Alignment

The alignment module maps indexed and trimmed reads to the reference genome, filters high-quality results, and identifies PCR duplicates.

Workflow

graph LR
    A[Trimmed FASTQ] --> B[STAR Alignment]
    B --> C[BAM Filtering]
    C --> D[Mark Duplicates]
    D --> E[BigWig Tracks]

STAR Alignment

STAR is recognized for its high speed and accuracy in RNA-seq alignment.

Features

• Support for split-read mapping (splicing).
• Optimized for large genomes.
• Integrated sorting and indexing.

Configuration

By default, STAR is configured with:

• outFilterMultimapNmax: 20
• outSAMunmapped: Within
• quantMode: GeneCounts

Post-Alignment Filtering

After alignment, the pipeline filters the resulting BAM files to remove low-quality mappings and ensure data integrity.

BAM Filtering

• Minimizes noise by discarding reads with low mapping quality scores.
• Standard filter: q >= 255 (uniquely mapped reads only) or as configured.

Duplicate Identification (Picard)

Picard MarkDuplicates is used to identify and flag PCR duplicates, which is crucial for accurate quantification in low-input libraries.

Parameters & Defaults

Parameter	Default	Description
outFilterMultimapNmax	20	Max alignments for a read to be considered.
outSAMunmapped	Within	Controls if unmapped reads appear in output.
quantMode	GeneCounts	Enables gene-level quantification during alignment.

---

Results

Location	Description
results/alignments/star/	BAM files and STAR log files.
results/alignments/star_markdup/	BAM files with flagged PCR duplicates.
results/qc/star/	Alignment statistics and log files.
results/alignments/bigwig/	Genome browser tracks (BigWig).