Abstract
BACKGROUND: Copy number variation (CNV) analyses-often inferred from DNA-methylation data-depict alterations of DNA quantities across chromosomes and have improved tumour diagnostics and classification. For the analyses of larger case series, CNV-features of multiple samples have to be combined to reliably interpret tumour-type characteristics. Established workflows mainly focus on the analyses of singular samples and do not support scalability to high sample numbers. Additionally, only plots showing the frequency of the aberrations have been considered. RESULTS: We present the Cumulative CNV (CCNV) R package, which combines established segmentation methods and a newly implemented algorithm for thorough and fast CNV analysis at unprecedented accessibility. Our work is the first to supplement well-interpretable CNV frequency plots with their respective intensity plots, as well as showcasing the first application of penalised least-squares regression to DNA methylation data. CCNV exceeded existing tools concerning computing time and displayed high accuracy for all available array types on simulated and real-world data, verified by our newly developed benchmarking method. CONCLUSIONS: CCNV is a user-friendly R package, which enables fast and accurate generation and analyses of cumulative copy number variation plots.