Human variant calling & annotation pipeline

This bash pipeline for processing human whole-genome sequencing (WGS) data, including quality control, alignment, variant calling, and variant annotation.

---

Tools used

• FastQC – Quality control checks on raw FASTQ files
• MultiQC – Aggregates FastQC reports
• BWA – Aligns reads to the human reference genome
• Samtools – Converts and indexes BAM files
• Picard – Sorts BAM files and marks PCR duplicates
• GATK – Performs indel realignment, base recalibration, and variant calling
• Annovar – Annotates variants using multiple databases

---

Input requirements

1. list.txt – A file with one sample name per line (no extensions).
2. Paired-end FASTQ files for each sample (e.g., sample_R1.fastq.gz, sample_R2.fastq.gz).
3. Human reference genome (human_g1k_v37.fa) and known variant sites.
4. Annovar and the humandb/ directory with annotation databases.

---

How to run

Make sure your script is executable:

chmod +x variant_pipeline.sh

Run the script from your working directory:

./variant_pipeline.sh

Ensure the directory contains list.txt and all required FASTQ files.

---

Output

For each sample, this pipeline generates:

• Aligned, sorted, deduplicated, and indexed BAM files
• VCF files with called variants
• Annovar .avinput and .csv annotation outputs
• QC reports and summary via MultiQC

---

Annovar annotation databases

The script uses the following Annovar databases:

• refGene, cytoBand, snp138, esp6500siv2_all
• 1000g2015aug (AFR, EAS, EUR), avsift, clinvar
• dbnsfp30a, exac03, icgc21, caddgt20, gnomad

Ensure all databases are downloaded in ~/annovar/humandb/.

---

Quote to remember

"Bioinformatics is not something you are taught, it’s a way of life."

---

Author

Gerald Mboowa
Email: ...@gmail.com

---

License

This script is provided under the MIT License.