CNVkit: Genome-Wide Copy Number Detection and Visualization from Targeted DNA Sequencing

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Germline copy number variants (CNVs) and somatic copy number alterations (SCNAs) are crucial in genetic disorders and cancer.
• Massively parallel sequencing is a common method for detecting CNVs/SCNAs via read depth variations.
• Targeted re-sequencing presents challenges due to coverage gaps and capture biases, limiting comprehensive analysis.

Purpose of the Study:

• To present a novel method for copy number detection using targeted sequencing data.
• To introduce CNVkit, a software package designed for accurate and comprehensive copy number analysis.
• To address limitations of targeted sequencing for inferring copy number variations across the genome.

Main Methods:

• Developed CNVkit software integrating targeted and off-target sequencing reads for copy number inference.
• Normalized read counts against a pooled reference and corrected for GC content, target footprint, and repetitive sequences.
• Evaluated CNVkit performance against array comparative genomic hybridization (aCGH) for copy number changes.

Main Results:

• Achieved exon-level resolution in targeted regions and kilobase-level resolution in non-targeted regions.
• Successfully inferred genome-wide copy number at 100-kilobase resolution from targeted sequencing of 293 genes.
• Demonstrated that CNVkit effectively corrects for major sources of bias in sequencing read depth.

Conclusions:

• CNVkit provides a robust method for copy number detection in targeted sequencing, overcoming previous limitations.
• The software offers user-friendly features including visualizations and reporting for integration into analysis pipelines.
• CNVkit enables accurate and comprehensive copy number analysis across the genome from targeted sequencing data.