gatk_base_recalibrator

gatk_base_recalibrator recalibrates base quality scores to correct systematic sequencing errors and improve the accuracy of downstream variant calling.

Key Features:

• Two-Pass Processing: Operates as an integral component in a two-pass processing workflow that targets sites not cataloged in dbSNP.
• By-Locus Traversal: Performs locus-specific traversal and analysis to tailor recalibration to individual genomic sites.
• Error Identification (non-dbSNP mismatches): Treats mismatches with the reference genome at non-dbSNP sites as indicative of potential poor base quality.
• Covariate-Based Recalibration: Generates recalibration using user-defined covariates including read group, reported quality score, cycle position, and dinucleotide context.
• Recalibration Table Generation: Produces recalibration tables that capture covariate-specific error profiles for downstream application.
• Statistical Adjustment: Applies statistical models to adjust reported base quality scores based on observed error patterns.

Scientific Applications:

• Genomic Variant Calling: Improves accuracy of single nucleotide polymorphism (SNP) and other variant detection by correcting systematic base-quality biases.
• Metagenomics: Enhances base-call quality in complex metagenomic datasets and supports downstream analyses and tools such as MetaGenSense.
• Phylogenetic Analysis: Improves sequence data quality for phylogenetic inference workflows and platforms such as NGphylogeny.fr.

Methodology:

Performs an initial pass through aligned sequencing data to identify sites requiring recalibration; generates covariate-specific recalibration tables (read group, reported quality score, cycle position, dinucleotide context); and applies statistical models to adjust base quality scores as part of a two-pass workflow.