PennCNV

PennCNV detects copy number variations (CNVs) from Illumina high-density SNP genotyping array data using a Hidden Markov Model to enable kilobase-scale resolution CNV identification.

Key Features:

• Hidden Markov Model Framework: Employs a Hidden Markov Model to analyze SNP array signal intensities for CNV calling.
• Input Data: Analyzes Illumina high-density SNP genotyping array data including total signal intensity and allelic intensity ratio at each SNP marker.
• Inter-SNP Distance: Incorporates the genomic distance between neighboring SNPs to inform CNV boundary assessment.
• Allele Frequency Incorporation: Uses SNP allele frequency data to refine copy number state inference.
• Pedigree Information: Can incorporate pedigree information to improve CNV identification accuracy.
• High-Resolution Detection: Achieves a median CNV detection size of approximately 12 kilobases.
• Empirical Performance: Applied to 112 HapMap individuals and reported about 27 CNVs detected per individual on average.
• Identification of Novel Variations: By excluding common rearrangements in lymphoblastoid cell lines, identified a 3.3% fraction of offspring CNVs not detected in parents (CNV-NDPs).

Scientific Applications:

• Human genetic variation and disease association: Characterizing CNVs from SNP array data to study contributions to phenotypic variation and disease susceptibility.
• Population genetics and evolutionary biology: Fine-scale mapping of CNVs for population-level and evolutionary analyses.
• Genetic disorder research and association studies: High-resolution CNV detection to support studies of genetic disorders and CNV-based association analyses.

Methodology:

Uses a Hidden Markov Model to analyze total signal intensity and allelic intensity ratio from Illumina SNP genotyping arrays, incorporating inter-SNP distance, SNP allele frequency, and pedigree information.