CNVcaller

CNVcaller detects copy number variations (CNVs) in population sequencing data from nonhuman genomes with complex structures by leveraging read-depth approaches to identify and refine integrated CNV regions for population-scale analyses.

Key Features:

• Read-depth methodology: Implements read-depth approaches to detect CNVs and refine integrated copy number variation regions (CNVRs).
• High computational efficiency: Operates 1–2 orders of magnitude faster than CNVnator and Genome STRiP on complex genomes, demonstrated by processing CNV detection for 232 goats in 1.4 days on a single compute node.
• Scalability to large sample sizes: Designed to handle population-scale sequencing data and large cohorts.
• Robustness to genome assembly issues: Mitigates effects of high proportions of gaps and misassembled duplications in nonhuman reference assemblies, with Mendelian consistency observed in sheep trios.
• Superior accuracy and sensitivity: Shows improved accuracy and sensitivity, particularly for detecting duplications, in evaluations using real sheep and human data.
• Fast generalized detection algorithms: Includes fast generalized detection algorithms to overcome computational barriers in large-scale, complex-genome datasets.
• VCF genotype output: Produces results as VCF-format genotype files for downstream integration.

Scientific Applications:

• Population genetics: Enables population genetic analyses by detecting and refining CNVs across large cohorts.
• Comparative and nonhuman genomics: Facilitates functional study of CNVs in diverse nonhuman species with complex genomic architectures.
• GWAS and QTL integration: Provides VCF genotype outputs that can be integrated into Genome-Wide Association Studies and Quantitative Trait Loci research to investigate CNV contributions to phenotypic variation and disease.

Methodology:

Uses read-depth methodologies and fast generalized detection algorithms to identify and refine CNVRs and exports results as VCF genotype files.