MolPILE - large-scale, diverse dataset for molecular representation learning

Jakub Adamczyk, Jakub Poziemski, Franciszek Job, Mateusz Król, Maciej Makowski

Keywords: datasets & benchmarks, molecular representation learning, chemical foundation models, chemoinformatics

TL;DR: We present MolPILE, a large-scale, diverse and curated dataset for molecular representation learning and pretraining ML models

Abstract: The size, diversity, and quality of pretraining datasets critically determine the generalization ability of foundation models. Despite their growing importance in chemoinformatics, the effectiveness of molecular representation learning has been hindered by limitations in existing small molecule datasets. To address this gap, we present MolPILE, large-scale, diverse, and rigorously curated collection of 222 million compounds, constructed from 6 large-scale databases using an automated curation pipeline. We present a comprehensive analysis of current pretraining datasets, highlighting considerable shortcomings for training ML models, and demonstrate how retraining existing models on MolPILE yields improvements in generalization performance. This work provides a standardized resource for model training, addressing the pressing need for an ImageNet-like dataset in molecular chemistry.

Primary Area: datasets and benchmarks

Submission Number: 5902