Error Parsing DNA PDB file

HADDOCK
1

Good afternoon. I am attempting to dock a protein with a DNA PDB construct created in PyMol using the fnab command from a unique sequence. I have additionally used the alter commands to combine the chains and renumber the nucleotides so that it is one single chain with cohesive numbering. I was previously using an online webserver at Sequence to Structure to create the PDB file of my DNA sequence, however it seems that this web page has become unresponsive some time in the last month. The resulting PDB file that PyMol has created using the fnab command opens fine for me, but I am getting a parsing error when attempting to upload to HADDOCK. See error below:

Error in PDB file.
Issue when parsing the PDB file at line 1 .
Your PDB contains an known residue PDBLine: HETATM 813 P DA A 1 4.474 -7.823-133.449 1.00 0.00 A P that should be using ATOM record, please check the HADDOCK library for a list of supported modified residues.

Am I correct in assuming that there is an issue with the HETATM designation in the file? I have seen other posts where the PDB-Tools web server is suggested as a means of correcting the PDB file, but am struggling to implement it. The PDB file below is the file in question. I don’t want to paste as text as there are a very large number of lines.
DNA-Sequence.pdb (166.9 KB)

If anyone could recommend a fix for this file, or an alternate means of generating a PDB file of my unique DNA sequence that is user friendly, I’d be very much appreciative.

2

Simply edit the file and replace HETATM by ATOM (and two spaces as the column alignment is important)

3

Thank you! You are incredibly helpful. Is there any way to prevent HADDOCK from docking to the ends of DNA? In biological system, DNA would be contiguous and therefor any models HADDOCK produces where the protein is docking to snipped ends of DNA are not useful for my purpose. Will selecting only nucleic acids in middle of DNA sequence be sufficient for this purpose?

4

Thank you! You are incredibly helpful. Is there any way to prevent HADDOCK from docking to the ends of DNA? In biological system, DNA would be contiguous and therefor any models HADDOCK produces where the protein is docking to snipped ends of DNA are not useful for my purpose. Will selecting only nucleic acids in middle of DNA sequence be sufficient for this purpose?

You can indeed only select residues in the central part (I would exclude a few bases on both sides in that case). But depending on your protein, there might still be solutions possible in which the protein contact the ends. You will have to try and see. And if needed extend your DNA on both side and exclude a longer fragment on both side when defining the DNA residues.

And if you don’t have any information about the binding site on the DNA, only define the DNA bases as passive (i.e. not as active).
But identifying the correct site is a hard problem…