gatk_variant_eval

gatk_variant_eval evaluates variant calls and computes metrics to support accurate discovery and genotyping of genetic variation across multiple samples for studies of human disease, ancestry, and evolution.

Key Features:

• Unified Analytic Framework: Enables simultaneous discovery and genotyping of variation across multiple samples, providing sensitive and specific results across diverse sequencing technologies and experimental designs.
• Initial Read Mapping: Aligns raw sequence reads to a reference genome to provide the foundation for downstream analyses.
• Local Realignment Around Indels: Performs local realignment to correct misalignments caused by insertions and deletions and improve variant call accuracy.
• Base Quality Score Recalibration: Adjusts base quality scores to account for systematic sequencing errors and improve data reliability.
• SNP Discovery and Genotyping: Identifies potential single nucleotide polymorphisms (SNPs) and genotypes them across samples.
• Machine Learning Integration: Applies machine learning algorithms to discriminate true genetic variation from artifacts introduced by next-generation sequencing technologies.
• Application in Large-scale Projects: Applicable to datasets such as the 1000 Genomes Project, including deep whole-genome, whole-exome capture, and multi-sample low-pass sequencing data.

Scientific Applications:

• Human Disease Research: Produces accurate variant calls to support identification of genetic factors associated with disease.
• Ancestry and Evolution Studies: Enables cataloging of genetic variation across populations to inform studies of ancestry and evolutionary processes.

Methodology:

Computational steps explicitly include initial read mapping; local realignment around indels; base quality score recalibration; SNP discovery and genotyping; and machine learning algorithms to distinguish true variants from sequencing artifacts.