University of Cambridge > > Artificial Intelligence Research Group Talks (Computer Laboratory) > Machine learning as an assay for high-dimensional biology

Machine learning as an assay for high-dimensional biology

Add to your list(s) Download to your calendar using vCal

  • UserSara Mostafavi, University of Washington, USA
  • ClockTuesday 15 March 2022, 13:15-14:15
  • HouseZoom.

If you have a question about this talk, please contact Mateja Jamnik.

Join us on Zoom

The growing availability of hundreds of different functional genomic assays across thousands of individuals presents an exciting opportunity to understand the inner workings of biological systems, and in turn identify molecular causes of disease. Toward this goal, machine learning (ML) provides a powerful toolkit to integrate diverse datasets, uncovering hidden structure that can reveal how different layers of biological systems relate to each other. However, to harness the power of ML for biology, we need to be able to tune it so to distinguish meaningful structure from those that arise because of artifact and noise. In this talk, I’ll present our recent approaches for leveraging heterogeneous data to guide discovery of meaningful biological structure. In particular, I will first describe our deep learning approach to combining a large compendium of epigenomic data, in order to learn the relationship between non-coding genomic sequence and regulatory activity across the immune system (Yoshida et al., Cell 2019; Maslova et al., PNAS 2020 ). I will then focus on the challenging task of understanding molecular causes of complex disease. Here, I will describe our techniques for revealing mechanistic insights into Alzheimer’s disease by combining large and heterogeneous gene expression datasets from the brain (Beebe-Wang et al., Nature Comm 2021). Together, these examples illustrate that ML is an important “assay” that can synthesize information across multiple experimental assays, in order to uncover hidden complexities of biology and yield system level hypotheses.

Bio: Sara Mostafavi is an Associate Professor at the Paul G. Allen School of Computer Science and Engineering at University of Washington (UW). Previously (until Sept 2020), she was a faculty member at the University of British Columbia (UBC), and Vector Institute (Toronto). She is a recipient of a Canada Research Chair in Computational Biology, a Canada CIFAR Chair in Artificial Intelligence (CIFAR-AI), and is a CIFAR fellow in the Child and Brain Development program. Before UBC , Sara did her postdoctoral fellowship with Daphne Koller at Stanford University and obtained a PhD in Computer Science from the University of Toronto in 2011.

This talk is part of the Artificial Intelligence Research Group Talks (Computer Laboratory) series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity