Skip to Content »

 Carl de Boer

Talk Title:

Learning the biochemical and evolutionary rules of gene expression with gene regulatory “Big Data”

Date/Time:
Wednesday, April 20th, 2022 @ 11:00am ~ 12:00pm (Pacific Time)

Location:
Virtually on Zoom:
Link: https://sfu.zoom.us/j/62763817981?pwd=cFJaVDN1a3U4NVJ5Uy9welZJazVWUT09
Meeting ID: 627 6381 7981
Password: 735871

Affiliation:
Assistant Professor, School of Biomedical Engineering

Bio:
Carl de Boer is an Assistant Professor in the School of Biomedical Engineering at the University of British Columbia. He did his PhD in the lab of Tim Hughes at the University of Toronto, and was a postdoctoral fellow in Aviv Regev’s lab at the Broad institute until the end of 2019, after which he moved to his current position at UBC.

Abstract:
In this talk, I will present two recent works from our lab based on high throughput gene expression measurements of random DNA. We measured the gene regulatory activities of over 100 million random promoter sequences in the model eukaryote yeast S. cerevisiae. We then used these data to train interpretable and deep learning models that relate DNA sequence to gene expression. The interpretable model captures the biochemical activities of the proteins (transcription factors) that regulate gene expression, and their interactions with chromatin from scratch, and revealed an unexpected prevalence of weak gene regulatory interactions. The deep learning (transformer) model enabled us to query the evolution of gene expression, and revealed an unexpectedly dynamic evolutionary space for promoter sequences while providing insight into the evolutionary past and future of promoter sequences. The “Big Data” we generated in these studies has proven to be exceptionally rich, and we hope it will serve as a standard dataset for machine learning for gene regulation.

——————————
Introductory Speaker:
Bradley R. Jones, Joy Lab, UBC

Talk Title:
Evolutionary modelling of the HIV persistent reservoir
——————————