rnateam.de-novo.yaml

---
id: denovo
name: De novo transcriptome reconstruction with RNA-seq
description: >-
  In this tutorial, we will identify what transcripts are present in the G1E and
  megakaryocyte cellualr states and which transcripts are differentially
  expressed between the two states. We will use a de novo transcript
  reconstruction strategy to infer transcript structures from the mapped reads
  in the absence of the actual annotated transcript structures. This will allow
  us to identify novel transcripts and novel isoforms of known transcripts, as
  well as identify differentially expressed transcripts.
title_default: denovo
tags:
  - "RNA"

steps:
  - title: Introduction
    content: >-
      The data provided here are part of a Galaxy tutorial that analyzes RNA-seq
      data from a study published by Wu et al. in 2014 <a
      href="http://genome.cshlp.org/content/early/2014/10/12/gr.164830.113.abstract">DOI:10.1101/gr.164830.113</a>.
      The goal of this study was to investigate “the dynamics of occupancy and
      the role in gene regulation of the transcription factor Tal1, a critical
      regulator of hematopoiesis, at multiple stages of hematopoietic
      differentiation.” To this end, RNA-seq libraries were constructed from
      multiple mouse cell types including G1E - a GATA-null immortalized cell
      line derived from targeted disruption of GATA-1 in mouse embryonic stem
      cells - and megakaryocytes. This RNA-seq data was used to determine
      differential gene expression between G1E and megakaryocytes and later
      correlated with Tal1 occupancy. This dataset (GEO Accession: GSE51338)
      consists of biological replicate, paired-end, poly(A) selected RNA-seq
      libraries. Because of the long processing time for the large original
      files, we have downsampled the original raw data files to include only
      reads that align to chromosome 19 and a subset of interesting genomic loci
      identified by Wu et al.
    backdrop: true
  - title: Introduction
    content: >-
      In this tutorial, we will identify what transcripts are present in the G1E
      and megakaryocyte cellualr states and which transcripts are differentially
      expressed between the two states. We will use a de novo transcript
      reconstruction strategy to infer transcript structures from the mapped
      reads in the absence of the actual annotated transcript structures. This
      will allow us to identify novel transcripts and novel isoforms of known
      transcripts, as well as identify differentially expressed transcripts.
    backdrop: true
  - title: Data upload
    content: >-
      Due to the large size of this dataset, we have downsampled it to only
      include reads mapping to chromosome 19 and certain loci with relevance to
      hematopoeisis. This data is available at <a
      href="https://zenodo.org/record/583140#.WSW3NhPyub8">Zenodo</a>, where you
      can find the forward and reverse reads corresponding to replicate RNA-seq
      libraries from G1E and megakaryocyte cells and an annotation file of
      RefSeq transcripts we will use to generate our transcriptome database.
    backdrop: true
  - title: History options
    element: '#history-options-button'
    content: >-
      We will start the analyses by creating a new history. Click on this button
      and then "Create New"
    placement: left
  - title: Uploading the new data
    element: '#tool-panel-upload-button .fa.fa-upload'
    content: We need to upload data. Open the Galaxy Upload Manager
    placement: right
    postclick:
      - '#tool-panel-upload-button .fa.fa-upload'
      - '#btn-reset'
  - title: Uploading the input data
    element: '#btn-new'
    content: Click on Paste/Fetch Data
    placement: right
    postclick:
      - '#btn-new'
  - title: Uploading the input data
    element: .upload-text-column .upload-text .upload-text-content.form-control
    content: >-
      Insert the links here.<br> The input is available on <a
      href="https://zenodo.org/record/583140">Zenodo</a>.
    placement: right
    textinsert: >-
      https://zenodo.org/record/583140/files/G1E_rep1_forward_read_%28SRR549355_1%29

      https://zenodo.org/record/583140/files/G1E_rep1_reverse_read_%28SRR549355_2%29

      https://zenodo.org/record/583140/files/G1E_rep2_forward_read_%28SRR549356_1%29

      https://zenodo.org/record/583140/files/G1E_rep2_reverse_read_%28SRR549356_2%29

      https://zenodo.org/record/583140/files/Megakaryocyte_rep1_forward_read_%28SRR549357_1%29

      https://zenodo.org/record/583140/files/Megakaryocyte_rep1_reverse_read_%28SRR549357_2%29

      https://zenodo.org/record/583140/files/Megakaryocyte_rep2_forward_read_%28SRR549358_1%29

      https://zenodo.org/record/583140/files/Megakaryocyte_rep2_reverse_read_%28SRR549358_2%29

      https://zenodo.org/record/583140/files/RefSeq_reference_GTF_%28DSv2%29
  - title: Data types
    content: |-
      Make sure the datatypes are adjusted.
      <ul>
        <li>Set the datatype of the read files to fastqsanger</li>
        <li>Set the datatype of the annotation file to gtf and assign the Genome as mm10</li>
      </ul>
    backdrop: false
  - title: Importing data via links
    content: You will need to fetch the link to the annotation file yourself ;)
    backdrop: false
  - title: Uploading the input data
    element: '#btn-start'
    content: Click on "Start" to start loading the data to history
    placement: right
    postclick:
      - '#btn-start'
  - title: Uploading the input data
    element: '#btn-close'
    content: >-
      The upload may take a while.<br> Hit the close button to close this
      window.
    placement: right
    postclick:
      - '#btn-close'
  - title: Quality control
    content: >-
      For quality control, we use similar tools as described in <a
      href="https://galaxyproject.github.io/training-material/topics/sequence-analysis/">NGS-QC
      tutorial</a>: <a
      href="https://www.bioinformatics.babraham.ac.uk/projects/fastqc/">FastQC</a>
      and <a
      href="http://www.usadellab.org/cms/?page=trimmomatic">Trimmomatic</a>.
    backdrop: true
  - title: Quality control
    element: '#tool-search-query'
    content: >-
      Run FastQC on the forward and reverse read files to assess the quality of
      the reads. Search for FastQC tool
    placement: right
    textinsert: FastQC
  - title: Quality control
    element: '#tool-search'
    content: Click on the "FastQC" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fdevteam%2Ffastqc%2Ffastqc%2F0.68"]
  - title: Quality control
    element: '#tool-search'
    content: >-
      Run FastQC on the forward and reverse read files to assess the quality of
      the reads.
    position: left
  - title: Questions
    content: |-
      <ul>
        <li>What is the read length?</li>
        <li>Is there anything interesting about the quality of the base calls based on the position in the reads?</li>
      </ul>
    backdrop: false
  - title: Trimmomatic
    content: >-
      Trim off the low quality bases from the ends of the reads to increase
      mapping efficiency. Run Trimmomatic on each pair of forward and reverse
      reads.
    backdrop: true
  - title: Trimmomatic
    element: '#tool-search-query'
    content: Search for Trimmomatic tool
    placement: right
    textinsert: Trimmomatic
  - title: Trimmomatic
    element: '#tool-search'
    content: Click on the "Trimmomatic" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fpjbriggs%2Ftrimmomatic%2Ftrimmomatic%2F0.36.5"]
        .tool-old-link
  - title: Trimmomatic
    element: '#tool-search'
    content: Run Trimmomatic on each pair of forward and reverse reads.
    position: left
  - title: Quality control
    element: '#tool-search-query'
    content: Look for FastQC tool.
    placement: right
    textinsert: FastQC
  - title: Quality control
    element: '#tool-search'
    content: Click on the "FastQC" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fdevteam%2Ffastqc%2Ffastqc%2F0.68"]
  - title: Quality control
    element: '#tool-search'
    content: Re-run FastQC on trimmed reads and inspect the differences.
    position: left
  - title: Questions
    content: |-
      <ul>
        <li>What is the read length?</li>
        <li>Is there anything interesting about the quality of the base calls based on the position in the reads?</li>
      </ul>
    backdrop: false
  - title: Quality control
    content: >-
      Now that we have trimmed our reads and are fortunate that there is a
      reference genome assembly for mouse, we will align our trimmed reads to
      the genome. <br><br>Instead of running a single tool multiple times on all
      your data, would you rather run a single tool on multiple datasets at
      once? Check out the <a
      href="https://galaxyproject.org/tutorials/collections/">dataset
      collections</a> feature of Galaxy!
    backdrop: true
  - title: Mapping
    content: >-
      To make sense of the reads, their positions within mouse genome must be
      determined. This process is known as aligning or ‘mapping’ the reads to
      the reference genome.

      <br><br>Do you want to learn more about the principles behind mapping?
      Follow our <a
      href="https://galaxyproject.github.io/training-material/topics/sequence-analysis/">training</a>
    backdrop: true
  - title: Mapping
    content: >-
      In the case of a eukaryotic transcriptome, most reads originate from
      processed mRNAs lacking introns. Therefore, they cannot be simply mapped
      back to the genome as we normally do for reads derived from DNA sequences.
      Instead, the reads must be separated into two categories:
        <ul>
          <li>Reads contained within mature exons - these align perfectly to the reference genome</li>
          <li>Reads that span splice junctions in the mature mRNA - these align with gaps to the reference genome</li>
        </ul>
      Spliced mappers have been developed to efficiently map transcript-derived
      reads against genomes. <a
      href="https://ccb.jhu.edu/software/hisat2/index.shtml>HISAT</a> is an
      accurate and fast tool for mapping spliced reads to a genome. Another
      popular spliced aligner is <a
      href="https://ccb.jhu.edu/software/tophat/index.shtml">TopHat</a>, but we
      will be using HISAT2 in this tutorial.
    backdrop: true
  - title: Spliced mapping
    element: '#tool-search'
    content: Run HISAT2 on one forward/reverse read pair
    position: left
  - title: Spliced mapping
    element: '#tool-search-query'
    content: Search for HISAT2 tool.
    placement: right
    textinsert: HISAT
  - title: Spliced mapping
    element: '#tool-search'
    content: Click on the "HISAT2" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fhisat2%2Fhisat2%2F2.1.0"]
        .tool-old-link
  - title: Spliced mapping
    element: '#tool-search'
    content: |-
      Run the tool with the following settings:
        <ul>
          <li><b>Source for the reference genome to align against:</b>Use a built-in genome > Mouse (Mus Musculus): mm10</li>
          <li><b>Single-end or paired-end reads?</b>Paired-end</li>
          <li><b>Spliced alignment parameters:</b>Specify spliced alignment parameters</li>
          <li><b>Specify strand information:</b> Forward(FR)</li>
          <li><b>Spliced alignment options > Penalty for non-canonical splice sites:</b> 3</li>
          <li><b>Penalty function for long introns with canonical splice sites:</b> Constant [f(x) = B]</li>
          <li><b>Constant term (B):</b> 0.0</li>
          <li><b>Penalty function for long introns with non-canonical splice sites:</b> Constant [f(x) = B]</li>
          <li><b>Transcriptome assembly reporting:</b>Report alignments tailored for transcript assemblers including StringTie.</li>
        </ul>
    position: left
  - title: Spliced mapping
    content: >-
      Run HISAT2 on the remaining forward/reverse read pairs with the same
      parameters.
    backdrop: true
  - title: De novo transcript reconstruction
    content: >-
      Now that we have mapped our reads to the mouse genome with <i>HISAT</i>,
      we want to determine transcript structures that are represented by the
      aligned reads. This is called de novo transcriptome reconstruction. This
      unbiased approach permits the comprehensive identification of all
      transcripts present in a sample, including annotated genes, novel isoforms
      of annotated genes, and novel genes. While common gene/transcript
      databases are quite large, they are not comprehensive, and the de novo
      transcriptome reconstruction approach ensures complete transcriptome(s)
      identification from the experimental samples. The leading tool for
      transcript reconstruction is <i>Stringtie</i>. Here, we will use
      <i>Stringtie</i> to predict transcript structures based on the reads
      aligned by <i>HISAT2</i>.
    backdrop: true
  - title: De novo transcript reconstruction
    element: '#tool-search-query'
    content: Search for Stringtie tool.
    placement: right
    textinsert: Stringtie
  - title: De novo transcript reconstruction
    element: '#tool-search'
    content: Click on the "Stringtie" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fstringtie%2Fstringtie%2F1.3.3.2"]
        .tool-old-link
  - title: De novo transcript reconstruction
    element: '#tool-search'
    content: Run Stringtie on the HISAT2 alignments using the default parameters.
    position: left
  - title: Transcriptome assembly
    content: >-
      We just generated four transcriptomes with <i>Stringtie</i> representing
      each of the four RNA-seq libraries we are analyzing. Since these were
      generated in the absence of a reference transcriptome, and we ultimately
      would like to know what transcript structure corresponds to which
      annotated transcript (if any), we have to make a transcriptome database.
      We will use the tool <i>Stringtie - Merge</i> to combine redundant
      transcript structures across the four samples and the RefSeq reference.
      Once we have merged our transcript structures, we will use
      <i>GFFcompare</i> to annotate the transcripts of our newly created
      transcriptome so we know the relationship of each transcript to the RefSeq
      reference.
    backdrop: true
  - title: Transcriptome reconstruction
    element: '#tool-search-query'
    content: Search for Stringtie-merge tool.
    placement: right
    textinsert: Stringtie-merge
  - title: Transcriptome reconstruction
    element: '#tool-search'
    content: Click on the "Stringtie-merge" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fstringtie%2Fstringtie_merge%2F1.3.3"]
        .tool-old-link
  - title: Transcriptome reconstruction
    element: '#tool-search'
    content: >-
      Run Stringtie-merge on the Stringtie assembled transcripts along with the
      RefSeq annotation file we imported earlier.
        <ul>
          <li>Use batch mode to inlcude all four Stringtie assemblies as “input_gtf”.</li>
          <li>Select the “RefSeq GTF mm10” file as the “guide_gff”.</li>
        </ul>
    position: left
  - title: Transcriptome reconstruction
    element: '#tool-search-query'
    content: Search for GFFCompare tool.
    placement: right
    textinsert: GFFCompare
  - title: Transcriptome reconstruction
    element: '#tool-search'
    content: Click on the "GFFCompare" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fgffcompare%2Fgffcompare%2F0.9.8"]
        .tool-old-link
  - title: Transcriptome reconstruction
    element: '#tool-search'
    content: >-
      Run GFFCompare on the Stringtie-merge generated transcriptome along with
      the RefSeq annotation file.
        <ul>
          <li>Select the output of Stringtie-merge as the GTF input.</li>
          <li>Select <b>“Yes”</b> under Use Reference Annotation" and select the <b>"RefSeq GTF mm10"</b> file as the "Reference Annotation".</li>
        </ul>
    position: left
  - title: Analysis of the differential gene expression
    content: >-
      We just generated a transriptome database that represents the transcripts
      present in the G1E and megakaryocytes samples. This database provides the
      location of our transcripts with non-redundant identifiers, as well as
      information regarding the origin of the transcript.

      <br><br>We now want to identify which transcripts are differentially
      expressed between the G1E and megakaryocyte cellular states. To do this we
      will implement a counting approach using <b>FeatureCounts</b> to count
      reads per transcript. Then we will provide this information to
      <b>DESeq2</b> to generate normalized transcript counts (abundance
      estimates) and significance testing for differential expression.
    backdrop: true
  - title: Count the number of reads per transcript
    content: >-
      To compare the abundance of transcripts between different cellular states,
      the first essential step is to quantify the number of reads per
      transcript. <a
      href="http://bioinf.wehi.edu.au/featureCounts/">FeatureCounts</a> is one
      of the most popular tools for counting reads in genomic features. In our
      case, we’ll be using <b>FeatureCounts</b> to count reads aligning in exons
      of our <b>GFFCompare</b> generated transcriptome database.

      <br><br>The recommended mode is “union”, which counts overlaps even if a
      read only shares parts of its sequence with a genomic feature and
      disregards reads that overlap more than one feature.
    backdrop: true
  - title: Counting the number of reads per transcript
    element: '#tool-search-query'
    content: Search for FeatureCounts tool.
    placement: right
    textinsert: FeatureCounts
  - title: Counting the number of reads per transcript
    element: '#tool-search'
    content: Click on the "FeatureCounts" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Ffeaturecounts%2Ffeaturecounts%2F1.6.0.6"]
        .tool-old-link
  - title: Counting the number of reads per transcript
    element: '#tool-search'
    content: >-
      Run FeatureCounts on the aligned reads (HISAT output) using the GFFCompare
      transcriptome database as the annotation file.
        <ul>
          <li>Using the batch mode for input selection, choose the four HISAT2 aligned read files</li>
          <li><b>Specify strand information:</b> Stranded (Forward)</li>
          <li><b>Gene annotation file:</b> in your history, then select the annotated transcripts GTF file output by GFFCompare (this specifies the “union” mode)</li>
          <li>Expand <b>Advanced options</b></li>
          <li><b>GFF gene identifier:</b> enter “transcript_id”</li>
        </ul>
    position: left
  - title: Perform differential gene expression testing
    content: >-
      Transcript expression is estimated from read counts, and attempts are made
      to correct for variability in measurements using replicates. This is
      absolutely essential to obtaining accurate results. We recommend having at
      least two biological replicates.

      <br><br><a
      href="https://bioconductor.org/packages/release/bioc/html/DESeq2.html">DESeq2</a>
      is a great tool for differential gene expression analysis. It accepts read
      counts produced by FeatureCounts and applies size factor normalization:
        <ul>
          <li>Computation for each gene of the geometric mean of read counts across all samples</li>
          <li>Division of every gene count by the geometric mean</li>
          <li>Use of the median of these ratios as sample’s size factor for normalization</li>
        </ul>
    backdrop: true
  - title: Perform differential gene expression testing
    element: '#tool-search-query'
    content: Search for DESeq2 tool.
    placement: right
    textinsert: DESeq2
  - title: Perform differential gene expression testing
    element: '#tool-search'
    content: Click on the "DESeq2" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fiuc%2Fdeseq2%2Fdeseq2%2F2.11.40.2"]
        .tool-old-link
  - title: Perform differential gene expression testing
    element: '#tool-search'
    content: |-
      Run DESeq2 with the following parameters:
        <ul>
          <li>Specify “G1E” as the first factor level (condition) and select the count files corresponding to the two replicates</li>
          <li>Specify “Mega” as the second factor level (condition) and select the count files corresponding to the two replicates</li>
          <li><b>Visualising the analysis results:</b> Yes</li>
          <li><b>Output normalized counts table:</b> Yes</li>
        </ul>
    position: left
  - title: Perform differential gene expression testing
    element: '.history-right-panel .list-items > *:first'
    content: |-
      The first output of DESeq2 is a tabular file. The columns are:<ul>
        <li>Gene identifiers</li>
        <li>Mean normalized counts, averaged over all samples from both conditions</li>
        <li>Logarithm (base 2) of the fold change (the values correspond to up- or downregulation relative to the condition listed as Factor level 1)</li>
        <li>Standard error estimate for the log2 fold change estimate</li>
        <li><a href="http://data.princeton.edu/wws509/notes/c2s3.html">Wald</a> statistic</li>
        <li><i>p</i>-value for the statistical significance of this change</li>
        <li><i>p</i>-value adjusted for multiple testing with the Benjamini-Hochberg procedure which controls false discovery rate (<a href="http://www.biostathandbook.com/multiplecomparisons.html">FDR</a>)</li>
      </ul>
    position: left
  - title: Perform differential gene expression testing
    element: '#tool-search-query'
    content: Search for Filter tool.
    placement: right
    textinsert: Filter
  - title: Perform differential gene expression testing
    element: '#tool-search'
    content: Click on the "Filter" tool to open it
    placement: right
    postclick:
      - 'a[href$="/tool_runner?tool_id=Filter1"]'
  - title: Perform differential gene expression testing
    element: '#tool-search'
    content: >-
      Run Filter to extract genes with a significant change in gene expression
      (adjusted p-value less than 0.05) between treated and untreated samples
    position: left
  - title: Question
    content: |-
      <ul>
        <li>How many transcripts have a significant change in expression between these conditions?</li>
      </ul>
    backdrop: false
  - title: Filter
    content: Determine how many transcripts are up or down regulated in the G1E state.
    backdrop: true
  - title: Question
    content: |-
      <ul>
        <li>Are there more upregulated or downregulated genes in the treated samples?</li>
      </ul>
    backdrop: false
  - title: Perform differential gene expression testing
    element: '.history-right-panel .list-items > *:first'
    content: Rename your datasets for the downstream analyses
    position: left
  - title: Perform differential gene expression testing
    content: >-
      In addition to the list of genes, DESeq2 outputs a graphical summary of
      the results, useful to evaluate the quality of the experiment:

      <ul>
        <li>Histogram of <i>p</i>-values for all tests</li>
        <li><a href="https://en.wikipedia.org/wiki/MA_plot>MA plot</a>: global view of the relationship between the expression change of conditions (log ratios, M), the average expression strength of the genes (average mean, A), and the ability of the algorithm to detect differential gene expression. The genes that passed the significance threshold (adjusted p-value < 0.1) are colored in red.</li>
        <li>Principal Component Analysis (<a href="https://en.wikipedia.org/wiki/Principal_component_analysis">PCA</a>) and the first two axes</li>
        Each replicate is plotted as an individual data point. This type of plot is useful for visualizing the overall effect of experimental covariates and batch effects.
        <li>Heatmap of sample-to-sample distance matrix: overview over similarities and dissimilarities between samples</li>
      </ul>
    backdrop: false
  - title: Perform differential gene expression testing
    content: |-
      <ul>
        <li>Dispersion estimates: gene-wise estimates (black), the fitted values (red), and the final maximum a posteriori estimates used in testing (blue)</li>
        This dispersion plot is typical, with the final estimates shrunk from the gene-wise estimates towards the fitted estimates. Some gene-wise estimates are flagged as outliers and not shrunk towards the fitted value. The amount of shrinkage can be more or less than seen here, depending on the sample size, the number of coefficients, the row mean and the variability of the gene-wise estimates.
        <br><br>For more information about DESeq2 and its outputs, you can have a look at <a href="https://www.bioconductor.org/packages/release/bioc/manuals/DESeq2/man/DESeq2.pdf>DESeq2 documentation</a>.
      </ul>
    backdrop: false
  - title: Visualization
    content: >-
      Now that we have a list of transcript expression levels and their
      differential expression levels, it is time to visually inspect our
      transcript structures and the reads they were predicted from. It is a good
      practice to visually inspect (and present) loci with transcripts of
      interest. Fortunately, there is a built-in genome browser in Galaxy,
      <b>Trackster</b>, that make this task simple (and even fun!).

      <br><br>In this last section, we will convert our aligned read data from
      BAM format to bigWig format to simplify observing where our stranded
      RNA-seq data aligned to. We’ll then initiate a session on Trackster, load
      it with our data, and visually inspect our interesting loci.
    backdrop: true
  - title: Converting aligned read files to bigWig format
    element: '#tool-search-query'
    content: Search for bamCoverage tool.
    placement: right
    textinsert: bamCoverage
  - title: Converting aligned read files to bigWig format
    element: '#tool-search'
    content: Click on the "bamCoverage" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fbgruening%2Fdeeptools_bam_coverage%2Fdeeptools_bam_coverage%2F3.0.2.0"]
        .tool-old-link
  - title: Converting aligned read files to bigWig format
    element: '#tool-search'
    content: >-
      Run bamCoverage on all four aligned read files (HISAT output) with the
      following parameters: <ul>
        <li><b>Bin size in bases:</b> 1</li>
        <li><b>Effective genome size:</b> mm9 (2150570000)</li>
        <li>Expand the <b>Advanced options</b></li>
        <li><b>Only include reads originating from fragments from the forward or reverse strand:</b> forward</li>
      </ul>
    position: left
  - title: Perform differential gene expression testing
    element: '.history-right-panel .list-items > *:first'
    content: >-
      Rename the outputs to reflect the origin of the reads and that they
      represent the reads mapping to the PLUS strand.
    position: left
  - title: Perform differential gene expression testing
    content: Now we need to repeat bamCoverage step with some changes in settings.
    backdrop: true
  - title: Converting aligned read files to bigWig format
    element: '#tool-search'
    content: Click on the "bamCoverage" tool to open it
    placement: right
    postclick:
      - >-
        a[href$="/tool_runner?tool_id=toolshed.g2.bx.psu.edu%2Frepos%2Fbgruening%2Fdeeptools_bam_coverage%2Fdeeptools_bam_coverage%2F3.0.2.0"]
        .tool-old-link
  - title: Converting aligned read files to bigWig format
    element: '#tool-search'
    content: >-
      Run bamCoverage on all four aligned read files (HISAT output) with the
      following parameters: <ul>
        <li><b>Bin size in bases:</b> 1</li>
        <li><b>Effective genome size:</b> mm9 (2150570000)</li>
        <li>Expand the <b>Advanced options</b></li>
        <li><b>Only include reads originating from fragments from the forward or reverse strand:</b> reverse</li>
      </ul>
    position: left
  - title: Perform differential gene expression testing
    element: '.history-right-panel .list-items > *:first'
    content: >-
      Rename the outputs to reflect the origin of the reads and that they
      represent the reads mapping to the MINUS strand.
    position: left
  - title: Trackster based visualization
    element: '#visualization .dropdown a[href$="/visualization/list"]'
    content: >-
      On the center console at the top of the Galaxy interface, choose “
      Visualization” -> “New track browser”.
    placement: left
  - title: Trackster based visualization
    content: |-
      <ul>
        <li>Name your visualization someting descriptive under “Browser name:”</li>
        <li>Choose “Mouse Dec. 2011 (GRCm38/mm10) (mm10)” as the “Reference genome build (dbkey)</li>
        <li>Click “Create” to initiate your Trackster session</li>
      </ul>
    backdrop: false
  - title: Trackster based visualization
    content: |-
      Click “Add datasets to visualization”<ul>
        <li>Select the “RefSeq GTF mm10” file</li>
        <li>Select the output files from Stringtie</li>
        <li>Select the output file from GFFCompare</li>
        <li>Select the output files from bamCoverage</li>
      </ul>
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Using the grey labels on the left side of each track, drag and arrange the
      track order to your preference.
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Hover over the grey label on the left side of the “RefSeq GTF mm10” track
      and click the “Edit settings” icon. <ul>
        <li>Adjust the block color to blue (#0000ff) and antisense strand color to red (#ff0000)</li>
      </ul>
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Repeat the previous step on the output files from StringTie and
      GFFCompare.
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Hover over the grey label on the left side of the “G1E R1 plus” track and
      click the “Edit settings” icon. <ul>
        <li>Adjust the color to blue (#0000ff)</li>
      </ul>
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Repeat the previous step on the other three bigWig files representing the
      plus strand.
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Hover over the grey label on the left side of the “G1E R1 minus” track and
      click the “Edit settings” icon. <ul>
        <li>Adjust the color to red (#ff0000)</li>
      </ul>
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Repeat the previous step on the other three bigWig files representing the
      minus strand.
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Adjust the track height of the bigWig files to be consistent for each set
      of plus strand and minus strand tracks.
    backdrop: false
  - title: Trackster based visualization
    content: >-
      Direct Trackster to the coordinates: chr11:96191452-96206029, what do you
      see?
    backdrop: false
  - title: Question
    content: |-
      <ul>
        <li>What do you see?</li>
      </ul>
    backdrop: false
  - title: Conclusion
    content: >-
      In this tutorial, we have analyzed real RNA sequencing data to extract
      useful information, such as which genes are up- or down-regulated by
      depletion of the Pasilla gene and which genes are regulated by the Pasilla
      gene. To answer these questions, we analyzed RNA sequence datasets using a
      reference-based RNA-seq data analysis approach.
    backdrop: true
  - title: Key points
    content: |-
      <ul>
        <li>De novo transcriptome reconstruction is the ideal approach for identifying differentially expressed known and novel transcripts.</li>
        <li>Differential gene expression testing is improved with the use of replicate experiments and deep sequence coverage.</li>
        <li>Visualizing data on a genome browser is a great way to display interesting patterns of differential expression.</li>
      </ul>
    backdrop: true