Events @ Warwick Chemistry
New computational methods for charting chemical degradation space.
Abstract
Stability is among the most central chemical properties for which no predictive methods exist. Characterizing stability is also a practical bottleneck in many chemical and materials development applications. For example, the inability to predict stability has made it commonplace for computational screening projects to conclude with slow and manual expert curation of practical structures by organic chemists. In this talk, I’ll review the challenges associated with modeling stability and argue that the reaction prediction problem is the pivotal but solvable obstacle. The second half of the talk will focus on methods for accelerating the identification of degradants from high-throughput stability experiments. The challenge here is that degradation chemistry often yields complex product mixtures that are not found in databases. To address this, we are developing deductive machine learning models capable of ingesting multiple information sources to deduce degradants.
Microsoft Teams meeting
Join on your computer or mobile app
Click here to join the meetingLink opens in a new window
Meeting ID: 377 186 058 539
Passcode: JZTbdZ
Download TeamsLink opens in a new window | Join on the webLink opens in a new window