Unsupervised Learning to Unravel Differential Cell Fate Outcomes

By Kristen Naegle

Biological Engineering, Washington University at St. Louis, St. Louis, MO

Published on

Abstract

Cells are constantly receiving cues from the outside world and responding to them by altering their physiological fate by transducing this signal via intracellular biochemical networks. An important mechanism that many cell networks utilize to transduce these signals is the regulation of protein tyrosine phosphorylation. The addition of a phosphate group to a tyrosine residue can cause changes in protein activity, localization, and interactions.  Given the sheer size and complexity of these biochemical networks, computational methods are needed to help unravel how network flow is established and altered by changes in cell context, such as different tissue types or alterations as we seen in diseases like cancer.  In this talk, I will introduce computational methods we have developed that have pointed to specific network alterations in HER2 ovexpressing cells that result in increased migration and the cell-based experiments we have used to test computationally-derived hypotheses.

Bio

Kristen Naegle Dr. Naegle has a bachelor's and master's degrees in Electrical Engineering from University of Washington and a Ph.D. in Bioengineering from MIT (2010). She trained jointly with Forest White and Doug Lauffenburger to build methods to uncover the function of novel tyrosine phosphorylation in the epidermal growth factor receptor network. As a postdoctoral associate with Michael Yaffe, she experimentally tested the hypothesis that robust clustering of tyrosine phosphorylation dynamics predicted novel, transient protein interactions. She joined the faculty in Biomedical Engineering at Washington University in St. Louis in 2012, where her lab continues to develop computational and molecular tools, integrating across systems approaches, to infer and test the function of tyrosine phosphorylation on proteins and within networks. Dr. Naegle's computational approaches are diverse in scope, covering unsupervised and supervised learning and evolutionary approaches, but share the fundamental ideas of reproducibility, robustness, and accessibility. 

Sponsored by

Cite this work

Researchers should cite this work as follows:

  • Kristen Naegle (2017), "Unsupervised Learning to Unravel Differential Cell Fate Outcomes," http://nanohub.org/resources/25400.

    BibTex | EndNote

Time

Location

MJIS 1001, Purdue University, West Lafayette, IN

Tags

  1. nano/bio