Work

ATOMICA

Year: 2026

Type: software

Abstract: ATOMICA v1.0.0 Learning Universal Representations of Intermolecular Interactions with ATOMICA This release includes: ATOMICA model training and inference code downstream benchmark and evaluation code documentation and usage examples Preprint is available at: https://www.biorxiv.org/content/10.1101/2025.04.02.646906 Model checkpoints are available at: https://huggingface.co/ada-f/ATOMICA Study datasets are available through Harvard Dataverse: https://doi.org/10.7910/DVN/4DUBJX The code is released under the MIT License.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.22269794

Authors: Fang, Ada, Desgagné, Michael, Zhang, Zaixi, Zhou, Andrew, Loscalzo, Joseph, Pentelute, Bradley, Zitnik, Marinka

Institutions: Massachusetts Institute of Technology, Harvard University

Language: English

Topic: Computational Drug Discovery Methods

Subfield: Computational Theory and Mathematics

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7207877768

Cites:

Cited by:

Related to: 10