Snakemake pipeline for preprocessing of the DRAGEN bam and remapping them to GRCh37 genome for the follow up mosaic variant calling algorithms

The pipeline is originally wirtten by Danny Antaki, and re-implemented by Xin Xu, with help form Martin Breuss and Xiaoxu Yang

---

Before start, make sure you have:

java for Linux.

Picard Tools from BROAD institute.

BWA for mapping.

SAMtools for post-alignment processing.

GATK 3.8 for this pipeline.

Sambamba to help with the fast processing.

---

Input:

Below are headers of the input file list

UNIQ_ID

Unique IDs for any input Dragen bam file.

SAMPLE_ID

The ID of the sample, might be shared between several bams.

BAM_PATH

The path to the input Dragen bam.

---

Config files:

Below are files you need to prepare for the annotation scripts, saved in the file snake_conf.yaml

input_files

Path to the input file list.

out_dir

Path to the output directory.

scratch_dir

Path to the scratch files. Note that the number of temporary files will be euqal to two- or three-times the number of total listed variants.

temp_dir

Path to the temporary files. The temporary files will be deleted after the process is successfully finished.

java

Path to your java jre.

picard

Path to your picard.jar.

bwa

Path to your BWA.

samtools

Path to your SAMtools.

gatk

Path to your GenomeAnalysisTK.jar

sambamba

Path to your executable sambamba

hg19_fasta

Your reference genome.

mills_indel

vmills indel vcf file (corresponding to your reference genome file).

gp1000_indel

List for common indels from the 1000 Genome Project in vcf format (corresponding to your reference genome file).

db_gap

Snp list of your dbsnp file in vcf format (corresponding to your reference genome file).

---

Output:

The final output bam is in the recaled_bams folder.