GenomeAlignment

RNA-seq Alignment and Variant Calling Pipeline (SLURM)

Overview

GenomeAlignment is a SLURM-based RNA-seq alignment and variant calling pipeline designed for high-performance computing (HPC) environments. The pipeline processes pre-trimmed RNA-seq reads, aligns them to a reference genome using STAR, and performs RNA-seq–aware variant calling using GATK SplitNCigarReads and bcftools. The workflow is automated to process multiple sequencing runs and samples with minimal manual intervention.

Pipeline Workflow

For each RNA-seq sample, the pipeline performs:

Genome Alignment

Aligns reads to a reference genome using STAR and produces coordinate-sorted BAM files with gene-level read counts.

RNA-seq BAM Processing

Uses GATK SplitNCigarReads to properly handle spliced alignments prior to variant calling.

Variant Calling

Calls SNPs and indels using bcftools mpileup and bcftools call, producing compressed and indexed VCF files.

Requirements

System

Linux HPC environment SLURM workload manager

Software

STAR 2.7.11b samtools 1.22.1 bcftools 1.17 GATK 4.6.0.0

Each Run* directory is processed automatically FASTQ files must be pre-trimmed (e.g., using fastp) Single-end reads are assumed Reference Genome and Index The STAR genome index must be generated before running the pipeline.

Running the Pipeline

Submit the job using SLURM:

sbatch rnaseq_variant_calling.sh The pipeline will: Loop through all Run* directories Process all *_trimmed.fastq.gz files Skip samples where a .vcf.gz already exists

Output Files

For each sample, the pipeline generates: *_Aligned.sortedByCoord.out.bam — aligned reads *.split.bam — RNA-seq processed BAM *.vcf.gz — variant calls *.vcf.gz.tbi — VCF index

STAR gene count files

SLURM log files: pipeline.out pipeline.err

Notes

Input FASTQ files must be trimmed before running the pipeline Variant calls are unfiltered; downstream filtering is recommended Designed for RNA-seq variant discovery, not DNA-seq Modify STAR parameters for paired-end data if required