BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:In silico identification of reactivity driven MIEs - Charles Gong\
 , University of Cambridge
DTSTART:20201028T150000Z
DTEND:20201028T153000Z
UID:TALK151207@talks.cam.ac.uk
CONTACT:Lisa Masters
DESCRIPTION:Thousands of new molecules are being created every year\; ensu
 ring that they are safe for human exposure is paramount. While classical t
 oxicology has been focused on animal models\, there is an increasing focus
  towards predictive computational models based on data\, with machine lear
 ning models showing great promise. However\, these models are viewed with 
 suspicion due to lack of interpretability. The conception of the adverse o
 utcome pathway (AOP) further provides a framework for understanding the me
 chanism of toxicity\, with the molecular initiating event (MIE) being of p
 articular interest to chemists. Identifying MIEs based on chemical structu
 re can improve our understanding of why molecules are toxic.\n\nInformatio
 n regarding MIEs in toxicological data is rare\, but a recent Seahorse ass
 ay ran by the US EPA contains multi-class labels (as opposed to binary lab
 els). This allowed construction of structural alerts specific to each mech
 anism\, but the small sample space limited the scope of this study.\n\nCom
 bined data for mitochondrial toxicity was extracted from ChEMBL\, ToxCast 
 and the EPA Seahorse assay. A neural network model was built from the data
 \, and the resulting classifier was shown to outperform commercially avail
 able toxicity screening software such as DEREK Nexus. Analysis of the mode
 l uncovered certain structural features most significant to these predicti
 ons\, and a visualization can be generated to aid chemists in decision-mak
 ing\, allowing some insight to be gleaned into the chemistry of MIEs relat
 ed to mitochondrial toxicity.\n
LOCATION:Zoom - Meeting ID: 924 3189 5042 Passcode: 306870
END:VEVENT
END:VCALENDAR