- modeller_usage - salilab.org

ALIGN_CODE
by barwellj＠aston.ac.uk 07 Apr '08

07 Apr '08

Hello modellers, I have just started using the programme and have already ran into some difficulty. I am trying to do a homology model .The error message I get suggests that my alignment file does not match my pdb file. Although, all my files are in the same directory as modeller and both written in the same case. I'm not sure why it is not working? Before running the programme I edited the PDB file using DEEPVIEW to remove the antagonist from the structure. Also I used GENEDOC to produce the alignment and saved it in a PIR format. Not sure if this has something to do with my problem? attached is my PDB file (called a.pdb), my alignment file (a.pir) and my script file (called script.py) and the log file. Any help would really be appreciated

2 1

Re: [modeller_usage] modeller_usage Digest, Vol 7, Issue 19
by Vadim Patsalo 07 Apr '08

07 Apr '08

Hello, There seems to be a problem with your a.pir file. In particular, instead of structure:a.pdb:29::122::::: the file should read: structure:a.pdb:29:A:122:A:::: Note the difference, I am telling modeller to look for residue 29 of chain A in the pdb. otherwise, it looks like it looks for residue 29 of chain " ". Take a look at the manual to see how the PIR format works: http://www.salilab.org/modeller/manual/node421.html. Best, Vadim > Hello modellers, > > I have just started using the programme and have already ran into some > difficulty. I am trying to do a homology model .The error message I get > suggests that my alignment file does not match my pdb file. Although, all > my files are in the same directory as modeller and both written in the > same case. > > I'm not sure why it is not working? Before running the programme I edited > the PDB file using DEEPVIEW to remove the antagonist from the structure. > Also I used GENEDOC to produce the alignment and saved it in a PIR format. > Not sure if this has something to do with my problem? > > attached is my PDB file (called a.pdb), my alignment file (a.pir) and my > script file (called script.py) and the log file. > > Any help would really be appreciated

1 0

Help
by matziast＠med.uth.gr 06 Apr '08

06 Apr '08

Thank you for your immediate reply, I am trying to install modeller9v3 on Scientific Linux(2.6-18-53.1.4.3I5xen). It doesn't have any glibc, only glib=2.0 but I think that it has to be installed glib=2.2. And a last question, the scripts must be written in Python and they also have to be finished with .py? Thank you in advance and I am sorry for my so simple and boring questions.

2 1

Help
by matziast＠med.uth.gr 04 Apr '08

04 Apr '08

Hi, I am a new user of modeller and specifically modeller9v3. I am a student of Biochemistry and Biotechnology in Greece and I need this program for my diploma thesis. I read in the page of modeller that I must install python and I have to write my scripts in this language. On top of that, I must install glib>=2.2 and glibc>=2.3 but I have some problems in their installation. If I do not install them, the modeller9v3 will not work? (I am sorry for my errors in English). Thank you in advance. Maria.

2 1

inducing kink in a helix
by Marta Rossini 01 Apr '08

01 Apr '08

Hi all, I am using Modeller to model a structure in which there is a portion - not present in the template - that I want to force to be an alpha-helix. If I introduce the secondray structure restraints in the entire corresponding region of sequence, I obtain a straightened helix. In the sequence, however, there is a Proline in correspondence of which I want to induce a little distortion of the helix. If I force the secondary structure only before and after the proline, by excluding it, I obtain some very strange angles of an helix with respect to the other, that is a too pronounced bending. Is there in Modeller a way to induce a "controlled" distortion (that in my case corresponds to control the angle between the CO and NH groups of the proline!)? thanks _________________________________________________________________ Explore the seven wonders of the world http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE

2 1

FW: inducing kink in a helix
by Marta Rossini 01 Apr '08

01 Apr '08

> > Hi all, > I am using Modeller to model a structure in which there is a portion - not present in the template - that I want to force to be an alpha-helix. > If I introduce the secondray structure restraints in the entire corresponding region of sequence, I obtain a straightened helix. In the sequence, however, there is a Proline in correspondence of which I want to induce a little distortion of the helix. > If I force the secondary structure only before and after the proline, by excluding it, I obtain some very strange angles of an helix with respect to the other, that is a too pronounced bending. > Is there in Modeller a way to induce a "controlled" distortion (that in my case corresponds to control the angle between the CO and NH groups of the proline!)? > thanks > _________________________________________________________________ > Explore the seven wonders of the world > http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=QBRE _________________________________________________________________ News, entertainment and everything you care about at Live.com. Get it now! http://www.live.com/getstarted.aspx

1 0

MODELLER and USER FORMAT restraints
by Yingang Feng 29 Mar '08

29 Mar '08

Hi, In modeller9v3, the USER FORMAT of rsr file is not supported. However, it seems MODELLER FORMAT rsr file do not support the atom selection with name like 'CA:10' but have to use the number indice of atom. This makes the NMR restraint conversion difficult. How to deal with large number restraints easily? Best Wishes, Yingang Feng National Laboratory of Biomacromolecules Institue of Biophysics, Chinese Academy of Sciences Datun Road 15, Beijing 100101, P. R. China

2 1

symmetry and molecular dynamics
by JEAN WATERMEYER 28 Mar '08

28 Mar '08

Hi all We are trying to build a homotetramer model but having some difficulty maintaining symmetry. I've read the previous posts but maybe someone can help me further - I'm new to Modeller and to Python. I have imposed symmetry: s1 = selection(self.chains['A']).only_atom_types('CA') s2 = selection(self.chains['B']).only_atom_types('CA') s3 = selection(self.chains['C']).only_atom_types('CA') s4 = selection(self.chains['D']).only_atom_types('CA') self.restraints.symmetry.append(symmetry(s1, s2, 1.0)) self.restraints.symmetry.append(symmetry(s1, s3, 1.0)) self.restraints.symmetry.append(symmetry(s1, s4, 1.0)) but when I run automodel the molecular dynamics step pulls the monomers away from the template - so that although the subunits are identical, the symmetry is broken. We can think of a few ways to get around this: 1) do no molecular dynamics, only minimisation (maybe someone can tell me how?) 2) run molecular dynamics only on selected regions of the model so that the core remains fixed in place (again, how would I do this?) 3) learn Python and write restraints using symmetry matrices (if someone has implemented something like this before I'd love to see their code!) Looking forward to your input Jean Jean Watermeyer Division of Medical Biochemistry IIDMM University of Cape Town Health Sciences Anzio Rd Observatory 7925 South Africa Tel. 021 4066832

2 1

What are the considerations to model a cytosol protein using nuclear protein.
by Saravanan 20 Mar '08

20 Mar '08

I would like to model a protein which subcellular location is cytosol. I have found a better template which subcellular location is nuclear. My queries: 1) What are the considerations or constrainsts to be used to model a cytosol protein using nuclear protein. Is is correct to do like this. 2) Is it possible to specify constrainsts to model the environment based on tissue specific. Thank you in advance for any help Please don't print this e-mail unless you really need to. Save Trees.. P.Saravanan --------------------------------- Looking for last minute shopping deals? Find them fast with Yahoo! Search.

1 0

DOPE profile of target structure
by Hanudatta Atreya 19 Mar '08

19 Mar '08

Dear all, Should the residue-wise DOPE profile of the target modeled structure necessarily match that of the template? This is because in the tutorial online, the DOPE profile shown indeed follows very closely that of the template. In my case, the modeled structure has 1.0 Angstrom RMSD with the template. However, the residue wise DOPE profile do not match with that of template. Also, other structure quality scores as I checked with PROCHECK, MOLPROBITY etc. are poor. So does this indicate non-optimal modeling? Especially the Clash-scores are high. Should I carry out one more mimimization step myself to loosen the distortions? Thanks With regards Atreya --------------------------------- Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.

2 1