genomic-project

Variant Calling Workflow – README

Overview

This repository contains a fully reproducible Snakemake workflow that processes raw whole‑exome sequencing (WES) reads (paired‑end FASTQ) to high‑confidence genetic variants (VCF/BCF). The pipeline integrates:

Stage	Tool	Purpose
QC	FastQC	Assess raw read quality
Alignment	Bowtie2	Map reads to the hg38 reference genome
Post‑process	SAMtools	Convert, sort & index alignments
Variant Calling	BCFtools (+ SAMtools mpileup)	Detect SNPs & indels, produce VCF/BCF

All software dependencies are isolated in a dedicated Conda environment (genomics_env) to guarantee identical results across machines.

---

Quick Start

# 1 Clone repository and enter project dir
$ git clone <repo‑url>
$ cd variant‑pipeline

# 2 Create Conda env (+ Mamba recommended for speed)
$ conda env create -n genomics_env -f env.yml   # or: mamba env create ...
$ conda activate genomics_env

# 3 Download / link raw FASTQs & reference index
$ mkdir data
$ cp /path/to/subset_SRR099957_*fastq.gz data/
$ cp -r /path/to/hg38_bowtie2_index reference/
$ cp /path/to/hg38.fa reference/

# 4 Dry‑run (shows DAG & commands)
$ snakemake -npr --use-conda

# 5 Execute workflow (single core, 4 GB RAM)
$ snakemake --cores 1 --resources mem_mb=4000 --use-conda

Tip: Increase --cores to parallelise multi‑sample runs or use a Snakemake cluster profile for HPC/cloud execution.

---

Input Files

Path	Description
data/subset_SRR099957_1.fastq.gz	Forward reads
data/subset_SRR099957_2.fastq.gz	Reverse reads
reference/hg38.*.bt2	Bowtie2 index files
reference/hg38.fa	Reference genome FASTA

All other paths are generated automatically by Snakemake as defined in config/config.yaml.

---

Output Artifacts

After successful execution, you will find:

results/
├── fastqc/
│   ├── subset_SRR099957_1_fastqc.html
│   └── subset_SRR099957_2_fastqc.html
├── alignments/
│   ├── subset_SRR099957.sorted.bam
│   └── subset_SRR099957.sorted.bam.bai
└── variants/
    ├── subset_SRR099957.bcf
    └── subset_SRR099957.vcf

FastQC reports can be opened in any browser; BAM/VCF files can be inspected with IGV or command‑line tools like samtools and bcftools.

---

Configuration

Edit config/config.yaml to adjust:

• samples: FASTQ filenames or pattern for multi‑sample runs.
• reference: path to Bowtie2 index & FASTA.
• threads / memory: per‑rule resource usage.

---

Extending the Pipeline

The modular Snakefile makes it easy to:

1. Add read trimming – insert a rule using fastp or Trimmomatic before alignment.
2. Switch aligner – replace align_reads_bowtie2 with BWA‑MEM2 or HISAT2.
3. Add variant filtering / annotation – append rules invoking bcftools filter, ANNOVAR, or VEP.
4. Run joint calling – create a rule that merges or jointly calls across multiple samples.

---

Troubleshooting

Issue	Fix
Conda solve is slow or conflicts	Use Mamba (conda install mamba -n base -c conda-forge).
bowtie2 cannot find index	Ensure paths in config.yaml point to all six .bt2 files.
Snakemake stops on missing files	Run with -npr to view expected filenames, then check config.yaml.

---

Citation

If you use this workflow, please cite:

• Köster & Rahmann, Snakemake (2012)
• Grüning et al., Bioconda (2018)
• Langmead & Salzberg, Bowtie 2 (2012)
• Li et al., SAMtools (2009)
• Danecek et al., BCFtools (2021)
• Babraham Bioinformatics, FastQC (2019)

---

License

MIT License – see LICENSE file.

---

Author & Acknowledgements

Developed and maintained by Pranav Dhruv Tandon (UBC). Based on community best practices in genomics data analysis.

Special thanks to the open‑source developers of FastQC, Bowtie2, SAMtools/BCFtools, Snakemake, and the Bioconda community.