geng-lee / transindel

Indel caller for DNA-seq or RNA-seq

Introduction

transIndel is used to detect indels (insertions and deletions) from DNA-seq or RNA-seq data by parsing chimiric alignments from BWA-MEM.

Prerequisites

Samtools/1.0 or newer (http://www.htslib.org/) Python 3.6 or newer (https://www.python.org/) Python packages:

• Pysam/0.13.0 or newer (https://pypi.org/project/pysam)
• HTSeq/0.6.1 or newer (https://pypi.python.org/pypi/HTSeq)

Getting Soure Code

git clone git://github.com/cauyrd/transIndel.git
cd transIndel

Running transIndel

STEP 1: Build new BAM file with redefined CIGAR string

• analyzing DNA-seq data (whole genome seq/exome-seq/targeted capture)

   python transIndel_build_DNA.py -i input_bam_file -o output_bam_file [options]
  

• analyzing RNA-seq data

   python transIndel_build_RNA.py -i input_bam_file -r reference_genome_fasta -g gtf_file -o output_bam_file [options]
  

Options:

-h, --help            show this help message and exit
-i INPUT, --input INPUT
                    Input BAM file
-o OUTPUT, --output OUTPUT
                    Output BAM file
-r REF, --ref REF     reference genome used for analyzing RNA-seq data
-g GTF, --gtf GTF     gene annotatino file used for analyzing RNA-seq data
-s SPLICE_BIN, --splice_bin SPLICE_BIN
                    splice site half bin size (default: 20)
-m MAPQ, --mapq MAPQ  minimal MAPQ of read from BAM file for supporting
                    Indel (default: 15)
-l LENGTH, --length LENGTH
                     Maximum deletion length to be detected (default:1000000)
-v, --version         show program's version number and exit

Input:

input_bam_file   			:input BAM file is produced by BWA-MEM and is sorted and indexed.
reference_genome_fasta (for RNA-seq)    :reference genome in FastA format
gtf_file (for RNA-seq)    		:gene annotation file in GTF format

Output:

your_output_bam_file			:BAM file for CIGAR string redefinement.

transIndel generates the following optional fields in output BAMs

Tag| Meaning
--------------------------------------------------------------------------------------
OA | original representative alignment; format: (pos,CIGAR)
JM | splicing junction reads; infered from GTF or splicing motif (used in RNA-seq BAM)

STEP 2: Call indel

• Option 1: using transIndel_call.py script

   python transIndel_call.py -i input_bam_from_transIndel_build -o output_vcf_filename_prefix [options]	
  

Options:

-h, --help            show this help message and exit
-i INPUT, --input INPUT
                        Input BAM file
-o OUTPUT, --output OUTPUT
                    output VCF file prefix
-c AO, --ao AO        minimal observation count for Indel (default: 4)
-d DEPTH, --depth DEPTH
                 minimal depth to call Indel (default: 10)
-f VAF, --vaf VAF     minimal variant allele frequency (default: 0.1)
-l LENGTH, --length LENGTH
                 minimal Indel length (>=1) to report (default: 10)
-m MAPQ, --mapq MAPQ  minimal MAPQ of read from BAM file to call Indel
                 (default: 15)
-t REGION             Limit analysis to targets listed in the BED-format
                 FILE or a samtools region string
-v, --version         show program's version number and exit

Input:

input_bam_file   			:input BAM file is produced by transIndel_build.py

Output:

output_vcf_file   			:Reported Indels with VCF format

• Option 2: using existing variant caller (e.g. VarDict, freebayes, GATK)

   following the specific variant caller's manual