Calling copy number variations from single-cell multiomic data to understand cancer evolution

• Applicant: Professorin Dr. Maria Colomé-Tatché
• Subject Area: Bioinformatics and Theoretical Biology
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 553739126

Aneuploidy and copy number variations (CNVs) are known to be associated with many diseases, particularly cancer. Despite its clinical relevance, the role of aneuploidy in tumorigenesis and the impact of CNV heterogeneity on tumor fitness remain poorly understood. This knowledge gap calls for the development of advanced techniques to comprehensively investigate aneuploidy and CNVs at the single cell level. Multiple layers of genomic information can be measured using single cell sequencing technologies. One of the most recently developed techniques is single cell multiome sequencing, where the gene expression and the chromatin state of the same single cell are interrogated. My group has recently developed a computational method to call single cell CNVs from single cell open chromatin data (scATAC-seq). We have shown how our method robustly identifies the copy number states of the genome, compared to a ground truth obtained using whole genome DNA sequencing. In this proposal, our aim is to leverage the power of single cell multiome data to integrate genomic, transcriptomic, and epigenomic information to comprehensively study CNVs at the single cell level and shed light on the mechanisms of gene regulation in aneuploid cells. We will begin by comprehensively benchmarking existing CNV calling methods that use single cell RNA sequencing (scRNA-seq) data, and determine their performance. This benchmarking study will be a valuable resource for the community and in turn it will inform us about the best strategy to call CNVs from expression data. We will then focus on the development of novel methods to call CNVs from single cell multiome data, harnessing the complementary insights offered by both scRNA- seq and scATAC-seq. Finally, we will apply the developed methods to single cell multiome data from chromosomally unstable cancers, which will be produced as part of the project. Applying the newly developed methods to these datasets, we will unravel the intricate relationship between expression, chromatin accessibility, and copy number variations during tumor evolution and response to treatment. This project will provide a robust computational framework to investigate single-cell CNV heterogeneity from multiome data. The proposed research will significantly advance our understanding of the role of aneuploidy in disease, and the computational methods developed will have far-reaching applications in the oncology field and beyond.

• DFG Programme: Research Grants