Installation

Table of Contents

• Prerequisites
• pip install
• Manual install
• Configure SV²
• Check installation

---

Prerequisites

• python 2.7
  • cython
  • numpy
  • pandas
  • pybedtools
  • pysam 0.9+
  • scikit-learn 0.19+
• bedtools 2.25.0 or later

---

Users can install with pip or manually install SV².

Install with pip

Recommended

$ pip install sv2

---

Manual Install

Source Files 💾

$ wget https://github.com/dantaki/SV2/releases/download/sv2v1.4.0/sv2-1.4.1.tar.gz
$ tar -zxvf sv2-1.4.1.tar.gz
$ cd sv2-1.4.1/

$ python setup.py install [--prefix <PYTHONPATH>]

• define --prefix <PYTHONPATH> for local installation
• ignore numpy compilation warnings

---

Configure

Download required resource files

$ sv2 -download

SV² requires one FASTA file to run.

• hg19 FASTA

• hg38 FASTA

• mm10 FASTA

Configure command

required before genotyping

$ sv2 -hg19 <hg19.fa> -hg38 <hg38.fa> -mm10 <mm10.fa>

Fasta files must be indexed with samtools faidx FASTA

---

Check Installation

$ sv2 -help
                       ____
  _____________   ___ |___ \
 /   _____/\   \ /   // ___/
 \_____  \  \   Y   //_____)
 /        \  \     /
/_________/   \___/

Support Vector Structural Variation Genotyper
Version 1.4.3    Author: Danny Antaki <dantaki at ucsd dot edu>    github.com/dantaki/SV2

  sv2 -i <bam ...> -v <vcf ...> -b <bed ...> -snv <snv vcf ...> -p <ped ...> [OPTIONS]

input arguments: github.com/dantaki/SV2/wiki/Options#input-arguments

  Note: input arguments can take multiple files, separated by space <-i BAM1 BAM2 ...>

  -i, -bam    ...     bam file(s)
  -v, -vcf    ...     vcf files(s) of SVs
  -b, -bed    ...     bed files(s) of SVs
  -snv        ...     snv vcf files(s), must be bgzipped and tabixed
  -p, -ped    ...     ped files(s)

genotype arguments: github.com/dantaki/SV2/wiki/Options#genotype-arguments

  -g, -genome         reference genome build <hg19, hg38, mm10> [default: hg19]
  -pcrfree            GC content normalization for pcr free sequences
  -M                  bwa mem -M compatibility, split-reads flagged as secondary instead of supplementary
  -merge              merge overlapping SV breakpoints after genotyping
  -min-ovr            minimum reciprocal overlap for merging [default: 0.8]
  -no-anno            genotype without annotating variants   

  -pre                preprocessing output directory, skips preprocessing
  -feats              feature extraction output directory, skips feature extraction

classifier arguments: github.com/dantaki/SV2/wiki/Options#classifier-arguments

  -load-clf           add custom classifiers (-load-clf <clf.json>)
  -clf                define classifier for genotyping [default:default]

config arguments: github.com/dantaki/SV2/wiki/Options#config-arguments

  -download           download required data files
  -hg19               hg19 fasta file
  -hg38               hg38 fasta file
  -mm10               mm10 fasta file
  
optional arguments:
 
  -L, -log            log file for standard error messages [default: STDOUT]
  -T, -tmp-dir        directory for temporary files [default: working directory]
  -s, -seed           random seed for preprocessing shuffling [default: 42]
  -o, -out            output prefix [default: sv2_genotypes]
  -O, -odir           output path, location for sv2 output directories [default: working directory]

  -h, -help           show this message and exit

If you get this error: Error detail: Resource temporarily unavailable there is not enough memory to run SV² in your shell.

---