ExomeDepth

ExomeDepth detects copy number variants (CNVs) from exome sequencing read count data by modeling technical variability with an internal reference to improve detection of small and heterozygous exon-spanning deletions.

Key Features:

• Read count modeling: Uses a robust statistical model for exome sequencing read count data to call CNVs.
• Optimized reference set: Constructs an optimized, dataset-specific reference set tailored to each sample cohort.
• Internal reference model: Controls technical variability by using an internal reference rather than relying on external samples or sample combinations.
• Small CNV sensitivity: Detects small and heterozygous deletions spanning one to two exons that are challenging for traditional read depth methods.
• Empirical call rates: Produced between 170 and 250 exonic CNV calls per sample in an application to 24 patients with primary immunodeficiencies.
• Validated discoveries: Enabled discovery of causative deletions in the genes GATA2 and DOCK8 in applied studies.
• Implementation: Provided as an implementation in the R programming environment.

Scientific Applications:

• Mendelian disorder genetics: Detection of CNVs in targeted exome sequencing experiments to investigate genetic causes of Mendelian disorders.
• Primary immunodeficiency studies: Applied to exome data from 24 patients with primary immunodeficiencies, yielding high numbers of exonic CNV calls and identifying causative deletions in GATA2 and DOCK8.
• Small-scale CNV discovery: Identification of small, heterozygous exon-spanning deletions relevant to studies of complex traits and rare variant analyses.

Methodology:

Models exome read counts, constructs a dataset-specific optimized reference set, uses an internal reference model to control technical variability, and calls CNVs including small heterozygous deletions spanning one to two exons; implemented in R.