Community-Driven Copy Number Variant Discovery at Scale: Results from a Rare Disease Genomics Hackathon.

Area of Science:

• Genomics
• Rare Diseases
• Bioinformatics

Background:

• Copy number variants (CNVs) are significant contributors to rare genetic diseases.
• Detecting and interpreting CNVs from short-read genome sequencing (srGS) data is challenging, particularly at scale.
• Vast amounts of existing srGS data remain underutilized for identifying clinically relevant CNVs.

Purpose of the Study:

• To develop and apply scalable CNV analysis workflows for srGS data.
• To accelerate the diagnosis and genetic discovery in rare disease cohorts.
• To address the challenge of CNV detection and interpretation in large-scale datasets.

Main Methods:

• Utilized Parliament2 for structural variant (SV) calling and Mosdepth/SLMSuite for read-depth-based quality control and CNV detection.
• Applied R Shiny for visualization and developed an SV/CNV variant database with annotation.
• Employed DBSCAN clustering for allele frequency estimation and a 3-way annotation strategy for interpretation.

Main Results:

• Developed and applied scalable CNV analysis pipelines to srGS data from three rare disease cohorts.
• Identified high-confidence CNVs and streamlined interpretation, yielding 39 candidate pathogenic SVs within 2 days.
• Demonstrated rapid filtering, prioritization, and visualization of clinically relevant variants.

Conclusions:

• A community-driven Hackathon successfully demonstrated the feasibility of scalable CNV analysis.
• The developed tools and workflows significantly accelerate diagnosis and discovery in rare disease research.
• This approach enhances the utility of existing srGS data for identifying genetic causes of rare diseases.