PanCNV-Explorer: A pan-cancer resource to analyze copy number variations

Abstract

Introduction: Copy number variations (CNVs) are structural genomic alterations that involve changes in the number of copies of specific DNA regions. These variations can include deletions, duplications, and more complex rearrangements, and play a critical role in cancer progression by amplifying oncogenes, deleting tumor suppressor genes, or altering other key genomic regions. Despite the importance of CNVs in cancer biology, there is a lack of comprehensive resources that aggregate CNV data across multiple cancer types, impeding the exploration of their role in different malignancies. The PanCNV-Explorer was developed to fill this gap by providing a pan-cancer resource that integrates CNV data for over 21 cancer types. Methods: The PanCNV-Explorer was developed, utilizing data from repositories such as COSMIC, DepMap, DGV, and dbVAR. Amplifications and deletions can be visualized, with detailed chromosome-specific plots highlighting CNV patterns. In addition, a genome browser has been implemented to display regions of interest for searching for genes of potential relevance. Results: The database contains over 5 million CNVs from 15,809 samples, spanning 21 primary cancer types. Pathogenic CNVs from ClinVar and DepMap were compared to benign data from DGV and dbVAR, allowing a comprehensive analysis across tissues. CNV patterns were visualized for individual cancers. We demonstrate with use cases, focusing on common oncogenes such as MYC and tumor suppressor genes such as TP53, the ability of the resource to search for cancer driver genes. For example, chromosome 12 in pancreatic cancer showed frequent amplifications in the region containing the KRAS oncogene. In another example, we show how driver gene candidates already described in certain cancer entities can also be found with high frequency in other entities. Conclusion: The PanCNV-Explorer provides a valuable resource for pan-cancer CNV analysis, facilitating the exploration of CNV-driven tumor heterogeneity. The integrated visualization tools and genome browser enable detailed exploration of genomic regions, enhancing cancer research. As a valuable and easy-to-use platform that allows researchers to quantify and compare CNVs across cancer types, the PanCNV-Explorer is an essential tool for advancing cancer genomics research. The main page is available at https://mtb.bioinf.med.uni-goettingen.de/pancnv-explorer/.