Improving Predictions of Molecular Properties with Graph Featurisation and Heterogeneous Ensemble Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Parker, Michael L., Mahmoud, Samar, Montefiore, Bailey, Öeren, Mario, Tandon, Himani, Wharrick, Charlotte, Segall, Matthew D.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915580054863872
author Parker, Michael L.
Mahmoud, Samar
Montefiore, Bailey
Öeren, Mario
Tandon, Himani
Wharrick, Charlotte
Segall, Matthew D.
author_facet Parker, Michael L.
Mahmoud, Samar
Montefiore, Bailey
Öeren, Mario
Tandon, Himani
Wharrick, Charlotte
Segall, Matthew D.
contents We explore a "best-of-both" approach to modelling molecular properties by combining learned molecular descriptors from a graph neural network (GNN) with general-purpose descriptors and a mixed ensemble of machine learning (ML) models. We introduce a MetaModel framework to aggregate predictions from a diverse set of leading ML models. We present a featurisation scheme for combining task-specific GNN-derived features with conventional molecular descriptors. We demonstrate that our framework outperforms the cutting-edge ChemProp model on all regression datasets tested and 6 of 9 classification datasets. We further show that including the GNN features derived from ChemProp boosts the ensemble model's performance on several datasets where it otherwise would have underperformed. We conclude that to achieve optimal performance across a wide set of problems, it is vital to combine general-purpose descriptors with task-specific learned features and use a diverse set of ML models to make the predictions.
format Preprint
id arxiv_https___arxiv_org_abs_2510_23428
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Improving Predictions of Molecular Properties with Graph Featurisation and Heterogeneous Ensemble Models
Parker, Michael L.
Mahmoud, Samar
Montefiore, Bailey
Öeren, Mario
Tandon, Himani
Wharrick, Charlotte
Segall, Matthew D.
Machine Learning
We explore a "best-of-both" approach to modelling molecular properties by combining learned molecular descriptors from a graph neural network (GNN) with general-purpose descriptors and a mixed ensemble of machine learning (ML) models. We introduce a MetaModel framework to aggregate predictions from a diverse set of leading ML models. We present a featurisation scheme for combining task-specific GNN-derived features with conventional molecular descriptors. We demonstrate that our framework outperforms the cutting-edge ChemProp model on all regression datasets tested and 6 of 9 classification datasets. We further show that including the GNN features derived from ChemProp boosts the ensemble model's performance on several datasets where it otherwise would have underperformed. We conclude that to achieve optimal performance across a wide set of problems, it is vital to combine general-purpose descriptors with task-specific learned features and use a diverse set of ML models to make the predictions.
title Improving Predictions of Molecular Properties with Graph Featurisation and Heterogeneous Ensemble Models
topic Machine Learning
url https://arxiv.org/abs/2510.23428