May 2013 - modeller_usage

Error with the Mutate script
by Arjun Ray 08 May '13

08 May '13

Dear Sir, I am trying to mutate a residue THR61. Using the script provided on the website, I am getting following error: rayarjun$ python /usr/bin/mutate.py 1AV1.pdb 61 THR A > 1AV1.log File "/usr/bin/mutate.py", line 36 def refine(atmsel): ^ IndentationError: unindent does not match any outer indentation level Could you kindly help identifying the cause of it? Regards Arjun

2 3

change restraints relative weights
by Pietro Sormanni 03 May '13

03 May '13

dear Users, I am sorry for the naif question, but is there a way to change the relative importance of restraints? I am trying to add a secondary structure (both strand and sheet for H bonds) restraints to a model, in a region that do not align on the template structure (hence it shouldn't contain homology derived restraints). However only a few (up to 2) hydrogen bonds are formed rather than the whole beta sheet. In case you wish to delve further into this I attach at the end the piece of code that I use to add the restraints: Thank you very much for your help. Regards, Pietro from modeller import * from modeller.automodel import * # Load the automodel class # i tried also with automodel but there was no significant difference mdl = loopmodel(env, alnfile=already_done_alignment_filename, \ knowns = pdb_id, sequence = seq_id, \ loop_assess_methods=(assess.DOPE, assess.normalized_dope), \ assess_methods=(assess.DOPE, assess.normalized_dope) ) mdl.make_initial_model(aln) # Construct typical comparative modeling restraints mdl.mkhomcsr(selection(mdl), aln) mdl.restraints.add(secondary_structure.strand(mdl.residue_range('102:A', '108:A'))) # i know both of these regions are do align on anything (chain B is actually a 7 residues peptide I am adding, should be free to move anywhere). The residue numbering is read from initial model generated above mdl.restraints.add(secondary_structure.strand(mdl.residue_range('125:B:', '131:B'))) mdl.restraints.add(secondary_structure.sheet(mdl.atoms['N:102:A'], mdl.atoms['O:131:B'],sheet_h_bonds=-6)) mdl.restraints.make(selection(mdl), restraint_type='stereo', spline_on_site=False) mdl.restraints.condense() mdl.restraints.write(mdl.csrfile) # make the model env.edat.nonbonded_sel_atoms=1 # so that the selection interacts energetically with the unselected region. Actually I don't select anything manually, I added it just to be sure.. mdl.outputs = [] atmsel = mdl._check_select_atoms() mdl.multiple_models(atmsel) mdl.write_summary(mdl.outputs, 'models')

2 1

how to build the loop of protein model using modeller
by aixintiankong 03 May '13

03 May '13

Dear, The last email have some errors please see this one. In my system ,the loop is part of the active pocket of the protein. when the ligand is absent, the loop is disordered. if the ligand is present , the loop can transform into helix. In order to simulate the disordered loop transform into helix , i should build a model that the ligand and cofactor are in the disoreded loop site. In my system, there is a NAD+ cofactor in the active site and interact with the ligand.when the ligand and cofactor NAD+ coexist, the disordered loop can transform into helix. In the PDB(protein date base) i find the complex with the liand,NAD+ and the state of the loop is helix. i also find a structure of the protein without the liand, NAD+ and the state of the loop is disordered. The sequences similarity of the two structure is 100% and the RMSD between the two structures of the protein is 0.275 . i don't know how to build a model that the protein, NAD+ and ligand coexsit and the loop of protein is disordered. could give me some advice and tell me how to do it detailed? thank you very much.

2 1