A framework for variation discovery and genotyping using next-generation DNA sequencing data.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Advances in sequencing technology enable comprehensive genetic variation cataloging.
• Processing large volumes of raw sequencing data requires sophisticated computational methods for high-quality variant calls.

Purpose of the Study:

• To present a unified analytic framework for simultaneous discovery and genotyping of genetic variation in multiple samples.
• To achieve sensitive and specific variant calling across diverse sequencing technologies and experimental designs.

Main Methods:

• The framework integrates read mapping, local realignment around indels, and base quality score recalibration.
• It employs SNP discovery and genotyping, followed by machine learning to differentiate true variants from sequencing artifacts.
• The process is implemented in the Genome Analysis Toolkit (GATK).

Main Results:

• The unified framework demonstrates sensitive and specific results across five sequencing technologies and three experimental designs.
• Successful application to deep whole-genome, whole-exome capture, and low-pass 1000 Genomes Project datasets.

Conclusions:

• The presented analytic framework provides a robust approach for genetic variation analysis.
• It is applicable to various sequencing data types and experimental setups, supporting research in human genetics, ancestry, and evolution.