ngs-main-wes

This is a Dockerized "somatic paired-WES(Normal-Tumor) hg19 pipeline" which is refer to GATK best practice.

• Notes : This scripts based on related path
• Flow chart

---

Prepare env:

1. git clone to your main project directory (ex. OSCC)
2. put "ngs-main-wes.py" and "NGStools.py" in your main project directory (ex. OSCC)
3. prepare your own raw data in sub project directory (ex. 66xWES)

---

Usage:

python3 ngs-main-wes.py

---

Customizable：

• tag: wes_v3(docker image: https://hub.docker.com/r/adgh456/ngs-main/)
• genome coordinate: ucsc.hg19.fasta
• dbSNP: dbsnp_138.hg19.vcf
• COSMIC: CosmicAllMutsHeaderSorted.vcf
• seq_bed: agilent_region_OSCC_hg19_rmheader.bed
• gnomAD: gnomad.exomes.r2.0.2.rename.vcf.gz
• multi-thread: p = Pool(15)

---

Function：

• rmSAM: remove sam & sai file to release disk space
• getAllVCF: copy all subproject vcf to mainDir/annotation

---

• rawdataRename: rename raw data directory (ex. OC_631N to 631N)
• enterData: input patient ID (for parallel processing)

---

• CheckDir: mkdir ref_data (for reference and intermediate file), data (for storage)
• NGSMainWES: copy reference (ucsc.hg19.fasta, dbsnp_138.hg19.vcf, 'CosmicAllMutsHeaderSorted.vcf) to ref_data
• ReferenceIndex: build index for reference with samtools

---

• MergeFastq: merge multiple fastq into one (N.fq1, N.fq2, T.fq1, T.fq2)
• AfterQC: trimming raw fastq data with Afterqc (auto trimming), and produce a QC report
• FastqtoSam: align fastq reads to reference genome with BWA, and produce sam file
• SamtoSortbam: transform sam file (sequence alignment map) to bam file (binary alignment map) with picard, and build index with samtools
• MarkDuplicates: MarkDuplicates and AddOrReplaceReadGroups with picard, and build index with samtools
• BaseRecalibrator: BaseRecalibrator and ApplyBQSR with GATK4, and build index with samtools
• NormalforPONsOfMutect2: call variant using Mutect2 (tumor-only mode) without dbSNP and COSMIC with GATK4, and record patient to build Panel of Normal in normals_for_pon_vcf.args
• Mutect2: call variant using Mutect2 (Paired: Normal and Tumor) without dbSNP and COSMIC with GATK4
• Mutect2_v3: call variant using mutect2 (Paired: Normal and Tumor) with dbSNP and COSMIC with GATK4
• CheckVcf: record result (good_report.txt or bad_report.txt)
• CreatePONforCNV: create PON of CNV
• CNV: call copy number variant with GATK4 1.0.0.0.alpha
• MSIsensor: microsatellite instability
• Phial: clinical FDA drug relevence annotation (based on autoOncotator, please see https://github.com/shanghungshih/autoOncotator)
• ParaSNP: scoring variants in vcf, based on annovar annotation（Notes: process in data, not ref_data）
• CreatePONforMutect2: generate total vcf from normals of subproject patient
• Mutect2_PONs: call variant using Mutect2 with PONs and gnomAD