Inferring copy number variation from gene expression data: methods, comparisons, and applications to oncology

Abstract

Copy number variations (CNVs) are genomic events where the number of copies of a particular gene varies from cell to cell. Cancer cells are associated with somatic CNV changes resulting in gene amplifications and gene deletions. However, short of single-cell whole-genome sequencing, it is difficult to detect and quantify CNV events in single cells. In contrast, the rapid development of single-cell RNA sequencing (scRNA-seq) technologies has enabled easy acquisition of single-cell gene expression data. In this work, we employ three methods to infer CNV events from scRNA-seq data and provide a statistical comparison of the methods results. In addition, we combine the analysis of scRNA-seq and inferred CNV data to visualize and determine subpopulations and heterogeneity in tumor cell populations.