Evaluating the Calling Performance of a Rare Disease NGS Panel for Single Nucleotide and Copy Number Variants

Area of Science:

• Genetics and Genomics
• Bioinformatics
• Clinical Diagnostics

Background:

• Clinical next-generation sequencing (NGS) requires application-specific optimization for variant detection.
• Accurate identification of single nucleotide variants (SNVs) and copy number variants (CNVs) is critical for diagnosing rare genetic diseases.

Purpose of the Study:

• To analyze the performance of different variant detection programs for SNVs and CNVs using an NGS panel in patients with a rare disease.
• To evaluate the calling efficiency and accuracy of various bioinformatics tools for variant detection in a clinical setting.

Main Methods:

• Sequenced 30 genes in 83 patients with hereditary spastic paraplegia using an NGS panel.
• Compared variant calls from LifeScope, GATK UnifiedGenotyper, and GATK HaplotypeCaller against Sanger sequencing for SNVs and indels.
• Assessed copy number variant (CNV) detection using ExomeDepth, panelcn.MOPS, and CNVPanelizer for five multiexon deletions.

Main Results:

• 94% of SNVs and 20% of indels were consistently detected by all SNV callers.
• Two SNVs were missed due to rare reference alleles, and one was missed in a low-coverage region.
• ExomeDepth accurately detected all 5/5 multi-exon deletions, panelcn.MOPS detected 4/5, and CNVPanelizer detected 3/5.

Conclusions:

• NGS variant caller performance for SNVs depends on variant type and sequencing coverage.
• CNV detection algorithms can identify large deletions from NGS data, but sensitivity is affected by coverage, reference set, and deletion size.
• Incorporating multiple variant callers into NGS pipelines is recommended to maximize detection efficiency for clinical applications.