You are here

how to fix the ATP HETATM by using movemap or constraint file?

10 posts / 0 new
Last post
how to fix the ATP HETATM by using movemap or constraint file?

Dear all,

I'm doing docking peptides to kinase structure by using flexpepdock. In the meanwhile I need ATP in the kinase structure, therefore I generated a param file for ATP. But after the prepack mode, the ATP has been removed from the pocket to the somewhere in the structure. Here is my question:

1) how should I fix the ATP in the pocket without any movement? Movemap or constrain file?

2)I was told to use movemap to fix the ATP. But I don't know how's the rosetta numbering of the HETATM. 

for example, chain A is from 2-280, chain B is from 2-9, ATP(HETATM) is 381 in the chainA. In this case, which number is the ATP rosetta numbering?

Thank you!


Post Situation: 
Mon, 2017-11-20 06:43

I don't know that flexpepdock takes movemap files or constraint files.

I am assuming the PDB residues are ordered as chain A, chain B, then the ATP.  

If it takes movemap files, which seems less likely, then setting the second jump to false will probably keep the ATP in place relative to the first residue in the PDB.  (the first jump would be A->B, the second A->ATP).  This won't work great if it works at all.

If it takes constraint files - I think it does - try CoordinateConstraints on the ATP, or AtomPairConstraints between the ATP and the pocket (the latter will work much better).  AtomPairConstraints definitely, and CoordinateConstraints probably, take PDB numbering nowadays - so ATP is "381A" and a residue in chain A might be 50A, a residue in the peptide might be 5B.  

You can also run your protein through Rosetta (try score or score_jd2) with a renumbering flag - should be something like -renumber_pdb.

Mon, 2017-11-20 11:23

Thanks a lot! And I was wondering does the constraint file also work in prepack step? Because that's when the ATP is moved out of the pocket.  


Tue, 2017-11-21 04:58

Preacking only alters side-chain rotamers.  Also, a side-chain minimization step is added at the end. The ATP molecule should not move significantly unless it's clashing significantly. Have you cleaned the input pdb before running flexpepdock?  Make sure to remove water and other small molecules.  PyMol can be used for this purpose. As mentioned previously the order of different chains should be as chain A, chain B, then the ATP.  

Add the ATP molecule as HETATMs (here chain ID X)

HETATM    1  P1  ATP X   1      17.479  53.480 113.734  1.00 20.00           P  

If you want to freeze the ATP molecule during prepacking then I guess the way to do is to remove the lines declaring the CHI angles in the param file. Make sure the param file is created correctly. If it is then Rosetta will read it without throwing too many missing heavyatom: errors. 




Tue, 2017-11-21 06:40

Hi, thanks for your reply. I generated ATP.param file by


tutorial. And clean all the water. 

I've tried to maintain and remove the chi angles in the param file, but either way the ATP still cannot stay in the pocket. I attached my ATP.params file.

File attachments: 
Tue, 2017-11-21 08:31

Attach the clean pdb file too. 

Tue, 2017-11-21 08:59

Sure! So far I removed the water and other molecular except Mg2+ and ATP.

Wed, 2017-11-22 00:03

Fix the chain order. Peptide should be the second chain unless you specifically mention it by using -peptide_chain B flag. Also, replace the ATP related lines from the clean pdb with ATP.pdb content which is created during param file creation. Right now there are many missing heavyatom: errors. 

Wed, 2017-11-22 00:37

I've fixed the chain order. During I replacing the ATP lines from clean PDB with param file ATP.pdb. I noticed that the coordinate are different.  Should I replace the coordinate as well? 

Wed, 2017-11-22 02:10

I think I have figured it out this issue. The problem is the ATP naming from my pdb file is different from the ATP.pdb param file. That's why the ATP is moved after the flexpepdock. 

I replaced the atom name of ATP from my pdb with the ATP.pdb param file. And now the ATP stay in the pocket. 

Many thanks! 

Wed, 2017-11-22 06:21