genoCN

genoCN performs joint inference of copy number states and genotype calls from high-density SNP arrays to detect inherited copy number variations (CNVs) and somatic copy number aberrations (CNAs) in genomic DNA.

Key Features:

• Joint CN and genotype analysis: Simultaneously analyzes copy number states and genotype calls from high-density SNP arrays for integrated interpretation of genomic DNA.
• Components - genoCNV and genoCNA: Provides separate modules, genoCNV for inherited CNVs and genoCNA for somatic CNAs, to tailor analysis to each alteration type.
• Data-driven parameter estimation: Estimates model parameters directly from the input data rather than relying on predetermined parameters.
• Modeling of tissue contamination: Incorporates explicit modeling of tissue contamination effects in tumor samples to improve CNA inference.
• Matched tumor-normal leveraging: Uses genotype calls from matched normal tissue, when available, to refine CNA calls in corresponding tumor samples.
• Distinction between CNVs and CNAs: Accounts for biological differences between shorter, inherited CNVs and acquired, somatic CNAs in tumors.
• Empirical validation: Demonstrated on datasets including 162 HapMap individuals and a glioblastoma dataset, reporting CNV/CNA identification and high-quality genotype calls.

Scientific Applications:

• Germline CNV discovery: Identification and characterization of inherited copy number variations from SNP array data.
• Somatic CNA profiling in cancer: Detection and analysis of copy number aberrations in tumor samples, accounting for contamination and using matched normals.
• Genotype calling from SNP arrays: Generation of high-quality genotype calls concurrent with copy number analysis.
• Tumor-normal comparative studies: Comparative analysis of matched tumor and normal samples to distinguish somatic alterations from germline variation.

Methodology:

Implements a statistical framework that simultaneously analyzes copy number states and genotype calls from high-density SNP arrays, estimates model parameters from the data, explicitly models tissue contamination for CNA analysis, and leverages genotype calls from matched normal tissue when available.