Robustness of Massively Parallel Sequencing Platforms

Area of Science:

• Genomics
• Bioinformatics
• Clinical Sequencing

Background:

• High-throughput sequencing (HTS) technologies have enabled large-scale clinical sequencing projects.
• Ensuring accuracy and reproducibility is vital for using HTS data in diagnostics and prognostics.

Purpose of the Study:

• To assess the robustness and clinical usability of a widely used HTS platform.
• To evaluate the accuracy and reproducibility of variant calling from whole genome shotgun (WGS) data generated at different centers.

Main Methods:

• Generated WGS sequence data from two human individuals across two genome sequencing centers.
• Analyzed sequence data to characterize single nucleotide polymorphisms (SNPs) and insertions/deletions (indels).
• Utilized standardized bioinformatics tools (BWA, SAMtools, GATK) for variant calling.

Main Results:

• Observed significant discrepancies in variant call sets between the two sequencing centers.
• Disagreements were predominantly located in genomic regions with repeats and segmental duplications.
• A small fraction of discordant variants were found in functionally relevant regions like exons and promoters.

Conclusions:

• HTS platforms provide sufficient data for initial clinical assessments.
• Variant predictions from HTS require confirmation through orthogonal methods before clinical application.
• Robustness of variant calling across different centers needs further improvement for reliable clinical use.