Misannotation of multiple-nucleotide variants risks misdiagnosis

Area of Science:

• Genomics
• Bioinformatics
• Molecular Diagnostics

Background:

• Next-generation sequencing (NGS) analysis pipelines commonly misinterpret Multiple Nucleotide Variants (MNVs).
• MNVs are often incorrectly called as separate single nucleotide variants, deviating from their nature as single insertion-deletion events.
• This misannotation risks inaccurate diagnostic results and impacts downstream genetic interpretation.

Purpose of the Study:

• To evaluate the accuracy of widely used NGS pipelines, specifically GATK best practices, in calling Multiple Nucleotide Variants (MNVs).
• To identify alternative bioinformatics tools capable of correctly annotating MNVs.
• To propose improvements for NGS analysis guidelines to ensure accurate variant calling.

Main Methods:

• Utilized simulated sequencing data to assess variant calling accuracy.
• Re-analyzed diagnostic targeted gene panel sequencing data.
• Compared variant calls generated by GATK best practices against alternative bioinformatics tools.

Main Results:

• Demonstrated that GATK best practices pipeline miscalls Multiple Nucleotide Variants (MNVs).
• Identified alternative tools that accurately call and annotate MNVs.
• Highlighted the potential for misdiagnosis due to incorrect MNV calling in current standard practices.

Conclusions:

• Current GATK best practices for NGS data analysis are inadequate for accurately calling Multiple Nucleotide Variants (MNVs).
• Alternative variant calling tools offer a solution for correct MNV annotation in diagnostic settings.
• Integrating accurate MNV calling into established guidelines like GATK best practices is crucial for public genomic databases (e.g., gnomAD) and reliable clinical diagnoses.