SaProt_650M_AF2 is a 650 million parameter transformer model developed by Westlake University. It is trained on both protein sequences and structural information derived from AlphaFold2. The model supports fill-mask and other protein representation learning tasks and is intended to help researchers better understand and design proteins by incorporating structural context directly into the language model.
In the Foundation models & chat space, SaProt 650M AF2 takes a focused approach. It focuses on modeling protein sequences with awareness of 3D structure to improve predictions in bioinformatics and protein engineering. SaProt 650M AF2 is an open-source project aimed at bioinformatics researchers and computational biologists. The project is open source (MIT). SaProt 650M AF2 is available on the web and API.
Westlake University builds and maintains SaProt 650M AF2, and the product first shipped in 2020. The GitHub repository has been archived.
Latest indexed changes and source events
westlake-repl/SaProt_650M_AF2 verified by the PulseGate indexer
⚠ Archived
Other apps tracked under the same category.
