On the critical evaluation and confirmation of germline sequence variants identified using massively parallel...

Area of Science:

• Genomics and Molecular Biology
• Bioinformatics and Computational Biology

Background:

• Massively parallel sequencing (MPS) is increasingly used in research and clinical settings.
• The necessity for validating DNA variants identified by MPS using alternative methods remains a subject of discussion.
• Read depth statistics and genotype quality scores are crucial for evaluating variants directly from MPS data.

Purpose of the Study:

• To validate DNA variants identified through MPS.
• To investigate common sources of false-positive and false-negative findings in MPS data.
• To assess the influence of bioinformatics pipelines on variant evaluation metrics.

Main Methods:

• Validation of MPS-identified variants using Sanger sequencing.
• Simultaneous MPS and Sanger sequencing analysis of selected gene exons.
• Detailed examination of false-positive and false-negative results to identify error patterns.

Main Results:

• Identified common error sources including low read depth, incomplete reference genomes, indel realignment issues, and microsatellite-associated amplification errors.
• Demonstrated that manual revision of aligned reads, based on variant distribution patterns, can identify these errors.
• Observed that quantitative metrics (read counts, allelic balance) and genotype quality scores are pipeline-dependent.

Conclusions:

• Thorough manual review is essential for accurate variant interpretation from MPS data.
• The choice of bioinformatics pipeline significantly affects variant quality metrics.
• Establishing laboratory- and pipeline-specific thresholds for quantitative and quality scores is critical for reliable variant calling.