BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Interaction Profiles as a Universal Language for Generative Molecu
 lar Design with ShEPhERD–2
DTSTART:20261103T151500
DTEND:20261103T161500
DTSTAMP:20261028T032054Z
UID:e8fcf01b38ceb40d1df8dc515b5532e6fd68fbfbbdeed8ed15d9eec1
CATEGORIES:Conferences - Seminars
DESCRIPTION:Kento Abeywardane\, Kenji Walker\nThree-dimensional intermole
 cular interactions govern molecular recognition and are fundamental to the
  pharmacological activity of small-molecule drugs. We propose that an inte
 raction profile\, comprising shape\, electrostatics\, and directional phar
 macophores\, is a sufficient and transferable design specification for mol
 ecular design. We introduce ShEPhERD–2\, a 3D generative model that gene
 rates molecular structures conditioned on explicit interaction profiles. S
 hEPhERD–2 generates low-strain\, drug-like molecules more efficiently an
 d with greater fidelity to target interaction profiles than its predecesso
 r while enabling precise control over molecular design through pharmacopho
 re prioritization\, substructure constraints\, and composition of multiple
  interaction profiles. We demonstrate the application of a single model to
  bioisosteric fragment merging\, dual-target design\, selectivity engineer
 ing\, and modality hopping without task-specific retraining. These results
  establish interaction profiles as a chemotype-agnostic interface between 
 structure hypotheses and molecular design.\n\nKento Abeywardane is a 4th 
 year PhD student in the Coley Research Group at MIT studying chemical engi
 neering and computation. His thesis focuses on interaction aware 3D genera
 tive modeling for molecular design where he aims to imbue generative model
 s with biases for intermolecular interactions to improve learning and stee
 ring. He is interested in the applications of these generative approaches 
 to medicinal chemistry and catalyst design.\n\nKenji Walker is a 2nd year 
 PhD student in the Harvard-MIT Health Sciences and Technology program advi
 sed by Connor Coley and Angela Koehler. His interest is in improving small
  molecule generative models to enable therapeutics hypotheses that are oth
 erwise intractable.
LOCATION:https://epfl.zoom.us/j/68447908297?pwd=OU5JUGJUSUhZc0ZNYjQ2WENvYI
 NRdz09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR