BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Machine Learning for Toxicity Prediction Using Chemical Structures
 : Pillars for Success in the Real World  - Srijit Seal
DTSTART:20251215T123000Z
DTEND:20251215T130000Z
UID:TALK241324@talks.cam.ac.uk
CONTACT:Sam Nallaperuma-Herzberg
DESCRIPTION:Machine learning (ML) is increasingly valuable for predicting 
 molecular properties and toxicity in drug discovery. However\, toxicity-re
 lated end points have always been challenging to evaluate experimentally w
 ith respect to in vivo translation due to the required resources for human
  and animal studies\; this has impacted data availability in the field. ML
  can augment or even potentially replace traditional experimental processe
 s depending on the project phase and specific goals of the prediction. For
  instance\, models can be used to select promising compounds for on-target
  effects or to deselect those with undesirable characteristics (e.g.\, off
 -target or ineffective due to unfavorable pharmacokinetics). However\, rel
 iance on ML is not without risks\, due to biases stemming from nonrepresen
 tative training data\, incompatible choice of algorithm to represent the u
 nderlying data\, or poor model building and validation approaches. This mi
 ght lead to inaccurate predictions\, misinterpretation of the confidence i
 n ML predictions\, and ultimately suboptimal decision-making. Hence\, unde
 rstanding the predictive validity of ML models is of utmost importance to 
 enable faster drug development timelines while improving the quality of de
 cisions. This talk will emphasize the need to enhance the understanding an
 d application of machine learning models in drug discovery\, focusing on w
 ell-defined data sets for toxicity prediction based on small molecule stru
 ctures. We will focus on five crucial pillars for success with ML-driven m
 olecular property and toxicity prediction: (1) data set selection\, (2) st
 ructural representations\, (3) model algorithm\, (4) model validation\, an
 d (5) translation of predictions to decision-making. Understanding these k
 ey pillars will foster collaboration and coordination between ML researche
 rs and toxicologists\, which will help to advance drug discovery and devel
 opment.
LOCATION:SS03 Seminar Room\, Willam Gates building (Department of Computer
  Science and Technology)
END:VEVENT
END:VCALENDAR