TNBC-Variant-Pipeline: End-to-end GATK-based workflow for identifying clinically actionable variants in Triple-Negative Breast Cancer (TNBC).

This repository implements a reproducible oncology variant calling pipeline that follows GATK Best Practices to analyze raw sequencing data from a Triple-Negative Breast Cancer (TNBC) tumor sample (SRR8707685) in the Fudan University TNBC cohort. The objective of this project is to demonstrate how raw next-generation sequencing (NGS) data can be processed through alignment, quality recalibration, somatic variant calling, and functional annotation to identify high-impact mutations relevant to targeted therapy. This project is designed as a computational demonstration of a WES somatic variant calling workflow and does not constitute a clinical diagnostic analysis.

Biological Question: Can we identify clinically actionable mutations in a TNBC tumor sample that may inform targeted therapy strategies? Triple-Negative Breast Cancer is frequently associated with deficiencies in the Homologous Recombination (HR) DNA repair pathway, particularly involving BRCA1/2 alterations. Identifying high-impact variants in these genes can have therapeutic implications.

Selected sample SRR8707685 from the Fudan University TNBC cohort (SRP157974). This sample represents Triple-Negative Breast Cancer, a subtype frequently associated with deficiencies in the Homologous Recombination (HR) DNA repair pathway, specifically involving BRCA1/2 mutations.

Tech Stack & Standards

• Reference Genome: GRCh38 (hg38)- Chromosome 13 & 17
• Alignment: BWA-MEM
• Data Cleanup: GATK4 (MarkDuplicates, VariantFiltration, SelectVariants)
• Variant Calling: GATK HaplotypeCaller
• Annotation: GATK Funcotator (ClinVar, Gencode)

📁 Repository Structure

scripts/ # Pipeline execution scripts

env.yml # Reproducible Conda environment

results_preview.txt # Subset of annotated VCF

final_report # Summary of findings and clinical interpretations.

📊 Key Findings

The analysis identified several high-confidence, high-impact variants that characterize the aggressive mutational landscape of this Triple-Negative Breast Cancer (TNBC) sample.

This pipeline identifies multiple high-impact coding variants, including nonsense and frameshift mutations that are likely to disrupt protein function.

Example High-Impact Variants:-

Gene	Mutation Type	Predicted Effect
PABPC3	Frameshift	Loss of normal protein function
ATP12A	Nonsense	Premature stop codon
MRPL57	Frameshift	Truncated mitochondrial protein

High-impact variants were prioritized based on:

• Functional consequence (frameshift, nonsense)
• Read depth and variant allele fraction
• Annotation from ClinVar / cancer databases (via Funcotator)

Key Biological Insights:

1. Loss of Function in Tumor Suppressors: The identification of Nonsense mutations in genes like ATP12A and ASPSCR1 suggests a complete loss of protein function. In a clinical setting, these "truncation" events often disable the cell's natural ability to regulate growth.
2. Genomic Instability: The presence of multiple Frame-shift mutations (e.g., PABPC3, MAFG) indicates significant genomic instability. These mutations shift the reading frame of the DNA, leading to entirely dysfunctional proteins and contributing to the "triple-negative" phenotype, which lacks traditional hormone receptors.
3. Actionability: While TNBC is traditionally difficult to treat due to the lack of ER/PR/HER2 targets, identifying specific deleterious mutations in DNA repair or metabolic pathways (like those found on Chr 13 and 17) provides a roadmap for targeted therapy or PARP inhibitor eligibility in future clinical contexts.

This pipeline successfully transitioned from raw sequencing data to a refined list of clinically relevant variants, demonstrating that computational workflows can effectively nominate therapeutic targets in complex cancers.