Whole-genome sequencing analysis of genomic copy number variation (CNV) using low-coverage and paired-end strategies is highly efficient and outperforms array based CNV analysis

CNV analysis is an integral component to the study of human genomes in both research and clinical settings. Array-based CNV analysis is the current first-tier approach to analyzing CNV in clinical cytogenetics. Decreasing costs in high-throughput sequencing and cloud computing have opened doors for the development of sequencing-based CNV analysis pipelines with fast turnaround times. We carry out a systematic and quantitative comparative analysis of several low-coverage whole-genome sequencing (WGS) strategies to detect copy number variation (CNV) in the human genome and demonstrate that low-coverage WGS is an efficient and comprehensive tool for CNV analysis, in particular when compared to the array-based analysis used in clinical settings. We compared the CNV detection capabilities of WGS strategies (short-insert, 3kb-, and 5kb-insert mate-pair) each at 1X, 3X, and 5X coverages relative to each other and to all current high-density oligonucleotide arrays. As a benchmarking device, we used the Gold Standard and Silver Standard CNVs generated for the genome of 1000-Genomes-Project CEU subject NA12878. Overall, low-coverage WGS detects drastically more Gold Standard CNVs compared to arrays and is accompanied with many fewer CNV calls without secondary validation. Furthermore, we also show that WGS (at ≥1X coverage) is able to detect all seven validated deletion CNVs u003e100 kb in the NA12878 genome whereas only one of these deletions is detected by most arrays. Lastly, we show that the much larger 15 Mbp Cri-du-chat deletion can be clearly seen at even just 1X genomic sequencing coverage with short-insert paired-end WGS.