Closed srilekha1993 closed 3 months ago
the generated sequence fasta is default prepared_for_esm.fasta
not pdbbind_sequences.fasta
in README.md, so should run this command instead:
python scripts/extract.py esm2_t33_650M_UR50D ${DIFFDOCK_BASE}/data/prepared_for_esm.fasta embeddings_output --repr_layers 33 --include per_tok --truncation_seq_length 4096
Notice this command should be run at the facebook esm repository, not the DIFFDOCK repository.
README.md needs some correction. @gcorso
Hi, This URL i have referred for https://zenodo.org/records/10656052 for downloading BindingMOAD_2020_ab_processed dataset. But there is no moad_sequences_new folder inside BindingMOAD_2020_ab_processed. Can anyone please provide the correct link for the above dataset?