SSP Forum: Kelly Cochran on Deep Learning and Biology
The
Symbolic Systems Forum
presents
Deep Learning and Biology: Predicting Where Proteins Bind DNA Across Species
Kelly Cochran
Computer Science Department
Monday, January 31, 2022
12:15-1:15 pm
ABSTRACT:
What does research in computational biology look like, specifically at the intersection of deep learning and genomics? In this talk, we'll walk through an example research project to illustrate how comp-bio research can be driven by the knowns and unknowns of biology and how we can leverage deep learning to uncover biological insights. We'll begin with the concept of gene regulation: how our human bodies are made complex and kept alive by carefully orchestrated interactions between DNA and proteins. Other mammals have similar processes going on inside their cells – but how similar, exactly? Mice and humans both have hearts and lungs and livers, sure; but can we perform biological experiments in mice and reasonably expect the results to transfer to humans as well? We'll investigate this question through the lens of deep learning models trained to predict DNA-protein interactions. Note: no background in biology or machine learning is needed to follow this talk.
A NOTE ON THE RECORDING OF EVENTS:
If a decision has been made in advance to record an event and to make it available for later public viewing, the event announcement will usually state this. In many cases, however, decisions to record, and/or to make a recording available publicly, are not finalized before an event is announced. Availability decisions for recordings are often subject to what speakers prefer after an event has concluded, among other considerations that may include usage rights for material used in an event, as well as the need for, and practicality of, editing. When recordings are made publicly available, they will be linked within the original event announcement on the Symsys website in the days or weeks following an event. Unfortunately, we cannot follow up on individual requests for more information about whether and when a recording may become available if it is not yet posted publicly.