Machine Learning for Drug Discovery

Abstract

The current model of drug discovery -- the research involved in finding new medicines for the patients who need them -- is broken. It is expensive, high risk, is biased towards certain populations, has poor efficacy, and high societal costs. A new model is emerging that is data-driven and makes use of the latest developments in data science and machine learning. For example, tensor factorization techniques are being leveraged to identify new proteins to use as drug targets. Applying machine learning in a discovery application presents its own unique and exciting challenges, such as selecting the right data and model performance metrics. This talk will provide an overview of those challenges, along with new technology solutions. We will also discuss how to tackle the lack of diversity in biomedical data for machine learning in order to help close healthcare disparities.

Bio:

Alix Lacoste, PhD, is the Vice President of Data Science at BenevolentAI, an AI healthcare company that is transforming the way medicines are discovered, designed and developed. She leads a team that grows analytical, visual and machine learning tools for finding new ways to treat diseases. Previously at IBM Watson Health, Alix led AI-assisted research projects with academic and pharma partners. Alix holds a PhD in Molecular and Cellular Biology from Harvard University. She has presented her work in both biomedical and machine learning journals and conferences.

Meet in virtual Halligan Hall 102 (Sococo) or join Zoom Meeting:

https://tufts.zoom.us/j/98610939077

Meeting ID: 986 1093 9077

Passcode: See Colloquia email

One tap mobile

+13126266799,,98610939077#,,,,,,0#,,9008539197# US (Chicago)

+16465588656,,98610939077#,,,,,,0#,,9008539197# US (New York)

Dial by your location

+1 312 626 6799 US (Chicago)

+1 646 558 8656 US (New York)

+1 301 715 8592 US (Germantown)

+1 346 248 7799 US (Houston)

+1 669 900 6833 US (San Jose)

+1 253 215 8782 US (Tacoma)

Meeting ID: 986 1093 9077

Passcode: See Colloqiua email

Find your local number: https://tufts.zoom.us/u/adS4koag4r