AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Molformer: Motif-Based Transformer on 3D Heterogeneous Molecular Graphs

Fang Wu, Dragomir Radev, Stan Z. Li

AAAI 2023 · Graph learning › Graph self-supervised learning · approach: Molecular graph pretraining · 61 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add motif and atom nodes as heterogeneous molecular graphs (ablation)Molformer · QM7MAE: 132.2 -> 43.2Improved
add motif and atom nodes as heterogeneous molecular graphs (ablation)Molformer · QM8MAE: 0.0205 -> 0.0095Improved
add motif and atom nodes as heterogeneous molecular graphs (ablation)Molformer · PDBbindRMSD: 1.925 -> 1.386Improved
use separate heterogeneous motif nodes instead of multi-level feature fusion (ablation)Molformer · QM7MAE: 89.7 -> 43.2Improved
use separate heterogeneous motif nodes instead of multi-level feature fusion (ablation)Molformer · QM8MAE: 0.0154 -> 0.0095Improved
use separate heterogeneous motif nodes instead of multi-level feature fusion (ablation)Molformer · PDBbindRMSD: 1.427 -> 1.386Improved
add HSA and HMG instead of passing 3D coordinates directly as atom input features (ablation)Molformer · QM7MAE: 132.2 -> 46.5Improved
add HSA and HMG instead of passing 3D coordinates directly as atom input features (ablation)Molformer · QM8MAE: 0.0205 -> 0.0097Improved
add AFPS molecular readout (ablation)Molformer · QM7MAE: 46.5 -> 43.2Improved
add AFPS molecular readout (ablation)Molformer · PDBbindRMSD: 1.441 -> 1.386Improved
add multi-scale distance-constrained attention (ablation)Molformer · QM7MAE: 50.1 -> 46.5Improved
add multi-scale distance-constrained attention (ablation)Molformer · QM8MAENo clear effect

Later papers that compared with this method

2 later accepted papers report MolFormer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy