Current browse context:
q-bio.QM
Change to browse by:
References & Citations
Quantitative Biology > Quantitative Methods
Title: 3D-Transformer: Molecular Representation with Transformer in 3D Space
(Submitted on 4 Oct 2021 (v1), revised 8 Oct 2021 (this version, v2), latest version 7 Jan 2023 (v7))
Abstract: Spatial structures in the 3D space are important to determine molecular properties. Recent papers use geometric deep learning to represent molecules and predict properties. These papers, however, are computationally expensive in capturing long-range dependencies of input atoms; and have not considered the non-uniformity of interatomic distances, thus failing to learn context-dependent representations at different scales. To deal with such issues, we introduce 3D-Transformer, a variant of the Transformer for molecular representations that incorporates 3D spatial information. 3D-Transformer operates on a fully-connected graph with direct connections between atoms. To cope with the non-uniformity of interatomic distances, we develop a multi-scale self-attention module that exploits local fine-grained patterns with increasing contextual scales. As molecules of different sizes rely on different kinds of spatial features, we design an adaptive position encoding module that adopts different position encoding methods for small and large molecules. Finally, to attain the molecular representation from atom embeddings, we propose an attentive farthest point sampling algorithm that selects a portion of atoms with the assistance of attention scores, overcoming handicaps of the virtual node and previous distance-dominant downsampling methods. We validate 3D-Transformer across three important scientific domains: quantum chemistry, material science, and proteomics. Our experiments show significant improvements over state-of-the-art models on the crystal property prediction task and the protein-ligand binding affinity prediction task, and show better or competitive performance in quantum chemistry molecular datasets. This work provides clear evidence that biochemical tasks can gain consistent benefits from 3D molecular representations and different tasks require different position encoding methods.
Submission history
From: Fang Wu [view email][v1] Mon, 4 Oct 2021 05:11:23 GMT (9942kb,D)
[v2] Fri, 8 Oct 2021 05:59:37 GMT (9943kb,D)
[v3] Thu, 23 Dec 2021 03:17:08 GMT (9942kb,D)
[v4] Sun, 15 May 2022 11:27:25 GMT (1145kb,D)
[v5] Wed, 18 May 2022 08:10:34 GMT (1110kb,D)
[v6] Tue, 20 Dec 2022 07:28:58 GMT (1112kb,D)
[v7] Sat, 7 Jan 2023 05:33:24 GMT (1112kb,D)
Link back to: arXiv, form interface, contact.