sciCNV: High-throughput paired profiling of transcriptomes and DNA copynumber variations at single cell resolution

Abstract

Chromosome copy number variations (CNVs) are a near-universal feature of cancer however their effects on cellular function are incompletely understood. Single cell RNA sequencing (scRNA-seq) can reveal cellular gene expression however cannot directly link this to CNVs. Here we report new normalization methods (RTAM1 and -2) for scRNA-seq that improve gene expression alignment between cells, enhancing gene expression comparisons and the application of scRNA-seq to CNV detection. We also report sciCNV, a pipeline for inferring CNVs from RTAM-normalized data. Together, these tools provide dual profiling of transcriptomes and CNVs at single-cell resolution, enabling exploration of the effects of cancer CNVs on cellular programs. We apply these tools to multiple myeloma (MM) and examine the cellular effects of cancer CNVs +8q. Consistent with prior reports, MM cells with +8q22-24 upregulate MYC, MYC-target genes, mRNA processing and protein synthesis, verifying the approach. Overall, we provide new tools for scRNA-seq that enable matched profiling of the CNV landscape and transcriptome of single cells, facilitate deconstruction of the effects of cancer CNVs on cellular reprogramming within single samples.