Improving reaction prediction through chemically aware transfer learning
Pretraining of NERF models on chemically related mechanisms significantly improves the performance compared to pretraining by larger, mechanistically dissimilar reaction datasets.
Authors
- Elizabeth H. Krenske (ORCID: https://orcid.org/0000-0003-1911-0501)
- Xiangliang Zhang (ORCID: https://orcid.org/0000-0002-3574-5665)
- Olaf Wiest (ORCID: https://orcid.org/0000-0001-9316-7720)
- Angus B. Keto (ORCID: https://orcid.org/0000-0002-0471-9958)
- Taicheng Guo (ORCID: https://orcid.org/0000-0001-7919-6912)
- Nils Gönnheimer
Institutions
- University of Notre Dame (US)
- The University of Queensland (AU)
Publication Details
- Journal
- Digital Discovery
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1039/d4dd00412d
- Citations
- 2
- Primary Topic
- Computational Drug Discovery Methods
- Type
- article
- Field-Weighted Citation Impact
- 3.01