Home-made pipeline for SNV/CNV variant calling and annotation from fastq files

Overview

A comprehensive analysis pipeline for processing tumor sequencing data from raw FASTQ files to annotated variants and copy number variations (CNVs). This pipeline integrates multiple tools for quality control, variant calling, annotation, and prioritization.

Pipeline Workflow

Step 1: FASTQ to VCF Conversion

• Input: Paired-end FASTQ files (fq1, fq2)

• Process: Aligns reads to reference genome, produces BAM file, and calls variants using HaplotypeCaller

• Output: Sorted BAM file and VCF file with raw variants

Step 2: SNV Filtering

• Input: Raw VCF file from Step 1

• Process: Filters low-quality variants using quality metrics and custom criteria

• Output: Filtered variant list in text format

Step 3: Variant Annotation with CLEVER

• Input: Filtered variant file

• Tool: CLEVER (https://zenodo.org/records/15534528)

• Process: Annotates variants with functional and clinical information

• Output: Comprehensive variant annotation in TSV format

Step 4: Coverage Calculation

• Input: Aligned BAM file

• Process: Calculates read depth across target regions

• Output: Coverage files for autosomes and chromosome X

Step 5: Coverage File Preprocessing

• Gender Adjustment: Modifies chromosome X coverage for male samples (due to hemizygosity)

• Target Region Filtering: Focuses analysis on predefined target genes using BEDTools

• Batch Processing: Merges coverage files when analyzing multiple samples

Step 6: CNV Calling

• CANOES: Identifies CNVs using read-depth approach (https://github.com/ShenLab/CANOES)

• HMZDelFinder: Detects homozygous deletions in exome data

• Output: CNV calls from both methods

Step 7: CNV Annotation with PICNIC

• Input: CNV calls from Step 6

• Process: Annotates CNVs and prioritizes potentially pathogenic variants

• Reference: Detailed in PMID: 10.1136/jmedgenet-2019-106377

Step 8: Phenotype-Based Prioritization (Optional)

• Tool: PhenoPro (http://omics.fudan.edu.cn/phenopro/)

• Process: Ranks variants based on provided HPO terms to assist in clinical interpretation

Input Requirements

• FASTQ files: Paired-end sequencing data

• Sample information: Sample name and gender (Male/Female)

• Reference genome: hg19 (GRCh37)

• Target regions: Predefined gene panels for autosomes and chromosome X