MolPILE -- large-scale, diverse dataset for molecular representation learning
This provides a standardized resource for chemoinformatics researchers, addressing the need for an ImageNet-like dataset in molecular chemistry, though it is incremental as it builds on existing data sources.
The authors tackled the problem of limited datasets hindering molecular representation learning by introducing MolPILE, a large-scale, diverse dataset of 222 million compounds, and showed that retraining existing models on it improves generalization performance.
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.