README for MixClone

INTRODUCTION

MixClone is a comprehensive software package to study the subclonal structures of tumor genomes, including subclonal cellular prevalences estimation, allelic configuration estimation, absolute copy number estimation and a few visualization tools. It takes next-generation sequencing data of paired normal-tumor samples as input and integrates sequence information from both somatic copy number alterations and allele frequencies within a unified probabilistic framework. If you have any questions, please email [email protected]

INSTALL

Prerequisites

Mandatory

• Python (2.7). Python 2.7.3 is recommended.

• Numpy(>=1.6.1). You can download the source of Numpy from here.

• Scipy(>=0.10). You can download the source of Scipy from here.

• Pysam(>=0.7). Pysam_0.7X preferred, Pysam_0.8X tested and seems to be much slower (~4 times slower). To install Pysam, you also need to install Cython first.

Optional

• matplotlib(>=1.2.0) is required for a few visualization tools.

Although not mandatory, Linux system is recommended. Also, samtools is not required by MixClone, but can be useful for creating bam, bam index and fasta index files which are required by the pysam module of MixClone.

Install from source

Download the compressed source file MixClone-*.tar.gz and do as follows:

$ tar -xzvf MixClone-*.tar.gz
$ cd MixClone-*
$ python setup.py install

If you prefer to install MixClone other than the default directory, you can also use this command:

$ python setup.py install --prefix /home/yili/

There is also a bin/ folders under MixClone-*. The bin/ folder contains useful utilities, such as the R code to run BICseq and the python script to convert BICseq results to BED file. You can copy these two folders somewhere easily accessible.

USAGE

Overview

MixClone is composed of three modules:

• preprocess. Preprocess the reads aliments of paired normal-tumor samples in BAM format, the tumor genome segmentation file in BED format, and produce the *.MixClone.input.pkl file as output, which will be used for running the model.

• run_model. Take the *.MixClone.input.pkl as input, estimate the subclonal cellular prevalence, the absolute copy number and the allelic configuration of each segment, and produce the *.MixClone.output.pkl file as output, which will be used for postprocessing. If the user runs the model without specifying the number of subclonal populations, MixClone will run the model five times with subclonal number ranges from 1 to 5, and recommend the most likely model.

• postprocess. Take the *.MixClone.output.pkl file as input, and extract various output files, including the segments file with estimated parameters, the allele counts file, the subclonal analysis summary and a few plots.

The general workflow of MixClone is this

Tumor genome segmentation

MixClone requires a segmentation file of the tumor genome in BED format before running the package. We used BICseq in the original paper. To run a BICseq analysis, you can copy the commands in bin/BICseq.R (Li, Y., Xie, X. 2014, Bioinformatics) and paste them in a R interative shell. Or you can also run the R script from the command line:

$ R CMD BATCH bin/BICseq.R

Note that,normal.bam and tumor.bam must be in the same directory where you run the command. The R script will output a segments file segments.BICseq. Then you can use the other script bin/BICseq2bed.py (Li, Y., Xie, X. 2014, Bioinformatics) to convert the segments file into BED format:

$ BICseq2bed.py segments.BICseq segments.bed --seg_length 1000000

--seg_length Only convert segments with length longer than the threshold.

Preprocess

This part of README is based on JoinSNVMix. To preprocess the paired cancer sequencing data, execute:

$ MixClone.py preprocess REFERENCE_GENOME.fasta SEGMENTS.bed NORMAL.bam TUMOUR.bam INPUT_BASENAME --min_depth 20 --min_base_qual 10 --min_map_qual 10 --process_num 10

REFERENCE_GENOME.fasta The path to the fasta file that the paired BAM files aligned to. Currently, only the UCSC and ENSEMBL chromosome format are supported. Note that the index file should be generated for the reference genome. This can be done by running samtools as follows:

$ samtools faidx REFERENCE_GENOME.fasta

SEGMENTS.bed The BED file for the tumor genome segmentation.

NORMAL.bam The BAM file for the normal sample. The BAM index file should be generated for this file and named NORMAL.bam.bai. This can be done by running

$ samtools index NORMAL.bam

TUMOUR.bam The bam file for the tumour sample. As for the normal this file needs to be indexed.

INPUT_BASENAME The base name of the preprocessed input file to be created.

--min_depth Minimum depth in both normal and tumor sample required to use a site in the analysis.

--min_base_qual Minimum base quality required for each base.

--min_map_qual Minimum mapping quality required for each base.

--process_num Number of processes to launch for preprocessing.

Run model

After the preprocessed input file is created, we can run the probabilistic model of MixClone by execute:

$ MixClone.py run_model INPUT_BASENAME OUTPUT_BASENAME --max_copynumber 6 --subclone_num 2 --max_iters 30 --stop_value 1e-6

INPUT_BASENAME The base name of the preprocessed input file created in the preprocess step.

OUTPUT_BASENAME The base name of the output file with model parameters estimated to be created.

--max_copynumber The maximum copy number of each segment allows to take.

--subclone_num The number of subclones within the tumor sample. If not provided, go through [1, 5] and select the most likely model.

--max_iters Maximum number of iterations for training.

--stop_value Stop value of the EM algorithm for training. If the change of log-likelihood is lower than this value, stop training.

Postprocess

After the output file with model parameters estimated, we can extract various result files from the output file by execute:

$ MixClone.py postprocess OUTPUT_BASENAME

BASENAME The base name of the output file created in the run_model step.

Output files

*.MixClone.segments The segments file. It contains the genomic and subclonal information of each segment. The definition of each column in a *.MixClone.segments file is listed here:

Column	Definition
seg_name	Name of the segment
chrom	Chromosome of the segment
start	Start position of the segment
end	End position of the segment
normal_reads_num	Count of reads mapped to the segment in the normal sample
tumor_reads_num	Count of reads mapped to the segment in the tumor sample
LOH_frac	Fraction of LOH sites in the segment
LOH_status	FALSE -> no LOH; TRUE -> significant LOH; UNCERTAIN -> medium level LOH
log2_ratio	Log2 ratio between tumor_reads_num and normal_reads_num
copy_number	Estimated absolute copy number of the segment
allele_type	Estimated allelic configuration of the segment
subclone_prev	Estimated subclonal cellular prevalence of the segment
subclone_cluster	Estimated subclonal cluster label of the segment

*.MixClone.counts The allele counts file. It contains the allelic counts information of sites, which are heterozygous SNP sites in the normal genome. The definition of each column in a *.MixClone.counts file is listed here:

Column	Definition
seg_name	Name of the segment
normal_A	Count of bases match A allele in the normal sample
normal_B	Count of bases match B allele in the normal sample
tumor_A	Count of bases match A allele in the tumor sample
tumor_B	Count of bases match B allele in the tumor sample
chrom	Chromosome of the site
pos	Genomic position of the site

*.MixClone.summary The summary file about the subclonal analysis, including log-likelihood, subclonal labels and the corresponding cellular prevalences.

*.MixClone.heatmap The folder of BAF heat maps plotted for each segment (Li, Y., Xie, X. 2014, Bioinformatics).

*.MixClone.segplot.png The plot of estimated subclonal cellular prevalences and absolute copy numbers of all the segments of non-diploid allelic configuration.

*.MixClone.model_selection The summary about model selection if --subclone_num is not specified. It includes the log-likelihood and related change of each model with different number of subclones. Although MixClone selects the most likely model based on a heuristic described in the original paper, users can select the best model use their own judgement based on the log-likelihood related information.

Reference

Li, Y., Xie, X. (2015). MixClone: a mixture model for inferring tumor subclonal populations. BMC genomics.