Probabilistic graph-to-graph model for reaction prediction and retrosynthetic route prediction
Lead Research Organisation:
UNIVERSITY OF CAMBRIDGE
Department Name: Engineering
Abstract
This project aims to develop machine learning methodologies that predict the outcomes of chemical reaction given reagent and reactant. We also aim solve the inverse problem of devising a sequence of reactions to make a target compound from simple starting materials.
People |
ORCID iD |
| Gabor Csanyi (Primary Supervisor) | |
| David Kovacs (Student) |
Publications
Kovács D
(2024)
Machine Learning Force Fields for Molecular Chemistry
Kovács DP
(2021)
Quantitative interpretation explains machine learning models for chemical reaction prediction and uncovers bias.
in Nature communications
Kovács DP
(2021)
Linear Atomic Cluster Expansion Force Fields for Organic Molecules: Beyond RMSE.
in Journal of chemical theory and computation
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/T517677/1 | 30/09/2019 | 29/09/2025 | |||
| 2276922 | Studentship | EP/T517677/1 | 30/09/2019 | 07/12/2023 | David Kovacs |