GPS++: Reviving the Art of Message Passing for Molecular Property Prediction

Dominic Masters, Josef Dean, Kerstin Klaser, Zhiyi Li, Sam Maddrell-Mander +7 more
2/6/2023
cs.LG

No official code found. Best implementation: graphcore/ogb-lsc-pcqm4mv2 (78 stars, MIT, last commit 2024-07-25).

Abstract

We present GPS++, a hybrid Message Passing Neural Network / Graph Transformer model for molecular property prediction. Our model integrates a well-tuned local message passing component and biased global attention with other key ideas from prior literature to achieve state-of-the-art results on large-scale molecular dataset PCQM4Mv2. Through a thorough ablation study we highlight the impact of individual components and find that nearly all of the model's performance can be maintained without any use of global self-attention, showing that message passing is still a competitive approach for 3D molecular property prediction despite the recent dominance of graph transformers. We also find that our approach is significantly more accurate than prior art when 3D positional information is not available.

View on arXivView PDF

Code Implementations(1)

graphcore/ogb-lsc-pcqm4mv2Community100%
MIT

The OGB-LSC is the Large Scale Competition by Open Graph Benchmark to help accelerate research into machine learning on graph structured data

7815C++, CythonNov 16, 20222 years agoMIT

Similar Papers

Optimal message passing for molecular prediction is simple, attentive and spatial

90% match2025

Hierarchical Grammar-Induced Geometry for Data-Efficient Molecular Property Prediction

86% match1 repo262026

XIMP: Cross Graph Inter-Message Passing for Molecular Property Prediction

85% match2026

Cross-Modal Learning for Chemistry Property Prediction: Large Language Models Meet Graph Machine Learning

85% match2024

Investigating Graph Neural Networks and Classical Feature-Extraction Techniques in Activity-Cliff and Molecular Property Prediction

85% match2024

Adaptive Substructure-Aware Expert Model for Molecular Property Prediction

85% match2025

Cite this paper

@article{masters2023gps,
  title  = {GPS++: Reviving the Art of Message Passing for Molecular Property Prediction},
  author = {Dominic Masters and Josef Dean and Kerstin Klaser and Zhiyi Li and Sam Maddrell-Mander and Adam Sanders and Hatem Helal and Deniz Beker and Andrew Fitzgibbon and Shenyang Huang and Ladislav Rampášek and Dominique Beaini},
  year   = {2023},
  eprint = {2302.02947},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.02947v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.02947.svg)](https://codeofpaper.com/paper/2302.02947)

Discussion