nvidia/esm2_uniref_pretraining_data

ESM-2 Uniref Pretraining Data Dataset Description: UniRef, or UniProt Reference Clusters, are databases of clustered protein sequences from the UniProt Knowledgebase (UniProtKB) that group similar sequences to reduce redundancy and make data easier to work with for biological research. It offers different levels of clustering (UniRef100, UniRef90, and UniRef50) based on sequence identity, with each cluster containing a representative sequence, a count of member… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/esm2_uniref_pretraining_data.