Correlation-based Linking of Epigenomic Regions to Target Genes in Bulk and Single Cell Epigenomic Data

Single-cell sequencing has catalyzed a significant shift in biological modeling and hypothesis generation. In this study, we present advanced algorithms that utilize multiple modalities for peak-gene linking via correlation-based mechanisms. This approach provides a robust framework for bias correction and noise reduction at single-cell resolution. Our research introduces a novel algorithm that employs peak-gene linking to integrate epigenomic data, thereby translating it into transcriptomic data. This integration facilitates comprehensive secondary analyses on meta-cells and sub-cell types. Crucially, our methodology enables swift computation of modules from any single-cell assay, promoting exploration of intricate biological and disease mechanisms that might remain unmodeled with a pseudo-bulk approach. By combining snRNA-seq and snATAC-seq data, our method substantially outperforms equivalent tasks of gene expression estimation, showing promising results even with unpaired real data. Furthermore, the modeling of genomic peak modules using our algorithms uncovers additional signal potentially overlooked when examining single peaks. We envision correlation-based linking as a key aspect of future single-cell multiomic technology, as it allows for correlation between assays at the single-UMI level. As such, improved modeling of data at the single-cell level will enhance our understanding of complex gene regulatory networks and disease mechanisms.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2023

Author dc:creator

• James, Benjamin Thomas

Advisor dc:contributor.advisor

• Kellis, Manolis

Statement dc:rights

• Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
• Copyright retained by author(s)

Licence dc:rights.uri

• https://creativecommons.org/licenses/by-nc/4.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/151538

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/151538