Introduction

ScanIndel is a python program to detect indels (insertions and deletions) from NGS data by re-align and de novo assemble soft clipped reads.

Prerequisites

Softwares and Python packages:

• BedTools/2.17.0 (https://github.com/arq5x/bedtools2/releases)
• SAMtools/1.0 or less than 1.0 (http://samtools.sourceforge.net/)
• BWA/0.7.10 (http://bio-bwa.sourceforge.net/)
• BLAT (gfServer and gfClient)/34+ (http://genome.ucsc.edu/FAQ/FAQblat.html)
• freebayes/0.9.18 (https://github.com/ekg/freebayes)
• Inchworm assembler (http://inchworm.sourceforge.net)
• Python/2.7 (https://www.python.org/)
• Pysam/0.7.7 (https://code.google.com/p/pysam/)
• PyVCF/0.6.7 (https://github.com/jamescasbon/PyVCF)
• Biopython/1.64 (http://biopython.org/wiki/Main_Page)
• SciPy/0.14.0 and NumPy/1.8.1 (http://www.scipy.org/)

All softwares above are assumed to be installed in your searching path. Ask your admistrator for assistance if necessary.

Getting Soure Code

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

Running ScanIndel

command-line usage

python ScanIndel.py -i sample.txt -p config.txt [options]

Options:

 -F                 :setting min-alternate-fraction for FreeBayes (default 0.2)
 -o                 :setting output directory (default ./)
 -C                 :setting min-alternate-count for FreeBayes (default 2)
 -s                 :softclipping percentage triggering BLAT re-alignment (default 0.2)
 -t                 :setting -t for FreeBayes to provide a BED-format file limiting the analysis to these regions
 --min_percent_hq           :min percentage of high quality base in soft clipping reads (default 0.8)
 --lowqual_cutoff           :low quality cutoff value (default 20)
 --mapq_cutoff              :low mapping quality cutoff (default 1)
 --blat_ident_pct_cutoff        :Blat sequence identity cutoff (default 0.8)
 --gfServer_port            :gfServer service port number, changing this value to allow multiple ScanIndel running at a single machine (default 50000)
 --hetero_factor            :The factor about the indel heterogenirity and heterozygosity (default 0.1)
 --bam                  :the input file is BAM format
 --rmdup                :exccute duplicate removal step before realignment
 -h --help              :produce this menu
 -v --version               :show version of this tool

Input:

sample.txt                  :this file contains the listed samples to be analyzed (one per line), the input can be raw read FastQ file or aligned BAM file and use --bam when running (default name is sample.txt)
config.txt                  :this file contains the path of reference file for each BWA, BLAT and Freebayes (default name is config.txt)

Output:

The output files include the VCF file for detected variant and BAM files for BWA-MEM and BLAT mapping.

*.reads.bam             :BAM file for read after blat alignment.
*.contigs.bam               :BAM file for de novo assembled contigs after BWA and BLAT mapping.
*.mapping.indel.vcf         :VCF file includes putative INDELs from softclipping read re-alignment.
*.assembly.indel.vcf            :VCF file includes putative INDELs from de novo assembly.
*.merged.indel.vcf          :VCF file that include all putative INDELs by merging the results from *mapping.indel.vcf and *.assembly.indel.vcf

Example:

To run a test example, please go to folder 'example' and follow the README file to run test data.

Citing ScanIndel:

Yang, Rendong, et al. "ScanIndel: a hybrid framework for indel detection via gapped alignment, split reads and de novo assembly." Genome medicine 7.1 (2015): 1-12.