search

dauparas / LigandMPNN

MIT License
205 stars 40 forks source link

Ligand Recognition #32

Open pikaha2410 opened 2 months ago

pikaha2410 commented 2 months ago

Hi,

I am trying to design a Lysine Binding Protein using LigandMPNN, but it seems that my ligand, which is lysine, is not being recognized. Could you please explain how LigandMPNN considers ligands? Is there a specific format or requirement for the ligand input that I should know? lbp_ligmpnn

lbp_atomrecord Thank you!

pikaha2410 commented 2 months ago

It worked when I changed the residue name to something other than an amino acid, like "LYY" instead of "LYS".

lbugnon commented 2 months ago

Had exactly same issue and worked. However if ligand atoms are in the following format they are not recognized either

image

kimdn commented 3 weeks ago

Had exactly same issue and worked. However if ligand atoms are in the following format they are not recognized either image

Me too, same issue, I wonder how to fix it.

I can't see your image. Never mind, your image is nothing but copy from https://github.com/dauparas/LigandMPNN/issues/32#issuecomment-2196139897 I can your image from my gmail (automatic email from github due to my reply here).

lzygitk7 commented 3 weeks ago

Had exactly same issue and worked. However if ligand atoms are in the following format they are not recognized either image

Me too, same issue, I wonder how to fix it.

I can't see your image. Never mind, your image is nothing but copy from #32 (comment) I can your image from my gmail (automatic email from github due to my reply here).

Thank you very much. I just figured out it is the issue of occupancy in PDB files. The occupancy should be 1 instead of 0. For anyone who encounters this issue, it might be a solution.

kimdn commented 3 weeks ago

Had exactly same issue and worked. However if ligand atoms are in the following format they are not recognized either image

Me too, same issue, I wonder how to fix it.

I can't see your image. Never mind, your image is nothing but copy from #32 (comment) I can your image from my gmail (automatic email from github due to my reply here).

Thank you very much. I just figured out it is the issue of occupancy in PDB files. The occupancy should be 1 instead of 0. For anyone that encounter this issue, might be a solution.

You are my lifesaver. Thank you @lzygitk7 ! I had the same issue (no ligand recognition). Fixing occupancy solved!