search

chao1224 / MoleculeSTM

Multi-modal Molecule Structure-text Model for Text-based Editing and Retrieval, Nat Mach Intell 2023 (https://www.nature.com/articles/s42256-023-00759-6)
https://chao1224.github.io/MoleculeSTM
Other
188 stars 18 forks source link

Processing ".cif" file. #17

Closed Sharpiless closed 6 months ago

Sharpiless commented 6 months ago

Hi, great work! I wonder if it is possible to input a ".cif" file containing the molecular structure and extract its latent features?

chao1224 commented 6 months ago

Hi @Sharpiless,

Thank you for your interest!

By the .cif file, I assume you refer to the 3D geometry. If so, then the current version of MoleculeSTM does not support it. However, you can replace the 2D molecule encoder (GIN) or 1D molecule encoder (MegaMolBART encoder) with the 3D encoder (we have another work Geom3D that can be easily adopted here).