Introduction

PatternCNV is a versatile tool to facilitate CNV detection and interpretation from either Germline or Tumor/Normal pairs of Exome-seq samples. “Pattern” stands for common trends summarized from Exome-seq data. PatternCNV consists two basic components: 1) BAM2WIG conversion implemented with Perl and Bash scripts: converting large BAM files to the lightweight WIG format, allowing for less computationally intense read-depth comparisons and simpler coverage visualization; 2) a set of CNV functions implemented in R which use the WIG files to facilitate pattern-learning, CNV detection, and various post-processing and visualization functions. To avoid garbage-in garbage-out situations, multiple QC checks are included in the tool, from BAM2WIG script to several R functions. CNV results are investigated through interactive visualization or summary tables, across samples and/or genome.

Usage

module load bioinformatics/patterncnv/1.0

more readme.txt
# Create the necessary files to run PatternCNV
# Jared Evans
# 5/7/13
## STEP 1. Create an Exon Key:
# NOTE: start and stop of output key file are 1-base so it matches the 1-base WIG files
exon_key.sh <exon bed file> <capture kit bed file> <output file> <bin size (10)> <tool config.txt>
# EXAMPLE USAGE:
exon_key.sh reference/refflat_exons_hg19.bed SureSelect_All_Exon_V4+UTRs_with_annotation_hg19_.bed PatternCNV.Exon.Key.txt 10 config.txt

## STEP 2. Generate WIG files for each BAM:
bam2wig.sh <input bam file> <output dir> <bin size (10)> <min mapping quality (20)> <tool config.txt> <exon key>
# EXAMPLE USAGE:
bam2wig.sh LS1946_blood.sorted.bam output_dir/ 10 20 config.txt PatternCNV.Exon.Key.txt

Installation

Dir: /sw/bioinformatics/patterncnv/1.0

Griffith HPC Cluster Documentation

patterncnv

Introduction

Usage

Installation

Reference