README

SV simulation pipeline

Simulates SV of different purity levels and types. This was used to generate test data for validating the read counting used in SVclone.

How do I get set up?

Install these dependencies:

• Java Development Kit
• Samtools -- make sure samtools is in your $PATH
• Rubra -- install Rubra

These instructions will clone this repository, and install SimSeq and refpluspipeline.

git clone git@github.com:mcmero/sv_simu_pipe.git
cd sv_simu_pipe

git clone git@github.com:mcmero/refpluspipeline.git
cd refpluspipeline
./compile_javasim.sh

cd ..
git clone https://github.com/jstjohn/SimSeq.git

Create a reference directory under your sv_simu_pipe directory, and add the following files:

• hg19 chromosome reference (pick one chromosome, 12 is used in the example) -- generate an index with Bowtie2
• hg19 repeat track -- from UCSC (name it hg19_repeats.txt)

Configure the sv_simu_config.py configuration file, making sure the reference genome, repeats file, SimSeq dir and the directories for the java simulator are set correctly.

Running the pipeline

Run as:

./batch_simu.sh

This will generate 20 simulations: 20%, 40%, 60%, 80% and 100% purity mixtures for deletions, duplications, inversions and traslocations. This can be used as input for SV callers. VAFs can then be characterised using SVclone.