You are here

clean_pdb.py of Rosetta VS cleanATOM of PyRosetta

3 posts / 0 new
Last post
clean_pdb.py of Rosetta VS cleanATOM of PyRosetta
#1

Dear friends,
I am dealing with a PDB file missing some residues. I think I should clean it by clean_pdb.py of Rosetta or cleanATOM of PyRosetta.

In Rosetta, the clean_pdb ignores the missing positions and renumber all the residues from 1 to the end regardless of the missing gap.

In PyRosetta, the missing gap can still be remained.

Can I ask can I get the result similar to cleanATOM by using Rosetta's clean_pdb.py? I think I should clean the PDB first, then do loop modeling for the missing residues. Is that correct?

Thank you very much.

Yours sincerely
Cheng

Post Situation: 
Wed, 2014-09-24 03:02
lanselibai

I don't think there's an option in the distributed clean_pdb script not to renumber the residues.

Now-a-days, though, Rosetta itself is pretty good at cleaning up PDBs if you pass the appropriate options. I myself tend to use Rosetta directly, rather than relying on an external script. Try something like the following commandline:

main/source/bin/score_jd2.linuxgccrelease -ignore_unrecognized_res -ignore_zero_occupancy false -out:pdb -s FILE_TO_CLEAN.pdb

This hopefully will result in a FILE_TO_CLEAN_0001.pdb file which represents what Rosetta thinks the PDB looks like.

Sometimes it doesn't quite work, but usually in those cases it's a quick fix with a search/replace or deleting lines with a text editor.

(Edit: fixed commandline)

Fri, 2014-09-26 15:41
rmoretti

(I do not know why but it seems that recently I always encountered error when I submit a comment in this forum. It turned out to be okay when I re-edit the comment)

Hi R Moretti,
It really works, so amazing!

A new file of file_0001.pdb (and score.sc) was created, 1) with full hydrogens, 2) without filling the missing residues to re-number, 3) with score comment in the end

So I delete the score in the end to get a clean PDB file.

Thank you very much.

(The first trial did generate any file though the prompt in the terminal seemed to be okay. So I tried again and got those files. But anyway, the problem has been solved.)

(Just to correct, it is score_jd2 instead of sdore_jd2) :)

Yours sincerely
Cheng

Wed, 2014-09-24 12:34
lanselibai