PennCNV in whole-genome sequencing data.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• High-throughput sequencing data enhances genomic analysis.
• Copy-number variation (CNV) calling tools face accuracy limitations due to noisy sequencing data.
• Assessing existing CNV algorithms on new data types is crucial.

Purpose of the Study:

• Evaluate PennCNV's performance on whole genome sequencing (WGS) data.
• Adapt PennCNV for WGS by processing BAM files for coverage and B allele frequency (BAF).
• Compare PennCNV-Seq against other CNV calling tools.

Main Methods:

• Processed BAM files to extract log R ratio (LRR) and B allele frequency (BAF).
• Utilized high-quality NA12878 sample and 10 artificial samples with known CNVs.
• Compared PennCNV-Seq performance for deletions, duplications, copy number variations, and loss-of-heterozygosity (LOH).

Main Results:

• PennCNV-Seq demonstrated accuracy in identifying various CNVs.
• The tool performed well across different copy number states and LOH events.
• Results indicate robustness in calling CNVs from WGS data.

Conclusions:

• PennCNV-Seq accurately detects copy-number variations in WGS data.
• The algorithm can be integrated into existing CNV calling pipelines.
• Provides accurate copy number reporting for specific genomic regions.