Starr Hazard wrote:
Attempting to generalize to apply to a single ligand, I added "env.io.hetatm = True" to my python script and in the ali file _temp ERTF/. _target ERTY/. The HETATM entries for the ligand ARE in the template PDB file.
Note that Modeller will read _all_ of your HETATM entries once you turn this on, so if your template has multiple HETATMs, you must list them all in your alignment (or set the range in the alignment file header appropriately).
Modeller refuses to run with this and in the log file it writes that the alignment and the PDB file have different lengths EVEN THOUGH I have incremented the actual length listed in the ALI fileStructureX line in fields 3-6.
The alignment file header does not specify the length of the protein in the general case, but more usually the residue numbers of the starting and ending residues. So if your residue numbering is not sequential, this could be tripping you up. See http://salilab.org/modeller/9v1/manual/node380.html
(There is also a way to specify the real length of the sequence, by starting the final residue number with a '+'.)
If you still can't make sense of it, you'll need to post your full input files so that somebody can reproduce your problem.
In this particular case, the ligand in the template is covalently attached to a residue but I will want to model non-covalently bound ligands as well.
If the ligand is covalently attached, you probably don't want a chain break (/) in the sequence. Modeller also has no way to tell that covalent bonds exist (by default it'll only build generic distance restraints) so you may also want to add some restraints corresponding to the bonds and angles involved in that covalent bond.
Ben Webb, Modeller Caretaker
--
http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage