Identification of copy number variants from exome sequence data.

Area of Science:

• Genomics
• Bioinformatics
• Genetic Diagnostics

Background:

• Exome sequencing is cost-effective for genetic disease mutation detection.
• Computational prediction of copy number variants (CNVs) from exome data remains challenging, especially for smaller CNVs (1-4 exons).
• Standard array comparative genomic hybridization (aCGH) has limitations in exonic CNV discovery due to low probe density.

Purpose of the Study:

• To develop a protocol for detecting exonic CNVs, including shorter ones (1-4 exons).
• To combine computational prediction algorithms with a high-resolution custom CGH array for improved CNV detection.

Main Methods:

• Evaluated six published CNV prediction programs (ExomeCNV, CONTRA, ExomeCopy, ExomeDepth, CoNIFER, XHMM) and an in-house modification (ExCopyDepth).
• Tested computational predictions using a custom CGH array (exaCGH) designed to capture all exons.
• Assessed the sensitivity and false positive rate of each computational program for CNV detection.

Main Results:

• ExomeCopy and the in-house ExCopyDepth algorithm demonstrated the highest capability in detecting shorter CNVs.
• The custom exaCGH array was used for validation of computational predictions.
• Performance metrics (sensitivity, false positive rate) were calculated for each program.

Conclusions:

• Six computational programs were assessed for their ability to predict short (1-4 exon) CNVs.
• A protocol combining computational prediction and custom aCGH experiments is proposed for identifying and confirming shorter exonic CNVs.