search

YoshitakaMo / localcolabfold

ColabFold on your local PC
MIT License
527 stars 128 forks source link

Cyclic peptide Question: #232

Open aishiba0721 opened 2 months ago

aishiba0721 commented 2 months ago

Hello, I am using this project to predict the docking pose of cyclic peptides with proteins. I am using several examples from the original paper, such as 1SFI and 3AV9. The input format is in the form of a CSV file, and the sequence is formatted as described in the tutorial, with a colon separating the protein and cyclic peptide parts，such as “1SFI,IVGGYTCGANTVPYQVSLNXXSGYHFCGGSLINSQWVVSAAHCYKSGIQVRLXGEDNINVVEGNEQFISASKSIVHPSYNSNTLNNDIMLIKLKSAASLNSRVASISLPTXSCASXAGTQCLISGWGNTKSSGTSYPDVLKCLKAPILSDSSCKSAYPGQITSNMFCAGYLEGGKDSCQGDSGGPVVCSGKXXXXLQGIVSWGSXGCAQKNKPGVYTKVCNYVSWIKQTIASN:GRCTKSIPPICFPD”. However, I have noticed that the predicted peptides in the results are not forming cyclic peptides, but rather linear peptides. Thank you for any replies.

YoshitakaMo commented 2 months ago

AlphaFold2/ColabFold doesn't support a prediction of cyclic peptides currently...