monsoon-nlp commited on
Commit
5b357af
1 Parent(s): 628228d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -2
README.md CHANGED
@@ -10,11 +10,12 @@ tags:
10
  # Protein Pairs and Similarity
11
 
12
  Selected protein similarities within training, test, and validation sets.
13
- Each protein gets two similarities selected at random and (if possible) proteins within the top and bottom quintiles for similarity.
14
 
15
  The protein is represented by its UniProt ID and its amino acid sequence (using IUPAC-IUB codes where each amino acid maps to a letter of the alphabet, see: https://en.wikipedia.org/wiki/FASTA_format ).
16
 
17
- The distance column is cosine distance (identical = 0) on UniProt / SwissProt 1,024-dimension embeddings, downloaded in March 2024.
 
18
  For models and training scripts optimizing for similarity, you should use `(1 - distance)`.
19
 
20
  Train / test / validation splits are from [khairi/uniprot-swissprot](https://huggingface.co/datasets/khairi/uniprot-swissprot/tree/39d2f9010ceb2599a12397b79babc902f8c440e6)
 
10
  # Protein Pairs and Similarity
11
 
12
  Selected protein similarities within training, test, and validation sets.
13
+ Each protein gets two similarities selected at random and (usually) proteins within the top and bottom quintiles for similarity.
14
 
15
  The protein is represented by its UniProt ID and its amino acid sequence (using IUPAC-IUB codes where each amino acid maps to a letter of the alphabet, see: https://en.wikipedia.org/wiki/FASTA_format ).
16
 
17
+ The distance column is cosine distance (identical = 0) on UniProt / SwissProt 1,024-dimension embeddings, downloaded in March 2024, based on the https://huggingface.co/Rostlab/prot_t5_xl_uniref50 model.
18
+
19
  For models and training scripts optimizing for similarity, you should use `(1 - distance)`.
20
 
21
  Train / test / validation splits are from [khairi/uniprot-swissprot](https://huggingface.co/datasets/khairi/uniprot-swissprot/tree/39d2f9010ceb2599a12397b79babc902f8c440e6)