Closed zdf1122 closed 1 year ago
hello! does the top sequence in the .fasta output file also have those Xs? Could it be that your PDB file has residue jumps in those places?
hello! does the top sequence in the .fasta output file also have those Xs? Could it be that your PDB file has residue jumps in those places? thank you very much !
hello! does the top sequence in the .fasta output file also have those Xs? Could it be that your PDB file has residue jumps in those places?
Sorry to bother you again. I wonder if whether the higher the score, global_score and seq_recovery, the better the result.
Lower score is better; the score (negative log probability) represents model's uncertainty about the predictions.
I've got a similar output. From what I can tell the original PDB did have residue jumps that were labeled as "-" in pyMol and appear to come out as "X" in the ProteinMPNN output.
If I wanted to put this sequence into something like AlphaFold or ESM, then would I simply remove the "X"s? Or is there more to it?
e.g. "SEQVXXXKIXM" becomes "SEQVKIM"
hello! I found that my result concludes some abnormal amino acid X, I know that it means undefined amino acid. But how did this happen? When I run the sample, the result is normal. (length of my protein is 1366 aa )![image](https://user-images.githubusercontent.com/81453850/196085019-c82c61f4-7cde-40be-9087-7421c7277a2a.png)