OpenProteinSet: Training data for structural biology at scale
This provides a critical resource for researchers in structural biology and machine learning, enabling progress in protein-related tasks, though it is incremental as it builds on existing data generation methods.
The authors tackled the lack of accessible large-scale multiple sequence alignments (MSAs) for protein research by introducing OpenProteinSet, an open-source dataset of over 16 million MSAs with associated structural data, which they used to successfully retrain AlphaFold2.
Multiple sequence alignments (MSAs) of proteins encode rich biological information and have been workhorses in bioinformatic methods for tasks like protein design and protein structure prediction for decades. Recent breakthroughs like AlphaFold2 that use transformers to attend directly over large quantities of raw MSAs have reaffirmed their importance. Generation of MSAs is highly computationally intensive, however, and no datasets comparable to those used to train AlphaFold2 have been made available to the research community, hindering progress in machine learning for proteins. To remedy this problem, we introduce OpenProteinSet, an open-source corpus of more than 16 million MSAs, associated structural homologs from the Protein Data Bank, and AlphaFold2 protein structure predictions. We have previously demonstrated the utility of OpenProteinSet by successfully retraining AlphaFold2 on it. We expect OpenProteinSet to be broadly useful as training and validation data for 1) diverse tasks focused on protein structure, function, and design and 2) large-scale multimodal machine learning research.