Event

Abstract: DNA is compactly packed in the nucleus of a cell; around 2 meters of human DNA is stored in a 10 μm nucleus. Evidence suggests that 3D genome organization facilitates long-range interactions between transcriptional regulatory elements, which control tissue-agnostic and tissue-specific gene activation and epigenetic silencing. Additionally, the evolutionary conservation of standard features of 3D genome organization, such as A/B compartments, TADs, loops, and aberrant nuclear organization in diseases like cancer, further indicates its potential role in gene regulation. My long-term research goal is to study how DNA sequence variations influence the 3D genome organization in humans and across mammalian evolution, how variations in genome organization impact transcriptional and post-transcriptional gene regulation across human tissues, and how this regulatory infrastructure is misregulated in cancer.

In this talk, I will discuss previous work constructing sequence-to-function models for different steps of gene regulation, including 3D genome architecture, gene expression and alternative splicing. By interpreting these models using existing and novel techniques, we identified several regulatory elements implicated in gene regulation. These include transcription factors and G-quadruplexes implicated in trans-contacts and RNA-binding proteins implicated in tissue-specific alternative splicing. Furthermore, we leveraged interpretable machine-learning techniques to find typical transcriptomic signatures of solid tumors in the form of shared protein-coding genes, lncRNAs and splice junctions.

Toward the end, I will outline my vision for developing novel computational methods for large-scale multi-omics data analysis. The methods will drive innovation in three key pillars in genomics research: technology development, methods development, and data generation, which will yield actionable insights into cell-specific gene regulation in heterogeneous healthy and diseased individuals.

Anupama Jha website.