Sequence-Structure-Function Modeling for the 3D Genome

In this talk, Dr. Katherine Pollard, director of the Gladstone Institute of Data Science and Biotechnology, will discuss how the human genome sequence folds in three dimensions (3D) into a rich variety of locus-specific contact patterns.

Motivated by a lack of models for relating DNA sequence mutations to changes in genome structure and function, Dr. Pollard’s team has developed “Akita,” a deep convolutional neural network that can be used to accurately predict genome folding using only DNA sequence as the input. Akita has enabled rapid in silico predictions for effects of DNA sequence mutations on the 3D genome structure, including differences in genome folding across species and in disease cohorts. This prediction-first strategy exemplifies Dr. Pollard’s vision for a more proactive, rather than reactive, role for data science in biomedical research.

Questions can be submitted during the meeting to NLMEPInfo@mail.nih.gov.

This lecture is part of the NIH National Library of Medicine’s (NLM’s) Ada Lovelace Computational Health Lecture Series. Individuals with disabilities who need reasonable accommodation to participate in this event should contact Valerie Bartlett, bartletv@mail.nih.gov, or the Federal Relay at 1-800-877-8339. Requests should be made five days in advance.