Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Parser.py has some issues #36

Closed
jaideepm009 opened this issue Jun 14, 2024 · 9 comments
Closed

Parser.py has some issues #36

jaideepm009 opened this issue Jun 14, 2024 · 9 comments

Comments

@jaideepm009
Copy link

jaideepm009 commented Jun 14, 2024


Screenshot 2024-06-14 at 12 30 15 PM

I believe "r" should be replaced with "current_resname" in line 452 of parser.py file under DockQ directory?

@jaideepm009
Copy link
Author

But if i make the above change I get "Buffer has wrong number of dimensions (expected 2, got 1)". So I am not sure whats the issue.

@clami66
Copy link
Collaborator

clami66 commented Jun 14, 2024

Hi, could you attach the files you're trying to score?

@jaideepm009
Copy link
Author

jaideepm009 commented Jun 16, 2024

The files I am trying to process are 1cbv.pdb and the model file produced by rosettafold2 architecture. I am not able to attach them here :(

@clami66
Copy link
Collaborator

clami66 commented Jun 16, 2024

Try this:

  1. Update the repository with git pull and rerun pip install
  2. Since 1cbv contains a very small DNA molecule, it's likely that the short sequence is messing with the automatic mapping. So try to provide the mapping between native and model chains manually. It will be something like: --mapping ABC:DHL or similar (depends in what order Rosetta writes out the chains)

@jaideepm009
Copy link
Author

I was providing chain map manually. I did git pull and reinstalled DockQ. I am still getting error in buffer dimension. Here are the screenshots of my code.
Screenshot 2024-06-17 at 9 11 49 AM
Screenshot 2024-06-17 at 9 12 39 AM

@clami66
Copy link
Collaborator

clami66 commented Jun 17, 2024

I cannot help you further without the model unfortunately. You should be able to make a zip, then drag and drop it in a comment box here on the issue

@jaideepm009
Copy link
Author

Archive.zip

In model(model_00_pred.pdb): A corresponds to antigen chain, B corresponds to heavy chain, C corresponds to light chain. In native("1cbv.pdb): D = antigen, H = heavy, L = light. I need to calculate DockQ for antigen to heavy and antigen to light chains

@clami66
Copy link
Collaborator

clami66 commented Jun 17, 2024

Hi @jaideepm009

Chain A in your model is a peptide with unknown amino acids:

ATOM      1  N   UNK A   1     -14.695   1.700   8.805  1.00 33.72
ATOM      2  CA  UNK A   1     -16.016   2.020   8.281  1.00 33.72
ATOM      3  C   UNK A   1     -17.109   1.521   9.211  1.00 33.72
ATOM      4  O   UNK A   1     -17.328   0.316   9.344  1.00 33.72
ATOM      5  CB  UNK A   1     -16.594   1.112   7.207  1.00 33.72
ATOM      6  H   UNK A   1     -14.203   0.899   8.438  1.00 33.72
ATOM      7  HA  UNK A   1     -16.094   3.104   8.211  1.00 33.72
ATOM      8 1HB  UNK A   1     -17.641   1.395   7.016  1.00 33.72
ATOM      9 2HB  UNK A   1     -16.016   1.213   6.289  1.00 33.72
ATOM     10 3HB  UNK A   1     -16.562   0.078   7.547  1.00 33.72

But the native chain D is a NA (DNA?) molecule. So there is no way for DockQ to compare the two. The type of molecules need to be consistent across the two pdb files or it doesn't make sense to compare them

@jaideepm009
Copy link
Author

Yes we just realised that yesterday as well! Thanks for taking the time out looking into the issue.

@clami66 clami66 closed this as completed Jun 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants