[Colloquium] Reminder - Matthew McPartlon Dissertation Defense/May 22, 2023

[Megan Woodward]

This is an announcement of Matthew McPartlon’s Dissertation Defense
===============================================
Candidate: Matthew McPartlon

Date: Monday, May 22, 2023

Time: 9:00 am CST

Location: JCL 223

Remote Location: https://uchicago.zoom.us/j/95179647999?pwd=Q1BwTndCTDF0MjhnZUFhOVNMbi9UZz09<https://urldefense.com/v3/__https://www.google.com/url?q=https:**Auchicago.zoom.us*j*95179647999*pwd*3DQ1BwTndCTDF0MjhnZUFhOVNMbi9UZz09&sa=D&source=calendar&ust=1684785610252984&usg=AOvVaw3_Q2IwAa6KfKHdAwrN6uLu__;Ly8vLz8l!!BpyFHLRN4TMTrA!4xAcY7OaukpCFsdYiIyDzeLFt0P5KoejonxyJPCy30Ujt74DHn9Q1OQ6ZqpEk0mmyBemLWxW4mjsJt_tiDlLLL38oP6dt_Rf$>
Meeting ID: 951 7964 7999 Passcode: 917327

Dissertation Paper Title: Accelerating Protein Design with Deep Learning

Abstract: The human proteome comprises tens of thousands of proteins, each tailored for a specific
function by the selective pressures of evolution. The field of protein design seeks to develop
proteins with new or enhanced functions at will, ultimately bypassing the evolutionary clock.
In this thesis, we introduce general machine-learning methods for accelerating protein design,
focusing on modeling protein structure.

First, we propose an approach for fixed-backbone design (Chapter 2), the problem of
designing primary sequence and side-chain rotamers for a given backbone conformation.
Whereas classic approaches formulate sequence and rotamer design tasks separately, we offer
a single approach to predict both simultaneously. To realize this, we develop a deep neural
network that effectively leverages backbone coordinates. By exploiting backbone geometry,
we can efficiently represent atomic microenvironments at the coordinate level and ultimately
avoid discrete rotamer sampling. This results in more robust designs and accurate quality
estimates for downstream tasks.

Next, we introduce a framework for flexible protein-protein docking (Chapter 3), the
task of determining the structure of a protein complex given the unbound structures of its
components. Traditional docking methods are limited by their reliance on empirical physics-
based scoring functions, inability to accommodate conformational flexibility, and failure to
incorporate information on binding sites. To address these challenges, we propose an end-to-
end approach that can model conformational changes and target specific interactions while
significantly reducing computational time. As one of the pioneering deep learning methods
for this task, we uncover key determinants underlying our success and provide important
insights for future research. Finally, we highlight our method’s generality by extending it to
simultaneously dock and co-design the sequence and structure of antibody complementarity-
determining regions targeting a specified epitope

Advisors: Jinbo Xu & Andy Drucker

Committee Members: Jinbo Xu, Andy Drucker, Janos Simon


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20230522/c68ce6f4/attachment.html>