNPvv4AtQ/WNoiAsFGIchLbFsuxB7rDvIneQOi7cHhyv+jgKD6C78l7aSPr4sNHphTUIYaYX4I1EEGaDGAehpVxx3T08/Ox4bKNcbqILbNNQZEzHc6nPe2kmZZA2q2Fd2ZQnkw3Je+f+6wSZbMogY5reddpUTp7VMFPGEoQ2RoyD0FI2D3QyPlNiqlBmqlBhslBG6zLFcuPTVf30ZFMMdGdY1ZlhoMs9punNpenO2Wslup1Pl3c06cmm6cqapExZYCYIIMZBaDF/9sqTYsO11hTKtguJKcfF9WS+zFSxzGShYp874V68cz5VqDBRKPPYoUkmC2XGZ0oNv4VkUwY9OXfBnWNIPKNiBoxKTy7FK5+7nr6OdDOrRBDaAjEOQluilPJWPw91Z+vK/+CRQ7zlmrvmXW6hbFGYLHJkstiQ/J/d8CCbBjoAUL5ZRf7F1v7OI3cV9pNHFmdmVTP4wEXP4Y2/vKXVagiLjBgHoW2YLJT51588SUfGZFVnhv7ONKahCDzza/+pfaE1PHG4NY1txjQ4OFZAoz0/URr7zad63hLVmkZXRpqJlYh860Lb8Hff/QX/8pMnW63GrJiNK4925ft//EK2rZY9p4UgdY2DUupq4NXAIa31c52wjwOvAYrA48BbtNajSqktwMPAo07y27XWv+OkORO4BugAvg38vm73zSSEReVPXnEiA90ZDk8UODpV5Oh0iaPTRUamihydKi7IgrOlht9zbMALbMjrq1LKc0aYTZlk0wZpwwh4j3Xz+uA3H3KuffkajrfZsGfawLVfj5C8MUv5cP6GYutQFy/YvrrVVb5iaeTN4RrgH4FrfWHfBa7UWpeVUh8FrgT+zIl7XGu9IyafzwJvB+7ANg6vBG6ao97CMsLdJyFtGrzjvOMakp8qVhiZKlIsW96eDlrDZ27ZzXd9ezYvN7Szn4RtJpf3s9UxfTluu1KcJraKusZBa32r80bgD7vZd3k78LpaeSil1gO9WuvbnetrgdcixmFJ87kfPc5Hbnqk1WoIdVjdk3XGPaobE7ljIdoxqhefsYFzjx8KPMEr7+neHmw3FODG4b6tAPjfBGxZFXobsI9VWX8cTl4Kf7nQm5NZYK2kGWMObwW+6rveqpS6FxgH/lxr/WNgA7DXJ7PXCYtFKXU5cDnA5s2bm6CisBA82aJBYGF2HJ6ov8vGtT99ig9c9NxF0EZYKszLOCil3geUgS87QfuBzVrrYWeM4b+UUs+Zbb5a66uAq8DeQ3o+OgoLx0dfdyoffd2pgO3q4sF9Y+RL9p4MJcuiXNGUnX0ZyhVN2bL3f3B3g/PLlCxNyVkVXXLkSyH5srt1qBNWdvacKPl2mCs7eYbLbCVpU5E2DVKGczQVKcM92mFpZ+e8tBOXThmkvV31FKZhYCrso+E7KoXh7LhnODvH+XeXM43gpkbVXel84xRKcdZWcW0uBJmzcVBKvRl7oPol7sCy1roA9mZgWuu7lVKPAycA+4CNvuQbnTBhmXDDPXv5v//xQKvVaEtsA7Y0BtM/c9kOLtqR+FIvrCDmZByUUq8E/hR4odZ62he+GhjRWleUUtuA7cATWusRpdS4UuqXsAek3wj8w/zVF9qF156+gXV9OcoVd35/dQ2Ce/zta3e1Sj2hQT5x86M8dmjSXrgXGUOoLuL7leMGOX3zqpbqKiwsjUxlvR54ETCklNoLvB97dlIW+K7zY3GnrJ4HfEApVQIs4He01iNOVu+kOpX1JmQwelmRNg1esH019z8zyh9+7T60xptFZFn2udD+PDMywz98/7G6cpecuVGMwzKnkdlKr48J/kKC7A3ADQlxuwAZ8VrmfPp7v2jZauV2IZsyePsLtnmzecJrEfzz/k1vTUA13PTP+fetS/CvDzCN6Kwfd3whbv1DYF2EEVxr4I5HKFWdgWRG1iUEdcmlxUHhckdWSAtN5fNv3MlTw9OBRkUpvEFS/5THWeX74yf53I8en7U+mZRBX0ea3lyKvo40qzozrOqqemzt60jbA76+xtV0Bnb9A7yR8NCiM/c8kzLYNtTldb8IwlJFtfsi5Z07d+pdu6SvWrBnRH3u1se544kRUr6ZPIZStgfXfNU7q+vFdaZUeyBYKdvNd19nmr6O4KfXPeaicW687MkgtCNKqbu11jvnk4e8OQhLBsNQvPNFx/POFzWeplyxHBfeJSYLZcacbT7dz7j/PG/vC31wfNILq7evRE825RmRwKczaGACRsV5i5G9I4R2RoyDsKRxXW+ULYuKpSlbmkrFPlraf21hKkVvR5qubIo1vTkqzpqIitaRtPa1bVhGpgqMTJUYmSowPGX7ehqZKjI8VWSiUGaiUGbf6Mysde/KmJ4BOeWYXj51aZzXGUFoDWIchFlzYCzP//mX29l3dMZ2wYDdmDZxV88lh7u4LOWOUzgL3Owwoxrn+/jj04a8RQjthRgHYdbc8eTwsp+RtG11F6/dscHbZ9q/R7Tr6dQf7q1k9q9YVlUjELeK2T/ALQjthgxIC3Niulj2dj7zvKIC2lnToJ1w17mb5bxhWD5nb5bvaAUcwxHwtOrJxuWpg87krFDemuo6C3cTHjveDvu96+9tVRV6eO4tQm8edQ2NCr6J+I2OXyYVE+Z3wVGx8MZeRmeK9njLdIk3n7OVP3rZCa2uHmEONGNAeskZh+HJAvvH8nPKazazCxVze5prpIzqjmHBVcSNyGgvvppIh9IQk29Yxr9TWVQ2WFgkrc9VdJJeIVXi70VHRGro1cA9JdSVP9+w3lNFe99pS2OPM3hjEFbMGIQdXrGoxjvhlhUcq4gPt4+VGLmKU65fph148m8vkGm5S5AVN1upWLY480Pfa7UagrAsSRmKga4Mg91ZBrvsNSC//5X7gOpDj2smXIPhmQ0FL9g+xMWnb0RYHiwp45A2Fe+74GS+fm/VZ99snq9a+ZbUjKL1rO52bmU3u4ZmU+fzKnueimvsN4dSyLurfW2tiMH2sqU5NFHgUB0X34aClGl7jU05HmXTpmKgM8PFpy+SssKCs6SMg1KKt5+3jbeft63VqggLyE8fH+aRA+Oe2+5qI+131W35XHsvHbfeKWfgOu245Q668a6673ZlUqbdAGd8br5d999eeEjebaxThnN05DMpw5eHGx+UccvO+FyLu/FppwwZQF8ZLCnjIKwM/vIbP2P3ockFLSNjGrz4pNVsHuiMfQpOGb4G0TQCMl7j7ms03UbUa9h9Dbh/PwfpvxeWCmIchHnjThJwn9hLFYti2d7kp1SpPskXy9HrYsWi5IQXnbTHre6mI2MyOl1idNpeaNbsHsFixWKgK8v7XnVKczMWhGWCGAdhXozNlJbsJIFvP7if7z50wLmq9URfyzI18iYQTj/btwc3/ezfOtwXldka1/ALTiPp416KtNbV3fycWVgfuOg5/ObZx85OIWHREeMgNMTvXX+RMgHiAAAgAElEQVQv/33/s61Wo6mMzZRarcKK5CM3PSLGYQkgxkFoiKHuTKtVEFqAUvb7iut+Hd+1G3f+89ZzxYuPD+wa56Y1fYPs/rEXof1ZcovghKXN7oMTvOzTt7ZaDaHJvPI56+jrSAPB7iVLayYLZSby7qfkXU8Xq+7U3/HCbVx5/smLrfayZcUtghOWPrm02WoVhAXgf35+oL5QDa6742kxDm2GvDkILaUSXrPgDFyWypa3/sA/C8q/MK3kW6Ng+Xxm+N1oaODK/3ywdTcoLDire7Lc9p5fJS37Y3jIm4OwIPz3/c/ykZse8fYjDh/9fcvBfYYBZXul8jf6/tkqbgNfdBalrYSVx8LCMjJVbPpUZ6FB46CUuhp4NXBIa/1cJ2wA+CqwBdgDXKq1PqrsVT6fAS4ApoE3a63vcdK8CfhzJ9sPaa2/2LxbEZrFH3/tfoqV2jugtQtKQdo0bA+mfiPm27Ma7KNp2IZLKYVh2J5QVcjoGU6Y3+AFr33yRtVIunkbjidV14AaRjTPaHlVHZXP8Bqu/yI3LzdeVe/DP2DsnhM24OCtao7k5TPyjm2vn1fcwHStvCB0j05dOfXnz8v1Kps2FaZheCvKU/59MdzV4r5zsKfNyiLD5tHom8M1wD8C1/rC3gPcorX+iFLqPc71nwHnA9udz9nAZ4GzHWPyfmAn9tv+3UqpG7XWRxtR4L5nRvnoTY+wY3M/vTl74Ku2V9NaHjrdsGQZEryD1vZIGpWJejZt3LtoWC6Qfpb3k+TdNM4b6yufu47DEwXPbXaxoimUKhQrFoWSRaFsUSxX7GPFaulTm9bU3cpTWLm86MTVXPOWs1qtxpKkIeOgtb5VKbUlFHwR8CLn/IvAD7GNw0XAtdpuxW5XSvUrpdY7st/VWo8AKKW+C7wSuL4RHa7+yZP89IlhfvrEcCPigkOgKyjpCdgXF3jChdgnsVzaIJc2fJ46q0+lXrle+SHvnb6LWjK18q7KBNP5p1BWiwqmr3oRVT6ZcJkNlEH0ZmvJhHWby30Fw5LrrBqqvXEXd78L/4NDwO27s7+Fe66dtK4svjB/Wu3JVzPSoTziyvQ/0AR01FCyLM+FedL2rY12R77iOesaExQizGfMYa3Wer9zfgBY65xvAJ7xye11wpLCIyilLgcuB9i8eTMAn7z0NC7duYlbdx/m1l8c5pEDE/NQfeWgNfY/mX016/QBw0G1C8E1Hm4XQ2BsgmqXh9e9ovxhQVnQvi6S4FhGfJ7hsoN5QgP6GDFlhe7HL+vv9knUx0i+d7d7xi8b1DNBn0ie0Xr361X73oPdOcGyXD3t7raubMre3zqXpieXEmd7K5CmDEhrrbVSqmmdC1rrq4CrwJ6tBHa/8rnbhzh3+xDvvSA65c196tAEdwMjdK0BbRHYlazmTmI6Kktox7KaZemobGxZvnuoWZaXZ1A2qGdcnkl6Vp/eLCu625pOkA3fV+K9W0G96t97tSzC31Hk3l29oYIVK9vYd1RNV+veI/ok/ObCZS/1QXeloDubojeXprcjTV9H9bw3l7aNiBNmnwevOzOmjAUsQeZjHA4qpdZrrfc73UaHnPB9wCaf3EYnbB/Vbig3/IfzKD+A+3QEYM5xFzdBWCjm/fDi66IJb6E634eX8AOJu3BtfKbEeL7M2EzJOS8xPmOHPz0y7cVPFsp17/9jv34qlz5/U105oX2Yj3G4EXgT8BHn+A1f+LuUUl/BHpAecwzId4C/UUqtcuReDlw5j/KFJcCTR6aYcryqhtcfhPuticQH+8o1OjDAXy/PyJqHavJIH7y/fx5f3uE0+NIk9an7soz03dfUWUf18seH9UrKM6xrMD/fPYXqNlDvTl275acMxarODP0dmWD52t5Pw92DenymxOOHpyIG409veIA/veEB5sOP//TFbBronFceQuM0OpX1euyn/iGl1F7sWUcfAb6mlHob8BRwqSP+bexprI9hT2V9C4DWekQp9UHgLkfuA+7gtLA8+f4jB3nrNbKAUWgOfncbwsLT6Gyl1ydEvSRGVgNXJORzNXB1w9oJDWG/7pcC3Qxxferuudcl4Xvy9XdNzD1NsOtjqljGNBSVpd7pLswKecJfHsgK6SXOTLHC2R++hZmSPFUJtXFnLJmGCiwaNJyFZ/6FeaZRXcBnqurCPi/OOW4d6uK0jX08Z0MfHY7frPF8iYf3j4cW9AFEZ1x151IMdInH33ZkSRuHf/nxE3zoWw+3Wg1BWBJojbd2oFnc98woX79337zyuPkPz+OEtT1N0khoFkvaOKzv62i1CoIgzJM3/MsdZNNGwEWH62KD0LX/bYRwuH+tiB3pSxPNo5p3cA2PP3+/KxF/elBULIv9Y3n2HZ1hwjcA/70/eiHHr+lehJpbWJa0cXjVqet51amvarUaywr/lMvwTCF3hoqVIOOfnROXvjr7JUbOm/2SnAeB8Oj4RyN5JJ7X04/Q/c1GP8Kzh+LziNZffD0F9U7IO+Z7C363c5+5VDP/mno3kHfovuLrPZgH4bLq1U04Dwu0s05mulhhfKbERL40573Lx/PLY4fBJW0chObjXy+CrBcRmoTWPlcYluMaw3GJYemqiwy728uqyjlpLH9ay5+XRcWCsmVF4iytKVfs6+lihclCKbDx0GShzKRzdDchqtXjphSs782xcVUnGwc62Liqk02rnONAB+t6c6SWkdtwMQ6C0ETchWTVRsrC8jVekUbP14B5DWPIr1BSw2gFGsj4hrear12O3RBboQa2kYbYoqLx9PMa9EhjrWPvvR0mrHWkTbpzKXpyKXqyKbpzKYa6O+nO2i5CenIpup3wnlzak+nO2nFrenJkUsun8a+HGAdhzmgdbeySGge3kQo2hDrU0NhPgXFPjrUbRl+amk+k/rQJjXXSfWjnPsIO4WLSthOGgpRhYBh4Lq/dT8qZdZQyqzOYXJfY3rmTtjOVqqZxjuG8qudGfL4xae1zA9PAc9Ed0dGfxsvXCOhrqLh83VlXis60uaye6hcDMQ5CLAfH8/zWF+7g6HTJacStYKNu6Tn1xwr1CTdwadPwGj93bwO3ETaN+P0O3PT4B299ZSR5hw16gA0mVF5a5V1rjWfwk2R8GcaXOSu9VIxsfNp4mWhXaWKZcTL++1LQm0uzqjPDqs40q7oyDHRl6O+0w5b6znRiHIRYOjMm521fzXi+VB1w9MVXw4KRQRkdCAvnE3ALEToJu4OIKzOan/8OkmSiZdaSqVu2X7+EMsP3VKvM+DoOBoRltIZiJbi3Rvx3FryvyHfYiF6B+42XqVknoaqt/X3E/AYiZSTUeQ29av/ugjLNfBjqypj05GynhMcOdvGJ151GX2d6fpkuIGIchFh6cmn+/NWntFoNYZmy++AEf//9x5yZb7b33qoHW7tBtnwzjiyfjDvbqBEZSwedDVY9+fri8HWR+sZqSlZzN7KaKlaYKlY4MA6/ODjJ1+/dy5vP2dq8ApqMGAdBEBaFg+N5RqaKlCua1/zjT1qtTk1efspaTlrXY4+D+LrpUobCNA1f154iFbp2u/ziugDTpqLsjLU9d0Nvq2+zJmIcBEFYcPaPzXDuR3+wJPxsnbapn09eeho9ufbt8lkMxDgIghBLeEEkBBev+ccREhcuOnG5lMknLzmNPcNTjE6XGJkqcnS6aB+nioxMF8mX6u8F3t+ZZrArY7sP9w0Eu4PA/R1p+jszbHLWIQhzR4zDCuKff/gY/+9HT9CRNp19oE06Mia5lH2snht0pE3SptGUVbJQlaXGSte5rJZ1cmzqilm3DGLKr8rErAauVUboHqp1Ep9PUl3Wu0dC5SSWEa7LmHzakdHpEqPTJWCqppxpKO5630vFqd88EOOwgvjY/zwKwNjM8ljev5RQzt7MthfUqrdTpfC8nLr7OXteT535+2E5Nw/TSWP6PKa63lIj8p4H1hg5L89g+YZPH7scfOH2tVt+1edQDd9H7nWs7yO/D6Ow76NQ/rX8KznhQ91ZVrXxTKClgBiHZUqxbFEoB9143/eXL2MiXyZfqpAvWeTLFWaKFWZKFfIl+3zad+2ezxSrcjOlSuDNwe9nCV3d0tLeO7r6luHOCrEcYf+199TtT+8L88vVTO+la799m7WGstvPsoK8qxsKz+j5ndv53Xa7RsAwqmFKVRt999o1bvabr/3pcD8ZN8zwwh7YO1qVyZhkU4ZjAKvG0svX1VEFDa2/3KBM43mkTEVXduk1tUtPY6EhTvjzm1qtgiDYRlprxx62mcVeZE5Z3xtYZAjRRYd/8NLtvOTktYurWAJiHJYpV7z4OL5421PedWQBVEg+sjjJ1xdtOSuj27kvWhDamY2rOujImIlvJ25XXmemfZrk9tGkBvc8fZRS2Qp0GVS8BS3BBTRufGDBi7PKsboQhmpaHU6rqVi14/0LaypWsJyAbEQvTcVfdkx8w3mH9IrK2jM7ktKG5cNdMm7cSqTaTVCjm8HrbqgdH+hyCMg20oXhzyc4buDv2jAT4xP0inStxOkVHG9Iivfn7Y5hJHax+PQ2w/FG/bwb7dIJytb+buLcaQg2bW8cjk4X+bV/vq3VaghLDH+jYvoaIP9gqxkanHVl/IOgbl5hvzvEycQNyPoUCuZXzQNsY6zAeSgAlI71GeSdN6pTQriXo0/PeuVQJz+33rz7TqiPpHII3U+irK8csB903PEwe9ysHBg/858D/O2vPY/Xn7UZoTZzNg5KqROBr/qCtgF/CfQDbwcOO+Hv1Vp/20lzJfA27CG5d2utv1OvnP7ODP/w9rMpV5zpgfimCnp/QlMPQ1MFq+durnFTLKNTL/Gl9/JuoJxYnZLy8GUekInRuyGdEsIJpZ2TTgnlePcaVx+z0SmUX9x3FPju5qpTQGY2v6cGdKpRjv9eY/VPKsdyz62Ye/VNU9WuY8RqV6B7dL3fum+MrjdZ9y3SL7vc3xi/8JMnxTg0wJyNg9b6UWAHgFLKBPYBXwfeAnxaa/0Jv7xS6hTgMuA5wDHA95RSJ2ita87dUMCvHDc0VzWFJlHtugp2U9ndddE4fzdeUpzbUHlxvsapGufvfgt2iblxFbebrqY+1TKC92DHxd9fclzw/sAi+f60rnaDuuUH7kG7vn5C9xCKC5QZk245kTYVHWmTrmyKjoxJZ8akM5OiM2PSlYmGxZ13ZEy6siad6ep5LmViGNKV1AjN6lZ6CfC41vqpGn14FwFf0VoXgCeVUo8BZwE/nW/hR6eKHJksxP/Thf4h4/7pGvmHrMT8I0firJgyYxoF/7hJoOFx89XVcYm4OP8Yg7/hDMb500THXsL1FGjIYuKWI17XU2hQ0OuC8vXHe3EJaxTixgqq6wMU6UB3lq9/PJQuuvYgIc6IlmPUiDPdLjMV7GYLd6/51y8E1jKE9IncQ0xcWO9G1la4cbL3QutplnG4DLjed/0updQbgV3AH2utjwIbgNt9MnudsAhKqcuBywE2b67/+nfex34Q2OB7JZEybH//aVORSZlkTEU6ZZA2DTKmQc50zp0w+1yRcc5TprPRitcQ+hoLo9rghAf64hqAQIOjYhqEmMVV4dkacQ2rv5zExV++sQVVQx+/AZDBSEFIZt7GQSmVAS4ErnSCPgt8ELtb9IPAJ4G3ziZPrfVVwFUAO3furPvYes1bz+LpkanI9oVx14E9asO7ekV2+YrfTrHsO4a3akzauSxuZ7FmULY0ZauCveh5bgbS23lLuV4ng7t3BfeVrpLUtvo3W6kvG5dvvHBs6ALl24x7i03fDB3ivouEioiXTSBGOEl2NvnGbrAzq3yjgbmMyVB3htXdWYa6s6zusY9DPRlndXQGU7qP5kUz3hzOB+7RWh8EcI8ASqnPA990LvcBm3zpNjph8+bMY1dx5rGrmpHVohHecD28RWW9vYcj+/vWjK8ar+TyHENaqRpP7zpuhDLBuMUFx22ekyy7MPkm6xuNSNQhthoSdGi8yhLKa0a+Mfc2Cx1mk28S866zhPSj00UeOzjBkckixUrUYZ+hYKAraxsQ13B0Z5xjlqGeLB1pM1FvS2vHk4A902m6aJ//0rZBnruhLzHdcqIZxuH1+LqUlFLrtdb7ncuLgZ855zcC1ymlPoU9IL0duLMJ5S9JlHK2d0z+fQrCsiQ8wO8fZNehyQfhccCArDMZYGymxKGJPPtH8zw9Ms0zR6ft48g0RyYLPHJgoqn67/nIq5qaX7syL+OglOoCXga8wxf8MaXUDuwHjz1unNb650qprwEPYfd/XFFvppIg+BflBRcyRicVBBZIxsnq4CC+fzKB9uXnH8APTzAIL46MW1wZJxu3SDFRVvtkrZDu/kYzroGNue/ZyNauo2h8vcWVcbKtIpuyvQwXy8E3jYxpkEsb3mynXNqe8dSRsf0yVc9TXPC8da1RvgXMyzhoraeAwVDYb9WQ/zDw4fmUuRwZnS5y1t/cEvnR1qM3l6K3I02vsy9tTy6NOcdB1vAK6aTV2bGNcZKsr3HwPwEGVnhHGr9gY7aScAfwAwPwisigvX/lb3BVcm3ZuBXa/plU/hlPcSubzUAZySuo41Zgu5McGpL1zazy7tM3wcEIza5y7yVtKmeKayrUqNsfmcI6O9p+hfRKoFi2Zm0YAMbzZcbzZWCm+UrVIXYKpfsP7M4cCs00SvtWKKtQmvAsIneAPL5BCJXtnYdnUEUbpPB0UL8egbJjG1274YqUF2rgwjOnImX76sALJ7hyOYl6tr9e81dvhlb99LXS1sl7nm3zbMquWJqJfJmJfLmhsuvX6/y/E0tXvSUXyhb5UoVj+js4eX17bhcqxqENWNObq9uPWbE0pYpFoWwxWSgzNl1iPF9ifKZkG4kZ97rM2Ex8nPuP0gwsDVZFkzxsKQhCI/z4T1/MpoH227VOjMMSwX6arfqw/+pdz3BksgBE3TikTcVgV4bBrkzARUPZ0kzmy4znS4zNlCmUKvFuHai6dgjPFnGvk3YNi+5Y5s816E7CCiXWupqv3+aEdfTnHe/2I6hLNU30HgVhIVAKcimTTMogmzLIpg2yKXtPiVzaPmZTBuccP8TGVR2tVjcWMQ5LkK/fu5e/v2V3q9UQhBWLUnDhacfY3YPgrQVS+D2+Vh0Uul2IQVnFgbE8H/7WwxQrFgVnA65Cye56yvuO/nCAL/322Zy0bmG7o8Q4LEEuOXMTubTJeL4c67mTSFi0X9TdhtGNt+V9aX15xKWNXazkBPqj3Cd2f97uE7xSalZz5sPUSho3l16hIo7//Pdhv7kEdXf1dd9majk59L+lEHrLqev0L+FNSIfKrOqqIzq46b1wX5lhPcNhrmBYp6R7hOg9xTuudLTSNHSPYUeJ4frCFx75PuLqwKdDsMyq7rv2HGVylh4WtIZv3PfsrNI0k28/sF+MgxDFMBQX7Yj1PDIvvnT7U3zoWw9VZwnV6LqJ93YrCII7NdadKZVLB2dN5TLR7U3tayN47UvvboGaTZlk0wY9i7DtqBgHweP5WwZ4069ssee2J0xP9W9SFLceIG4To+AU1uAc+npz7N0w97zimxpr6WA5yesH2mOe/UogbsZW1cFfdRZZxD+WJ++kDclEfGMZ1ZljEceEnnx1xpo7Ay3ef1eoLCMoE+fvy/Tp6c3GMw26siZp04jMfvNPMw5PwY2flqzo70yTq7GKe6ER4yB4nLiuhyvPP7nVaiw49dZnhHcT9C/iijWGruGKWRDnX1wXXs+RZEzdhXDJxjTGa28N3fyegJOMqf+BICk+utYlrr7w6ZUcH7c4rqIdNy7l4PqYYH1UjX/4fsP6LnVOXNvDd/7wvJaVL8ZBWHH41ysIyxMdNmiht884Nx1+I+May4rWTBcqjEwXOTpV5KhzHJkucnS6xMikHTYyVWR0uhTr5ymJnlyK/s40/R0Z+jvT9Hak6e9I09+Zpq8jzWkb+xewhuojxkEQhGWHOyPIQNnbiOYr3vqfsRl7DVDwuhyJG5ux1wjVavDTpqLPadw3DXTS35Gmz2nw+5yG3m3s7esM/R1penKptt+zQoyDIAhLkrv2jPDzfWPeQs9wYz/W4OJP01CeK5o+xx3N+r4OejtSXsPvPtHbT/cZr9HvSJvLdl8QMQ6CICwJwvtvv/PL93B4ojDvfCuWZjxfZqZUYWSySMpU3sZY7rm7qZZ7nXHOU4a9eVbKsK8zPhl/urRp727nnRv2MVhO9dqVyaYNTlzb0xIDJMahDXAH8NyZOO6GQ+FN4u2wavxnf/g4N9yzt27+Z20dcApyD8F/Mu88pFM4PEmWGFl/cGJ5DZQdL1tb/zjdwxdx5cXpXrO8mEHPpLLj8ku+/3ilq/U5y/Ji7q/pdRGbb+O/s+T7XxzcvU/yzN7H2ULzBy/dzh+89IRFL3fZGIfJQpkrvnwPD+wdTZzCGB6kWik8dmgysDLTUNWFbIa7cC3krdNd7elOtXPl3RWfhhENC64C9cn6pvAFw/1pgwvrwov77LDoAr9EmZi8IH6hnz+PRmSJKS9O90b0VwkZ18svsS4S9G9UH38ewTpehLpIULReHg2VN8+6COoZLxumbGnKFYtSxfaLVrY0xbJF2bIoVzTFin0sWxbFsg6El5y4lKm4ZOem5EIWkGVjHB49MM6PfnG41Wq0JSNTxVarMCv8hskzNM6Kbv91QC7pHPcf2B8eTe+VG5e3F6ci6YnIBNMTSJN8TwGXC0l5N3pPkfpyDXw1PZH7DKZvuL68h4yqjprwPdSqk1A5MXn76yexThrJO1Qn/u+s+pCUnD5YJ1W5dvasOh+WjXE489gB7n//yylXLG+RDOAtsde6umzecsKq7gKiMt4rfFwc1SX5Oi6PpPN66X1xGs3IVJHJfDmwiMyd1+7OFa9Y1Xnf4fnxFasqFz/fvcaCttBcdq+LKyEusujNip97Hl5b4L8f/yY9bteZK1fW1rKYuy4sPzKmwcMffOWymxq9bIwDQF9HutUqNI19ozO86eo7KVWkRRRWLv79OZSqdn369w5x3yzC+4f43woCe4N4ccGVz4roPiVheVMRulas6c3y+R8/EXxDwdcN5b6VOGE7NvVz+ub23/N+WRmH5cJtjx3hU9/9hRgGYcXjvnWGhtOXPC86cXWsQ0E/4Qkd/t6MuPBPXHJaU/eFEOPQJB45MM4r/+7HrVZDEIQlwA8fbf746N/fspuPX3Ja0/IT49AEKpamvASe8k/f3B8a/FSh6+grsUvYHbf3yhwTX02rEmTd6/DrdzCcGmW48klxJJXRoA61dE/OM1hpiWWE42vo3rDeoUIblQ/rMKs08/2tqCboPYvfSrXM2epdW4fk34qKLX9Oeid+39X4nVua21U1b+OglNoDTAAVoKy13qmUGgC+CmwB9gCXaq2PKvvOPwNcAEwDb9Za3zNfHWpRqljMlCrkixVmShWmnaN7PVOqMFOskC8lxVuR+LyTxk0/l/2fF5KUoRjqzjLYnWGoO8tQd5YrLziJoe5sq1UTBGGJ0Kw3hxdrrY/4rt8D3KK1/ohS6j3O9Z8B5wPbnc/ZwGedY9P41/99kn/+4eNe415u0YKGjrRJV9YkmzJJm7Y73pRh2EfTHuRKOe6BU6bCNIzqdeDohJvBTeuDboCjT5hhrv7Jk955+IkpjErMxSdTX6R+Lg1k0kAxi3I/zdDDlmkkpyaUU0fjpnx/DeTTrN9S/Tzm/1tqXp3UltJaM12qMFUoM1WoMJEv2+fFMpOFMqdt7Oe9F5xMJtVa30sL1a10EfAi5/yLwA+xjcNFwLXaHnm5XSnVr5Rar7Xe36yCt6/p4bztq31TSX1TKcFzkawJemP0ywHeNE+/nH+qZ3VarBselqlOS7W0plTRFMrlZJ3COgTKsnWpR9zuZxGZOiKNmNJGdm+rJyHTUgUhnnufHuUNv3Qsx6/pbqkezTAOGrhZKaWB/6e1vgpY62vwDwBrnfMNwDO+tHudsKYZh3O3D3Hu9qFmZddUJgtlnh2d8a4V1UbUfdYIX4dRqtqwJj+gxOVcm2Y/Nfm3Ak3KK2wfYt1mRGSiErFuLELycYYzafZHrfLiywptnxkV8daw1MonfC/x+SS7zAjIJPnHoLquJlRyTB7hfEMysXn7dAvVb5w6sa49Iml0TFj0Ow6XH9Ahpszk8hLqo8Z9+PN2w296cD8/mOPA80s/9aPAdU82RY/jHLAnl6Inl+Z5G/p490u2L9j6imYYh3O11vuUUmuA7yqlHvFHaq21YzgaRil1OXA5wObNm5ug4vyYLJTRWntznuNI+lH6ee77v7MQ6gmCsMyZKJSZKJR5dizvhX3/kUN85pbdEdnv/dELm1LmvI2D1nqfczyklPo6cBZw0O0uUkqtBw454vsAv6OQjU5YOM+rgKsAdu7c2dIOiG8+8Czvuu7eVqogCILQMJUmjbPOyzgopboAQ2s94Zy/HPgAcCPwJuAjzvEbTpIbgXcppb6CPRA91szxhoXgrK0DnLi2h32jM57nVL8bi8XihSesDnT9KGKmThLuHoo6s4tM3wvLqJhBRBXsUvJP0Qumqy1DKO+60xP9YXWmDcalj5eJvvmF7zl8L16Yf7pnKONwPbrRqgEZQvnW/Y7iZGKm0dZzbtfolFR/ulrTdyP1Fp6iG6NXXLqITOz3oaJp6tR1WF9Xt0a+o3r/V+G8I/k6565r77Ab8GzKmcBimpim8uJa6ZJjvm8Oa4GvO5WSAq7TWv+PUuou4GtKqbcBTwGXOvLfxp7G+hj2VNa3zLP8BWdNT66p+7je+/RRLv7n22adTpwKCoIAtqHJpgyyKZNsyiCXto/ZtEEuZZJNN2eW07yMg9b6CSCyJE9rPQy8JCZcA1fMpowH942x5T3fqiljhnyqGArHz0rQd0p4KqhSQV8tfp8qgXM3X8NXRkyZphH0wxIZbHMGCvs60kwXywx0ZVCokGO+qrD7ZhJwyOfLU/vi3TB3tpU3oOgbnHRnToXTC4KwdCy/QC0AAAz4SURBVNAa8iWLfGlh11ctixXSFUtTAZaS/5W0qejJpTF9q0Vd4lYb+7sAAq/2MSsq/a+5sV02vvRxXQtxZfl1SyornCdhXb008WW5grX1D5ZFnP4JZVWP8bLhe/aXhU8usaxwN0oN2biy4r/LGukjXTqhe00oq5qmznfp785J+C0llVW9//iyknQOlF+jrGqauPqJ//+JW5WcVFb0/pPqN/n7Cf/+41f4x91/zD3X+X4rFhTKFQpli3ypwnkfZd60vXF43oY+dn3kVQDBHdF0dGc0NywQ77iuDsQ7rqH9u6+5bqstK5QuIhtOH9IrsnubnffB8TwP7x/n8cNTVCx73cOzozOcuK6Hk9f3csr6Xk5e38vWoS5v8ZwgCEKraHvj4McwFAaK9BJuN/OlCrsPTvLjxw7zsf95lHufHuXep0fnladS0JVJ0ZU16c6m+MOXnUBvbvm4LxcEwSaXNjlxbQ99nQv//60aWe3aSnbu3Kl37drVajWaTr1xFEEQhHpsG+ri91+6nYt2bAiEK6Xu1lrvnE/eYhzmgGVpHj88SamiA6tj3ZWp1cHl4ECy/3yyUOa6O54OuNlw00TyCuXr392ubGnHR4u9SKZQsvj1MzeSdfyyhHeXCw58R/OP7pIX1S+sQ828E/KYzT0m6hFJn5B3RO+EeombANDofSXkb+nG7kkQ5sOTf3tBaAxq/sZhSXUrtQs33LOX//sfD7RajUSuv/PpVqswK9zBvFntGQ2++Pi9fQmkieYBYBh2uH1MyDuUnnB4KG8CaRq8r1D+xmzvKZJ3vA6EdA3nUa/OomEN5h2pl2geJMUl1btfp1p5J9VLzbyDeQTrJZpH0u/FUHH10mDeAd3ifyuFksXqnmzAMDQLMQ5z4GWnrOXEtT08enCCd77oOE7b1D/rH031H6b5PxwjsTGpkX42+vnk8O61vn5GuJwF+EELgtAcxDjMgf7ODNe+7Swu+dxPue7Op7n49A1sX9vTarUEQRCaRmsdhi9h1vbm+NLbziZjGrzhC3fwzMh0q1USBEFoGmIc5sHmwU7+7W1nky9ZvOELd3BoIl8/kSAIwhJAjMM8OXFdD//6ludzeKLAG79wJ2PTpVarJAiCMG/EODSBMzav4qrf2skTh6d4yzV3Ml0st1olQRCEeSED0g5aa54dy7PnyJTto6RkeX5KCmUrEOb6MCmULKZLFf77/me9fO55epRT/tLe1Odb7z6X5xzT16pbEgRBmDMr1jiUKhYPPTvO3U8d9T4HxuuPGSiF5xY3mzJIm8kvXz1ZcWEhCMLSZMUYh6NTRe55umoI7t876rm83dDfwfO3DnDm5n5OXNdLRybkJz1lkHXOU44rb0EQhOXMsjYOo9NFXv0PP2Hv0ZlA+LreHBefvpFzjh/kzGNXsb6vo0UaCoIgtCfLdkD6qeEpPnHzoxHDAHBgPM/1dz7NP/3gcTKmweh0kbGZEhP5EtPFsjPOUKFcsbAsTbv7nxIEQWg2y/bN4YUf/2FdmYf3j3Pmh77XUH6uW4rw5t3vOf8kfueFx81FRUEQhLZl2RqHL73tbL79s/0MdWX497v3sn9sfgvUtLY3/glzxuZV88pXEAShHVm2xuHc7UOcu30IgFedegzX3fEUuw9N8ouDkxyZLMSm6c6mOH5NNyes7WZ9Xwdx486mUuzY3M/ztwyQW8q7DgmCINRgzvs5KKU2AdcCa7Hd01+ltf6MUuqvgLcDhx3R92qtv+2kuRJ4G1AB3q21/k69chZiP4ejU0XHUEzwmHMMG42ujMnxa3s4YU03xw52sqorw6+fsVEMgiAIbU+r93MoA3+stb5HKdUD3K2U+q4T92mt9Sf8wkqpU4DLgOcAxwDfU0qdoLWuzEOHObGqK8NZWwc4a+tAINw1GrsPTXDjfc9yx5Mj3P9MdQvPp4enufKCkxdbXUEQhEVnzsZBa70f2O+cTyilHgY21EhyEfAVrXUBeFIp9RhwFvDTuerQbFyjsefIFPc9M8pQd5bztg/R15mmI23yDhl4FgRhhdCUMQel1BbgdOAO4BzgXUqpNwK7sN8ujmIbjtt9yfZS25gsOjPFCn/xjZ/xH3fv5ZzjB/m73zid1T3ZVqslCIKw6Mx7nYNSqhu4AfgDrfU48FngOGAH9pvFJ+eQ5+VKqV1KqV2HDx+un6AJPH54kov/+X+54Z69vPtXj+fat54thkEQhBXLvN4clFJpbMPwZa31fwJorQ/64j8PfNO53Ads8iXf6IRF0FpfBVwF9oD0fHRshLufOsobv3AHmZTBNW85ixeesHqhixQEQWhr5vzmoGwHQ18AHtZaf8oXvt4ndjHwM+f8RuAypVRWKbUV2A7cOdfym8UzI9Ncfu0uVvdk+da7XyCGQRAEgfm9OZwD/BbwoFLqPifsvcDrlVI7sKe37gHeAaC1/rlS6mvAQ9gzna5oxUwll8ecGUlf27WXUsXiC29+Psf0i48lQRAEmN9spZ8Ace5Jv10jzYeBD8+1zGZwcDzPW6+5i58/O45S8MvbBvmTV5zIcau7W6mWIAhCW7FsHe81gsL2mfSLAxOyvacgCIKPOa+QXiwWYoU0wO6DE/z3/c9y4/3Psmd4mrSpeOEJq3nNacfw0pPX0pVdtp5FBEFY5jRjhfSKNQ4uWmt+tm+cG+/fx3/fv58D43k60iZvf8FW/ujlJy5YuYIgCAtFq91nLAuUUjxvYx/P29jHleefzK6njnLj/ftYL4PTgiCsYFa8cfBjGCrW55IgCMJKY0UPSAuCIAjxiHEQBEEQIohxEARBECKIcRAEQRAiiHEQBEEQIohxEARBECKIcRAEQRAiiHEQBEEQIrS9+wyl1GHgqVbrUYMh4EirlWgA0bO5iJ7NRfRsLidqrXvmk0Hbr5DWWrf17jtKqV3z9WGyGIiezUX0bC6iZ3NRSs3bIZ10KwmCIAgRxDgIgiAIEcQ4zJ+rWq1Ag4iezUX0bC6iZ3OZt55tPyAtCIIgLD7y5iAIgiBEEOPQIEqpTUqpHyilHlJK/Vwp9ftO+F8ppfYppe5zPhe0WlcApdQepdSDjk67nLABpdR3lVK7neOqFut4oq/e7lNKjSul/qAd6lQpdbVS6pBS6me+sNj6UzZ/r5R6TCn1gFLqjBbr+XGl1COOLl9XSvU74VuUUjO+ev1ci/VM/J6VUlc69fmoUuoVLdbzqz4d9yil7nPCW1mfSe1R836jWmv5NPAB1gNnOOc9wC+AU4C/Av6k1frF6LsHGAqFfQx4j3P+HuCjrdbTp5sJHACObYc6Bc4DzgB+Vq/+gAuAmwAF/BJwR4v1fDmQcs4/6tNzi1+uDeoz9nt2/q/uB7LAVuBxwGyVnqH4TwJ/2Qb1mdQeNe03Km8ODaK13q+1vsc5nwAeBja0VqtZcxHwRef8i8BrW6hLmJcAj2ut22LBo9b6VmAkFJxUfxcB12qb24F+pdT6Vumptb5Za112Lm8HNi6GLrVIqM8kLgK+orUuaK2fBB4Dzlow5XzU0lMppYBLgesXQ5da1GiPmvYbFeMwB5RSW4DTgTucoHc5r2pXt7qrxocGblZK3a2UutwJW6u13u+cHwDWtka1WC4j+E/XjnWaVH8bgGd8cntpnweHt2I/MbpsVUrdq5T6kVLqBa1Sykfc99yu9fkC4KDWercvrOX1GWqPmvYbFeMwS5RS3cANwB9orceBzwLHATuA/divne3AuVrrM4DzgSuUUuf5I7X9rtkWU9WUUhngQuDfnaB2rVOPdqq/JJRS7wPKwJedoP3AZq316cAfAdcppXpbpR9L4HsO8XqCDzAtr8+Y9shjvr9RMQ6zQCmVxv4ivqy1/k8ArfVBrXVFa20Bn2eRXn/robXe5xwPAV/H1uug+yrpHA+1TsMA5wP3aK0PQvvWKcn1tw/Y5JPb6IS1DKXUm4FXA7/pNBI43TTDzvnd2H35J7RKxxrfczvWZwr4NeCrblir6zOuPaKJv1ExDg3i9Dd+AXhYa/0pX7i/3+5i4GfhtIuNUqpLKdXjnmMPUP4MuBF4kyP2JuAbrdEwQuCJrB3r1CGp/m4E3ujMCPklYMz3ar/oKKVeCfwpcKHWetoXvlopZTrn24DtwBOt0bLm93wjcJlSKquU2oqt552LrV+IlwKPaK33ugGtrM+k9ohm/kZbMdK+FD/AudivaA8A9zmfC4B/Ax50wm8E1reBrtuwZ3vcD/wceJ8TPgjcAuwGvgcMtIGuXcAw0OcLa3mdYhur/UAJu3/2bUn1hz0D5J+wnxwfBHa2WM/HsPuX3d/p5xzZX3d+D/cB9wCvabGeid8z8D6nPh8Fzm+lnk74NcDvhGRbWZ9J7VHTfqOyQloQBEGIIN1KgiAIQgQxDoIgCEIEMQ6CIAhCBDEOgiAIQgQxDoIgCEIEMQ6CIAhCBDEOgiAIQgQxDoIgCEKE/w8lYCgA21mUIAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot.line()으로 그려봅니다.\n",
    "df3.plot.line()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "엉망입니다... 아무래도 x축을 sort해줘야 하는것 같습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17f8a32b0>"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sort_value와 비슷한 개념으로 sort_index가 있습니다. method chaining방법으로 그려봅니다.\n",
    "df3."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "attrs 컬럼을 보면 참 많은 정보들이 담겨있습니다.\n",
    "\n",
    "세미 콜론으로 분리되어있고, 분리되어있는 항목들이 달라보입니다.\n",
    "\n",
    "이중 우리는 유전자의 id 혹은 transcript id는 꼭 필요해 보입니다. 따로 분리 해야겠습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q? Apply를 이용하여 'ID=' 혹은 'Name='로 시작하는 데이터를 분리하여 df에 새로운 컬럼 'id'을 추가 해보세요**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                   external_name=oe %3D 0.79;logic_name=cpg\n",
       "1                                         logic_name=eponine\n",
       "2                                         logic_name=eponine\n",
       "3                                         logic_name=eponine\n",
       "4                                         logic_name=eponine\n",
       "5                                         logic_name=eponine\n",
       "6                                         logic_name=eponine\n",
       "7                                         logic_name=eponine\n",
       "8                                         logic_name=eponine\n",
       "9                                         logic_name=eponine\n",
       "10                                        logic_name=eponine\n",
       "11                                        logic_name=eponine\n",
       "12                                        logic_name=eponine\n",
       "13                                        logic_name=eponine\n",
       "14                                        logic_name=eponine\n",
       "15         ID=gene:ENSG00000223972;Name=DDX11L1;biotype=t...\n",
       "16         ID=transcript:ENST00000456328;Parent=gene:ENSG...\n",
       "17         Parent=transcript:ENST00000456328;Name=ENSE000...\n",
       "18         Parent=transcript:ENST00000456328;Name=ENSE000...\n",
       "19         Parent=transcript:ENST00000456328;Name=ENSE000...\n",
       "20         ID=transcript:ENST00000450305;Parent=gene:ENSG...\n",
       "21         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "22         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "23         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "24         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "25         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "26         Parent=transcript:ENST00000450305;Name=ENSE000...\n",
       "27         ID=gene:ENSG00000227232;Name=WASH7P;biotype=un...\n",
       "28         ID=transcript:ENST00000488147;Parent=gene:ENSG...\n",
       "29         Parent=transcript:ENST00000488147;Name=ENSE000...\n",
       "                                 ...                        \n",
       "2890629    ID=transcript:ENST00000420810;Parent=gene:ENSG...\n",
       "2890630    Parent=transcript:ENST00000420810;Name=ENSE000...\n",
       "2890631    ID=gene:ENSG00000227629;Name=SLC25A15P1;biotyp...\n",
       "2890632    ID=transcript:ENST00000456738;Parent=gene:ENSG...\n",
       "2890633    Parent=transcript:ENST00000456738;Name=ENSE000...\n",
       "2890634    Parent=transcript:ENST00000456738;Name=ENSE000...\n",
       "2890635    Parent=transcript:ENST00000456738;Name=ENSE000...\n",
       "2890636    ID=gene:ENSG00000237917;Name=PARP4P1;biotype=u...\n",
       "2890637    ID=transcript:ENST00000435945;Parent=gene:ENSG...\n",
       "2890638    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890639    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890640    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890641    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890642    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890643    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890644    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890645    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890646    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890647    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890648    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890649    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890650    Parent=transcript:ENST00000435945;Name=ENSE000...\n",
       "2890651    ID=gene:ENSG00000231514;Name=CCNQP2;biotype=pr...\n",
       "2890652    ID=transcript:ENST00000435741;Parent=gene:ENSG...\n",
       "2890653    Parent=transcript:ENST00000435741;Name=ENSE000...\n",
       "2890654          external_name=rank %3D 1;logic_name=firstef\n",
       "2890655          external_name=rank %3D 1;logic_name=firstef\n",
       "2890656    ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...\n",
       "2890657    ID=transcript:ENST00000431853;Parent=gene:ENSG...\n",
       "2890658    Parent=transcript:ENST00000431853;Name=ENSE000...\n",
       "Name: attrs, Length: 2890659, dtype: object"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 먼저 attrs를 봅시다.\n",
    "df['attrs']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'ID=transcript:ENST00000450305;Parent=gene:ENSG00000223972;Name=DDX11L1-201;biotype=transcribed_unprocessed_pseudogene;tag=basic;transcript_id=ENST00000450305;transcript_support_level=NA;version=2'"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 한 셀만 자세히 살펴보죠 (iloc이용)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 탐색이 끝났습니다. ';'으로 분리하고 'ID=' 혹은 'Name='로 시작하는 데이터를 가지고 오는 함수를 만들어야 하는군요\n",
    "\n",
    "import numpy as np # 원하는 조건이 없을 때, np.nan으로 missing값 처리 하려고\n",
    "\n",
    "def get_id(row):\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                               NaN\n",
       "1                               NaN\n",
       "2                               NaN\n",
       "3                               NaN\n",
       "4                               NaN\n",
       "5                               NaN\n",
       "6                               NaN\n",
       "7                               NaN\n",
       "8                               NaN\n",
       "9                               NaN\n",
       "10                              NaN\n",
       "11                              NaN\n",
       "12                              NaN\n",
       "13                              NaN\n",
       "14                              NaN\n",
       "15          ID=gene:ENSG00000223972\n",
       "16    ID=transcript:ENST00000456328\n",
       "17             Name=ENSE00002234944\n",
       "18             Name=ENSE00003582793\n",
       "19             Name=ENSE00002312635\n",
       "20    ID=transcript:ENST00000450305\n",
       "Name: attrs, dtype: object"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# apply 써봅니다.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "      <th>gene_len</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10469</td>\n",
       "      <td>11240</td>\n",
       "      <td>1.3e+03</td>\n",
       "      <td>.</td>\n",
       "      <td>.</td>\n",
       "      <td>external_name=oe %3D 0.79;logic_name=cpg</td>\n",
       "      <td>772</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10650</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10655</td>\n",
       "      <td>10657</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10678</td>\n",
       "      <td>10687</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10681</td>\n",
       "      <td>10688</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10707</td>\n",
       "      <td>10716</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10708</td>\n",
       "      <td>10718</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>11</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10735</td>\n",
       "      <td>10747</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>13</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10737</td>\n",
       "      <td>10744</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10766</td>\n",
       "      <td>10773</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10770</td>\n",
       "      <td>10779</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10796</td>\n",
       "      <td>10801</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10810</td>\n",
       "      <td>10819</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10870</td>\n",
       "      <td>10872</td>\n",
       "      <td>0.999</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1</td>\n",
       "      <td>.</td>\n",
       "      <td>biological_region</td>\n",
       "      <td>10889</td>\n",
       "      <td>10893</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>logic_name=eponine</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>11869</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000223972;Name=DDX11L1;biotype=t...</td>\n",
       "      <td>2541</td>\n",
       "      <td>ID=gene:ENSG00000223972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>lnc_RNA</td>\n",
       "      <td>11869</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000456328;Parent=gene:ENSG...</td>\n",
       "      <td>2541</td>\n",
       "      <td>ID=transcript:ENST00000456328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>11869</td>\n",
       "      <td>12227</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>359</td>\n",
       "      <td>Name=ENSE00002234944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12613</td>\n",
       "      <td>12721</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>109</td>\n",
       "      <td>Name=ENSE00003582793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>13221</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>1189</td>\n",
       "      <td>Name=ENSE00002312635</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   chr  source               type  start    end    score strand phase  \\\n",
       "0    1       .  biological_region  10469  11240  1.3e+03      .     .   \n",
       "1    1       .  biological_region  10650  10657    0.999      +     .   \n",
       "2    1       .  biological_region  10655  10657    0.999      -     .   \n",
       "3    1       .  biological_region  10678  10687    0.999      +     .   \n",
       "4    1       .  biological_region  10681  10688    0.999      -     .   \n",
       "5    1       .  biological_region  10707  10716    0.999      +     .   \n",
       "6    1       .  biological_region  10708  10718    0.999      -     .   \n",
       "7    1       .  biological_region  10735  10747    0.999      -     .   \n",
       "8    1       .  biological_region  10737  10744    0.999      +     .   \n",
       "9    1       .  biological_region  10766  10773    0.999      +     .   \n",
       "10   1       .  biological_region  10770  10779    0.999      -     .   \n",
       "11   1       .  biological_region  10796  10801    0.999      +     .   \n",
       "12   1       .  biological_region  10810  10819    0.999      -     .   \n",
       "13   1       .  biological_region  10870  10872    0.999      +     .   \n",
       "14   1       .  biological_region  10889  10893    0.999      -     .   \n",
       "15   1  havana         pseudogene  11869  14409        .      +     .   \n",
       "16   1  havana            lnc_RNA  11869  14409        .      +     .   \n",
       "17   1  havana               exon  11869  12227        .      +     .   \n",
       "18   1  havana               exon  12613  12721        .      +     .   \n",
       "19   1  havana               exon  13221  14409        .      +     .   \n",
       "\n",
       "                                                attrs  gene_len  \\\n",
       "0            external_name=oe %3D 0.79;logic_name=cpg       772   \n",
       "1                                  logic_name=eponine         8   \n",
       "2                                  logic_name=eponine         3   \n",
       "3                                  logic_name=eponine        10   \n",
       "4                                  logic_name=eponine         8   \n",
       "5                                  logic_name=eponine        10   \n",
       "6                                  logic_name=eponine        11   \n",
       "7                                  logic_name=eponine        13   \n",
       "8                                  logic_name=eponine         8   \n",
       "9                                  logic_name=eponine         8   \n",
       "10                                 logic_name=eponine        10   \n",
       "11                                 logic_name=eponine         6   \n",
       "12                                 logic_name=eponine        10   \n",
       "13                                 logic_name=eponine         3   \n",
       "14                                 logic_name=eponine         5   \n",
       "15  ID=gene:ENSG00000223972;Name=DDX11L1;biotype=t...      2541   \n",
       "16  ID=transcript:ENST00000456328;Parent=gene:ENSG...      2541   \n",
       "17  Parent=transcript:ENST00000456328;Name=ENSE000...       359   \n",
       "18  Parent=transcript:ENST00000456328;Name=ENSE000...       109   \n",
       "19  Parent=transcript:ENST00000456328;Name=ENSE000...      1189   \n",
       "\n",
       "                               id  \n",
       "0                             NaN  \n",
       "1                             NaN  \n",
       "2                             NaN  \n",
       "3                             NaN  \n",
       "4                             NaN  \n",
       "5                             NaN  \n",
       "6                             NaN  \n",
       "7                             NaN  \n",
       "8                             NaN  \n",
       "9                             NaN  \n",
       "10                            NaN  \n",
       "11                            NaN  \n",
       "12                            NaN  \n",
       "13                            NaN  \n",
       "14                            NaN  \n",
       "15        ID=gene:ENSG00000223972  \n",
       "16  ID=transcript:ENST00000456328  \n",
       "17           Name=ENSE00002234944  \n",
       "18           Name=ENSE00003582793  \n",
       "19           Name=ENSE00002312635  "
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 그럼 이제 id라는 컬럼을 새로 넣어 봅시다.\n",
    "\n",
    "df.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "아무래도 유전자 이름도 모르는곳은 저에게는 의미가 없어 보입니다.\n",
    "\n",
    "id 열이 missing인 행은 다 지워 버리기로 하죠."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chr</th>\n",
       "      <th>source</th>\n",
       "      <th>type</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>score</th>\n",
       "      <th>strand</th>\n",
       "      <th>phase</th>\n",
       "      <th>attrs</th>\n",
       "      <th>gene_len</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>11869</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000223972;Name=DDX11L1;biotype=t...</td>\n",
       "      <td>2541</td>\n",
       "      <td>ID=gene:ENSG00000223972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>lnc_RNA</td>\n",
       "      <td>11869</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000456328;Parent=gene:ENSG...</td>\n",
       "      <td>2541</td>\n",
       "      <td>ID=transcript:ENST00000456328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>11869</td>\n",
       "      <td>12227</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>359</td>\n",
       "      <td>Name=ENSE00002234944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12613</td>\n",
       "      <td>12721</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>109</td>\n",
       "      <td>Name=ENSE00003582793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>13221</td>\n",
       "      <td>14409</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456328;Name=ENSE000...</td>\n",
       "      <td>1189</td>\n",
       "      <td>Name=ENSE00002312635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>12010</td>\n",
       "      <td>13670</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000450305;Parent=gene:ENSG...</td>\n",
       "      <td>1661</td>\n",
       "      <td>ID=transcript:ENST00000450305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12010</td>\n",
       "      <td>12057</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>48</td>\n",
       "      <td>Name=ENSE00001948541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12179</td>\n",
       "      <td>12227</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>49</td>\n",
       "      <td>Name=ENSE00001671638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12613</td>\n",
       "      <td>12697</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>85</td>\n",
       "      <td>Name=ENSE00001758273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>12975</td>\n",
       "      <td>13052</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>78</td>\n",
       "      <td>Name=ENSE00001799933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>13221</td>\n",
       "      <td>13374</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>154</td>\n",
       "      <td>Name=ENSE00001746346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>13453</td>\n",
       "      <td>13670</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000450305;Name=ENSE000...</td>\n",
       "      <td>218</td>\n",
       "      <td>Name=ENSE00001863096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>14404</td>\n",
       "      <td>29570</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000227232;Name=WASH7P;biotype=un...</td>\n",
       "      <td>15167</td>\n",
       "      <td>ID=gene:ENSG00000227232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>14404</td>\n",
       "      <td>29570</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000488147;Parent=gene:ENSG...</td>\n",
       "      <td>15167</td>\n",
       "      <td>ID=transcript:ENST00000488147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>14404</td>\n",
       "      <td>14501</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>98</td>\n",
       "      <td>Name=ENSE00001843071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>15005</td>\n",
       "      <td>15038</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>34</td>\n",
       "      <td>Name=ENSE00001935574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>15796</td>\n",
       "      <td>15947</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>152</td>\n",
       "      <td>Name=ENSE00002030414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>16607</td>\n",
       "      <td>16765</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>159</td>\n",
       "      <td>Name=ENSE00003621279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>16858</td>\n",
       "      <td>17055</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>198</td>\n",
       "      <td>Name=ENSE00003553898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>17233</td>\n",
       "      <td>17368</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>136</td>\n",
       "      <td>Name=ENSE00003502542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>17606</td>\n",
       "      <td>17742</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>137</td>\n",
       "      <td>Name=ENSE00003475637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>17915</td>\n",
       "      <td>18061</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>147</td>\n",
       "      <td>Name=ENSE00003565697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>18268</td>\n",
       "      <td>18366</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>99</td>\n",
       "      <td>Name=ENSE00003477500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>24738</td>\n",
       "      <td>24891</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>154</td>\n",
       "      <td>Name=ENSE00003507205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>29534</td>\n",
       "      <td>29570</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000488147;Name=ENSE000...</td>\n",
       "      <td>37</td>\n",
       "      <td>Name=ENSE00001890219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>1</td>\n",
       "      <td>mirbase</td>\n",
       "      <td>ncRNA_gene</td>\n",
       "      <td>17369</td>\n",
       "      <td>17436</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000278267;Name=MIR6859-1;biotype...</td>\n",
       "      <td>68</td>\n",
       "      <td>ID=gene:ENSG00000278267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>1</td>\n",
       "      <td>mirbase</td>\n",
       "      <td>miRNA</td>\n",
       "      <td>17369</td>\n",
       "      <td>17436</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000619216;Parent=gene:ENSG...</td>\n",
       "      <td>68</td>\n",
       "      <td>ID=transcript:ENST00000619216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>1</td>\n",
       "      <td>mirbase</td>\n",
       "      <td>exon</td>\n",
       "      <td>17369</td>\n",
       "      <td>17436</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000619216;Name=ENSE000...</td>\n",
       "      <td>68</td>\n",
       "      <td>Name=ENSE00003746039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>ncRNA_gene</td>\n",
       "      <td>29554</td>\n",
       "      <td>31109</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000243485;Name=MIR1302-2HG;bioty...</td>\n",
       "      <td>1556</td>\n",
       "      <td>ID=gene:ENSG00000243485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>1</td>\n",
       "      <td>havana</td>\n",
       "      <td>lnc_RNA</td>\n",
       "      <td>29554</td>\n",
       "      <td>31097</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000473358;Parent=gene:ENSG...</td>\n",
       "      <td>1544</td>\n",
       "      <td>ID=transcript:ENST00000473358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890627</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26579589</td>\n",
       "      <td>26579690</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000258589;Name=ENSE000...</td>\n",
       "      <td>102</td>\n",
       "      <td>Name=ENSE00002035744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890628</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26549425</td>\n",
       "      <td>26549743</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000224240;Name=CYCSP49;biotype=p...</td>\n",
       "      <td>319</td>\n",
       "      <td>ID=gene:ENSG00000224240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890629</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26549425</td>\n",
       "      <td>26549743</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000420810;Parent=gene:ENSG...</td>\n",
       "      <td>319</td>\n",
       "      <td>ID=transcript:ENST00000420810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890630</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26549425</td>\n",
       "      <td>26549743</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000420810;Name=ENSE000...</td>\n",
       "      <td>319</td>\n",
       "      <td>Name=ENSE00001705100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890631</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26586642</td>\n",
       "      <td>26591601</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000227629;Name=SLC25A15P1;biotyp...</td>\n",
       "      <td>4960</td>\n",
       "      <td>ID=gene:ENSG00000227629</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890632</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26586642</td>\n",
       "      <td>26591601</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000456738;Parent=gene:ENSG...</td>\n",
       "      <td>4960</td>\n",
       "      <td>ID=transcript:ENST00000456738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890633</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26586642</td>\n",
       "      <td>26586766</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "      <td>125</td>\n",
       "      <td>Name=ENSE00001652425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890634</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26587398</td>\n",
       "      <td>26587560</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "      <td>163</td>\n",
       "      <td>Name=ENSE00001691206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890635</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26591548</td>\n",
       "      <td>26591601</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000456738;Name=ENSE000...</td>\n",
       "      <td>54</td>\n",
       "      <td>Name=ENSE00001632993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890636</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26594851</td>\n",
       "      <td>26634652</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000237917;Name=PARP4P1;biotype=u...</td>\n",
       "      <td>39802</td>\n",
       "      <td>ID=gene:ENSG00000237917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890637</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26594851</td>\n",
       "      <td>26634652</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000435945;Parent=gene:ENSG...</td>\n",
       "      <td>39802</td>\n",
       "      <td>ID=transcript:ENST00000435945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890638</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26594851</td>\n",
       "      <td>26595045</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>195</td>\n",
       "      <td>Name=ENSE00001663098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890639</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26600891</td>\n",
       "      <td>26601022</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>132</td>\n",
       "      <td>Name=ENSE00001744948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890640</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26611713</td>\n",
       "      <td>26611807</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>95</td>\n",
       "      <td>Name=ENSE00001647502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890641</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26614154</td>\n",
       "      <td>26614962</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>809</td>\n",
       "      <td>Name=ENSE00001702742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890642</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26615294</td>\n",
       "      <td>26615531</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>238</td>\n",
       "      <td>Name=ENSE00001747631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890643</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26621773</td>\n",
       "      <td>26621895</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>123</td>\n",
       "      <td>Name=ENSE00001687652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890644</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26622512</td>\n",
       "      <td>26622608</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>97</td>\n",
       "      <td>Name=ENSE00001752207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890645</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26623586</td>\n",
       "      <td>26623666</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>81</td>\n",
       "      <td>Name=ENSE00001670663</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890646</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26627943</td>\n",
       "      <td>26628022</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>80</td>\n",
       "      <td>Name=ENSE00001725096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890647</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26628271</td>\n",
       "      <td>26628437</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>167</td>\n",
       "      <td>Name=ENSE00001741452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890648</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26630647</td>\n",
       "      <td>26630749</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>103</td>\n",
       "      <td>Name=ENSE00001681574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890649</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26633345</td>\n",
       "      <td>26633431</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>87</td>\n",
       "      <td>Name=ENSE00001638296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890650</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26634523</td>\n",
       "      <td>26634652</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435945;Name=ENSE000...</td>\n",
       "      <td>130</td>\n",
       "      <td>Name=ENSE00001797328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890651</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>26626520</td>\n",
       "      <td>26627159</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000231514;Name=CCNQP2;biotype=pr...</td>\n",
       "      <td>640</td>\n",
       "      <td>ID=gene:ENSG00000231514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890652</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>26626520</td>\n",
       "      <td>26627159</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000435741;Parent=gene:ENSG...</td>\n",
       "      <td>640</td>\n",
       "      <td>ID=transcript:ENST00000435741</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890653</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>26626520</td>\n",
       "      <td>26627159</td>\n",
       "      <td>.</td>\n",
       "      <td>-</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000435741;Name=ENSE000...</td>\n",
       "      <td>640</td>\n",
       "      <td>Name=ENSE00001616687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890656</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogene</td>\n",
       "      <td>56855244</td>\n",
       "      <td>56855488</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...</td>\n",
       "      <td>245</td>\n",
       "      <td>ID=gene:ENSG00000235857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890657</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>pseudogenic_transcript</td>\n",
       "      <td>56855244</td>\n",
       "      <td>56855488</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>ID=transcript:ENST00000431853;Parent=gene:ENSG...</td>\n",
       "      <td>245</td>\n",
       "      <td>ID=transcript:ENST00000431853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890658</th>\n",
       "      <td>Y</td>\n",
       "      <td>havana</td>\n",
       "      <td>exon</td>\n",
       "      <td>56855244</td>\n",
       "      <td>56855488</td>\n",
       "      <td>.</td>\n",
       "      <td>+</td>\n",
       "      <td>.</td>\n",
       "      <td>Parent=transcript:ENST00000431853;Name=ENSE000...</td>\n",
       "      <td>245</td>\n",
       "      <td>Name=ENSE00001794473</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2405878 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        chr   source                    type     start       end score strand  \\\n",
       "15        1   havana              pseudogene     11869     14409     .      +   \n",
       "16        1   havana                 lnc_RNA     11869     14409     .      +   \n",
       "17        1   havana                    exon     11869     12227     .      +   \n",
       "18        1   havana                    exon     12613     12721     .      +   \n",
       "19        1   havana                    exon     13221     14409     .      +   \n",
       "20        1   havana  pseudogenic_transcript     12010     13670     .      +   \n",
       "21        1   havana                    exon     12010     12057     .      +   \n",
       "22        1   havana                    exon     12179     12227     .      +   \n",
       "23        1   havana                    exon     12613     12697     .      +   \n",
       "24        1   havana                    exon     12975     13052     .      +   \n",
       "25        1   havana                    exon     13221     13374     .      +   \n",
       "26        1   havana                    exon     13453     13670     .      +   \n",
       "27        1   havana              pseudogene     14404     29570     .      -   \n",
       "28        1   havana  pseudogenic_transcript     14404     29570     .      -   \n",
       "29        1   havana                    exon     14404     14501     .      -   \n",
       "30        1   havana                    exon     15005     15038     .      -   \n",
       "31        1   havana                    exon     15796     15947     .      -   \n",
       "32        1   havana                    exon     16607     16765     .      -   \n",
       "33        1   havana                    exon     16858     17055     .      -   \n",
       "34        1   havana                    exon     17233     17368     .      -   \n",
       "35        1   havana                    exon     17606     17742     .      -   \n",
       "36        1   havana                    exon     17915     18061     .      -   \n",
       "37        1   havana                    exon     18268     18366     .      -   \n",
       "38        1   havana                    exon     24738     24891     .      -   \n",
       "39        1   havana                    exon     29534     29570     .      -   \n",
       "41        1  mirbase              ncRNA_gene     17369     17436     .      -   \n",
       "42        1  mirbase                   miRNA     17369     17436     .      -   \n",
       "43        1  mirbase                    exon     17369     17436     .      -   \n",
       "50        1   havana              ncRNA_gene     29554     31109     .      +   \n",
       "51        1   havana                 lnc_RNA     29554     31097     .      +   \n",
       "...      ..      ...                     ...       ...       ...   ...    ...   \n",
       "2890627   Y   havana                    exon  26579589  26579690     .      +   \n",
       "2890628   Y   havana              pseudogene  26549425  26549743     .      +   \n",
       "2890629   Y   havana  pseudogenic_transcript  26549425  26549743     .      +   \n",
       "2890630   Y   havana                    exon  26549425  26549743     .      +   \n",
       "2890631   Y   havana              pseudogene  26586642  26591601     .      -   \n",
       "2890632   Y   havana  pseudogenic_transcript  26586642  26591601     .      -   \n",
       "2890633   Y   havana                    exon  26586642  26586766     .      -   \n",
       "2890634   Y   havana                    exon  26587398  26587560     .      -   \n",
       "2890635   Y   havana                    exon  26591548  26591601     .      -   \n",
       "2890636   Y   havana              pseudogene  26594851  26634652     .      -   \n",
       "2890637   Y   havana  pseudogenic_transcript  26594851  26634652     .      -   \n",
       "2890638   Y   havana                    exon  26594851  26595045     .      -   \n",
       "2890639   Y   havana                    exon  26600891  26601022     .      -   \n",
       "2890640   Y   havana                    exon  26611713  26611807     .      -   \n",
       "2890641   Y   havana                    exon  26614154  26614962     .      -   \n",
       "2890642   Y   havana                    exon  26615294  26615531     .      -   \n",
       "2890643   Y   havana                    exon  26621773  26621895     .      -   \n",
       "2890644   Y   havana                    exon  26622512  26622608     .      -   \n",
       "2890645   Y   havana                    exon  26623586  26623666     .      -   \n",
       "2890646   Y   havana                    exon  26627943  26628022     .      -   \n",
       "2890647   Y   havana                    exon  26628271  26628437     .      -   \n",
       "2890648   Y   havana                    exon  26630647  26630749     .      -   \n",
       "2890649   Y   havana                    exon  26633345  26633431     .      -   \n",
       "2890650   Y   havana                    exon  26634523  26634652     .      -   \n",
       "2890651   Y   havana              pseudogene  26626520  26627159     .      -   \n",
       "2890652   Y   havana  pseudogenic_transcript  26626520  26627159     .      -   \n",
       "2890653   Y   havana                    exon  26626520  26627159     .      -   \n",
       "2890656   Y   havana              pseudogene  56855244  56855488     .      +   \n",
       "2890657   Y   havana  pseudogenic_transcript  56855244  56855488     .      +   \n",
       "2890658   Y   havana                    exon  56855244  56855488     .      +   \n",
       "\n",
       "        phase                                              attrs  gene_len  \\\n",
       "15          .  ID=gene:ENSG00000223972;Name=DDX11L1;biotype=t...      2541   \n",
       "16          .  ID=transcript:ENST00000456328;Parent=gene:ENSG...      2541   \n",
       "17          .  Parent=transcript:ENST00000456328;Name=ENSE000...       359   \n",
       "18          .  Parent=transcript:ENST00000456328;Name=ENSE000...       109   \n",
       "19          .  Parent=transcript:ENST00000456328;Name=ENSE000...      1189   \n",
       "20          .  ID=transcript:ENST00000450305;Parent=gene:ENSG...      1661   \n",
       "21          .  Parent=transcript:ENST00000450305;Name=ENSE000...        48   \n",
       "22          .  Parent=transcript:ENST00000450305;Name=ENSE000...        49   \n",
       "23          .  Parent=transcript:ENST00000450305;Name=ENSE000...        85   \n",
       "24          .  Parent=transcript:ENST00000450305;Name=ENSE000...        78   \n",
       "25          .  Parent=transcript:ENST00000450305;Name=ENSE000...       154   \n",
       "26          .  Parent=transcript:ENST00000450305;Name=ENSE000...       218   \n",
       "27          .  ID=gene:ENSG00000227232;Name=WASH7P;biotype=un...     15167   \n",
       "28          .  ID=transcript:ENST00000488147;Parent=gene:ENSG...     15167   \n",
       "29          .  Parent=transcript:ENST00000488147;Name=ENSE000...        98   \n",
       "30          .  Parent=transcript:ENST00000488147;Name=ENSE000...        34   \n",
       "31          .  Parent=transcript:ENST00000488147;Name=ENSE000...       152   \n",
       "32          .  Parent=transcript:ENST00000488147;Name=ENSE000...       159   \n",
       "33          .  Parent=transcript:ENST00000488147;Name=ENSE000...       198   \n",
       "34          .  Parent=transcript:ENST00000488147;Name=ENSE000...       136   \n",
       "35          .  Parent=transcript:ENST00000488147;Name=ENSE000...       137   \n",
       "36          .  Parent=transcript:ENST00000488147;Name=ENSE000...       147   \n",
       "37          .  Parent=transcript:ENST00000488147;Name=ENSE000...        99   \n",
       "38          .  Parent=transcript:ENST00000488147;Name=ENSE000...       154   \n",
       "39          .  Parent=transcript:ENST00000488147;Name=ENSE000...        37   \n",
       "41          .  ID=gene:ENSG00000278267;Name=MIR6859-1;biotype...        68   \n",
       "42          .  ID=transcript:ENST00000619216;Parent=gene:ENSG...        68   \n",
       "43          .  Parent=transcript:ENST00000619216;Name=ENSE000...        68   \n",
       "50          .  ID=gene:ENSG00000243485;Name=MIR1302-2HG;bioty...      1556   \n",
       "51          .  ID=transcript:ENST00000473358;Parent=gene:ENSG...      1544   \n",
       "...       ...                                                ...       ...   \n",
       "2890627     .  Parent=transcript:ENST00000258589;Name=ENSE000...       102   \n",
       "2890628     .  ID=gene:ENSG00000224240;Name=CYCSP49;biotype=p...       319   \n",
       "2890629     .  ID=transcript:ENST00000420810;Parent=gene:ENSG...       319   \n",
       "2890630     .  Parent=transcript:ENST00000420810;Name=ENSE000...       319   \n",
       "2890631     .  ID=gene:ENSG00000227629;Name=SLC25A15P1;biotyp...      4960   \n",
       "2890632     .  ID=transcript:ENST00000456738;Parent=gene:ENSG...      4960   \n",
       "2890633     .  Parent=transcript:ENST00000456738;Name=ENSE000...       125   \n",
       "2890634     .  Parent=transcript:ENST00000456738;Name=ENSE000...       163   \n",
       "2890635     .  Parent=transcript:ENST00000456738;Name=ENSE000...        54   \n",
       "2890636     .  ID=gene:ENSG00000237917;Name=PARP4P1;biotype=u...     39802   \n",
       "2890637     .  ID=transcript:ENST00000435945;Parent=gene:ENSG...     39802   \n",
       "2890638     .  Parent=transcript:ENST00000435945;Name=ENSE000...       195   \n",
       "2890639     .  Parent=transcript:ENST00000435945;Name=ENSE000...       132   \n",
       "2890640     .  Parent=transcript:ENST00000435945;Name=ENSE000...        95   \n",
       "2890641     .  Parent=transcript:ENST00000435945;Name=ENSE000...       809   \n",
       "2890642     .  Parent=transcript:ENST00000435945;Name=ENSE000...       238   \n",
       "2890643     .  Parent=transcript:ENST00000435945;Name=ENSE000...       123   \n",
       "2890644     .  Parent=transcript:ENST00000435945;Name=ENSE000...        97   \n",
       "2890645     .  Parent=transcript:ENST00000435945;Name=ENSE000...        81   \n",
       "2890646     .  Parent=transcript:ENST00000435945;Name=ENSE000...        80   \n",
       "2890647     .  Parent=transcript:ENST00000435945;Name=ENSE000...       167   \n",
       "2890648     .  Parent=transcript:ENST00000435945;Name=ENSE000...       103   \n",
       "2890649     .  Parent=transcript:ENST00000435945;Name=ENSE000...        87   \n",
       "2890650     .  Parent=transcript:ENST00000435945;Name=ENSE000...       130   \n",
       "2890651     .  ID=gene:ENSG00000231514;Name=CCNQP2;biotype=pr...       640   \n",
       "2890652     .  ID=transcript:ENST00000435741;Parent=gene:ENSG...       640   \n",
       "2890653     .  Parent=transcript:ENST00000435741;Name=ENSE000...       640   \n",
       "2890656     .  ID=gene:ENSG00000235857;Name=CTBP2P1;biotype=p...       245   \n",
       "2890657     .  ID=transcript:ENST00000431853;Parent=gene:ENSG...       245   \n",
       "2890658     .  Parent=transcript:ENST00000431853;Name=ENSE000...       245   \n",
       "\n",
       "                                    id  \n",
       "15             ID=gene:ENSG00000223972  \n",
       "16       ID=transcript:ENST00000456328  \n",
       "17                Name=ENSE00002234944  \n",
       "18                Name=ENSE00003582793  \n",
       "19                Name=ENSE00002312635  \n",
       "20       ID=transcript:ENST00000450305  \n",
       "21                Name=ENSE00001948541  \n",
       "22                Name=ENSE00001671638  \n",
       "23                Name=ENSE00001758273  \n",
       "24                Name=ENSE00001799933  \n",
       "25                Name=ENSE00001746346  \n",
       "26                Name=ENSE00001863096  \n",
       "27             ID=gene:ENSG00000227232  \n",
       "28       ID=transcript:ENST00000488147  \n",
       "29                Name=ENSE00001843071  \n",
       "30                Name=ENSE00001935574  \n",
       "31                Name=ENSE00002030414  \n",
       "32                Name=ENSE00003621279  \n",
       "33                Name=ENSE00003553898  \n",
       "34                Name=ENSE00003502542  \n",
       "35                Name=ENSE00003475637  \n",
       "36                Name=ENSE00003565697  \n",
       "37                Name=ENSE00003477500  \n",
       "38                Name=ENSE00003507205  \n",
       "39                Name=ENSE00001890219  \n",
       "41             ID=gene:ENSG00000278267  \n",
       "42       ID=transcript:ENST00000619216  \n",
       "43                Name=ENSE00003746039  \n",
       "50             ID=gene:ENSG00000243485  \n",
       "51       ID=transcript:ENST00000473358  \n",
       "...                                ...  \n",
       "2890627           Name=ENSE00002035744  \n",
       "2890628        ID=gene:ENSG00000224240  \n",
       "2890629  ID=transcript:ENST00000420810  \n",
       "2890630           Name=ENSE00001705100  \n",
       "2890631        ID=gene:ENSG00000227629  \n",
       "2890632  ID=transcript:ENST00000456738  \n",
       "2890633           Name=ENSE00001652425  \n",
       "2890634           Name=ENSE00001691206  \n",
       "2890635           Name=ENSE00001632993  \n",
       "2890636        ID=gene:ENSG00000237917  \n",
       "2890637  ID=transcript:ENST00000435945  \n",
       "2890638           Name=ENSE00001663098  \n",
       "2890639           Name=ENSE00001744948  \n",
       "2890640           Name=ENSE00001647502  \n",
       "2890641           Name=ENSE00001702742  \n",
       "2890642           Name=ENSE00001747631  \n",
       "2890643           Name=ENSE00001687652  \n",
       "2890644           Name=ENSE00001752207  \n",
       "2890645           Name=ENSE00001670663  \n",
       "2890646           Name=ENSE00001725096  \n",
       "2890647           Name=ENSE00001741452  \n",
       "2890648           Name=ENSE00001681574  \n",
       "2890649           Name=ENSE00001638296  \n",
       "2890650           Name=ENSE00001797328  \n",
       "2890651        ID=gene:ENSG00000231514  \n",
       "2890652  ID=transcript:ENST00000435741  \n",
       "2890653           Name=ENSE00001616687  \n",
       "2890656        ID=gene:ENSG00000235857  \n",
       "2890657  ID=transcript:ENST00000431853  \n",
       "2890658           Name=ENSE00001794473  \n",
       "\n",
       "[2405878 rows x 11 columns]"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dropna(subset=['id'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "완성!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "여기까지 GFF데이터를 이용해서 데이터를 읽고, 정제하고 데이터를 이해하며 변형 해보는 시간을 가졌습니다.\n",
    "\n",
    "분명 같은 데이터에도 자신의 목적에 따라 하고싶은 데이터 탐색 방법이 다를겁니다. 여기서 끝내지 마시고 돌아가셔서 GFF파일에서 더 만져볼 것은 없는가? 아이디어 내보시고 코딩하시다보면 나만의 데이터가 생겼을 때 자신있게 탐색을 시작하실 수 있을 겁니다.\n",
    "\n",
    "아이디어가 필요한 분들께 마지막으로 문제 하나 만들어 드리고 본 notebook을 마치겠습니다.\n",
    "\n",
    "***\n",
    "\n",
    "### Q? GFF에서 임의의 타겟 영역을 지정해 주면, 해당하는 유전자와 type, strand direction을 출력해주는 스크립트 혹은 notebook 를 작성해보세요\n",
    "### 스크립트를 작성시 실행 코드 화면은 다음과 같습니다.\n",
    "`python genefinder.py chr01 18540951`\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "hint. filtering과 apply함수를 이용합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}