CONY

CONY detects copy number variations (CNVs) from sequencing read-depth signals using a Bayesian hierarchical model to infer absolute and relative copy numbers for individual samples and case-control pairs.

Key Features:

• Bayesian hierarchical model: Models sequencing read-depth data with a Bayesian hierarchical framework that integrates prior information and quantifies uncertainty for CNV detection.
• Reversible-Jump MCMC (RJMCMC) inference: Uses reversible-jump Markov chain Monte Carlo to perform model selection and parameter estimation across varying numbers of copy-number states.
• Data versatility: Applies to both individual-sample analyses for estimating absolute copy numbers and case-control pairs for identifying patient-specific relative CNVs from read-depth data.
• Performance evaluation: Evaluated via simulations and experimental data from the 1000 Genomes Project, demonstrating accuracy across high and low coverage levels.
• Comprehensive CNV detection: Detects both absolute and relative CNVs from sequencing read-depth signals.

Scientific Applications:

• Disease genetics: Identification of genetic copy-number alterations associated with diseases.
• Evolutionary biology: Analysis of copy-number variation in evolutionary biology studies.
• Personalized medicine: Detection of patient-specific CNVs to inform personalized medicine approaches.
• Large-scale genomic and clinical studies: Application in large-scale genomic projects and clinical diagnostics through accurate CNV detection across sample types and coverage levels.

Methodology:

Analyzes sequencing read-depth signals using a Bayesian hierarchical model with inference via reversible-jump Markov chain Monte Carlo (RJMCMC); performance assessed using simulations and experimental data from the 1000 Genomes Project.