- modeller_usage - salilab.org

Re: [modeller_usage] Modelling with SAXS data
by Edelmiro Moman 30 Oct '15

30 Oct '15

Yet another option would be converting your SAXS envelope to an (pseudo) electron density map, (rigid) docking your model into the map with something like SITUS and then refining your model to fit the (pseudo) electron density with something like NMFF. I have done it in the past with good results (it is very straightforward and you obtain a very nice trajectory/movie of the NM fitting). On 30/10/15 09:59, modeller_usage-request(a)salilab.org wrote: > [modeller_usage] Modelling with SAXS data -- Edelmiro Moman, PhD ProSciens, computer-aided drug design Caminho Bastida, 162 E-15405 Ferrol Galicia, Spain Phone : +352 621-511-423 E-mail : moman(a)prosciens.com http://prosciens.com

2 1

Modelling with SAXS data
by Ajasja Ljubetič 29 Oct '15

29 Oct '15

Hi all! First of all I would just like to say that I'm a very happy modeller user and that things have been running pretty smoothly:) We have obtained recently some SAXS data for a specific protein (that we hope is a monomer in solution). I have already built several models of this protein, but now I would like to refine it using SAXS data. As far as I know I have three options: - Use the SAXS modeller class <https://salilab.org/modeller/manual/node426.html> and hope for the best (I found an old thread <https://salilab.org/archives/modeller_usage/2011/msg00330.html> stating that it's no longer supported). - Use FoXS to screen models after I have generated them using modeller - Switch to IMP (I could only find one example <http://integrativemodeling.org/keystone2015/IMP_SAXS.pdf> of the IMP python interface) and add the restraints in IMP. Any comments and suggestions are welcome. in particular I'd like to know: how/if restraints be added in modeller based on SAXS data? And if it really is a very bad idea to use this unsupported functionality? Thank you & best regards, Ajasja Ljubetič, National Institute of Chemistry, Slovenia

2 1

Bug in documentation: automodel.get_model_filename() takes five arguments
by Ajasja Ljubetič 23 Oct '15

23 Oct '15

Hi! In the documentation <https://salilab.org/modeller/9.15/manual/node75.html> the signature for automodel.get_model_filename() is declared as get_model_filename(sequence, id1, id2, file_ext) But I got the following error: Traceback (most recent call last): File "d:/data/modeller/poly_modeller/ppmod/make_homology_model.py", line 97, in <module> a.make() File "c:\bin\Modeller\modlib\modeller\automodel\automodel.py", line 121, in make self.multiple_models(atmsel) File "c:\bin\Modeller\modlib\modeller\automodel\automodel.py", line 225, in multiple_models self.outputs.append(self.single_model(atmsel, num)) File "c:\bin\Modeller\modlib\modeller\automodel\automodel.py", line 308, in single_model self.pdb_ext) TypeError: get_model_filename() takes exactly 4 arguments (5 given) Of course one must include self as well, so the following works: def get_model_filename(self, sequence, id1, id2, file_ext): Perhaps in the documentation the self could be included as well? It's obvious in hindsight, but still took me 15 minutes to figure out. Or maybe a quick example could be added: How to move the pdb files into subfolder outdir class MyModel(automodel): def get_model_filename(self, sequence, id1, id2, file_ext): print self, sequence, id1, id2, file_ext return "outdir/test-prefix-{id1:02}-{id2:02}{file_ext}".format( id1=id1, id2=id2, file_ext=file_ext) PS: Is there some way to move all the generated files (including traces and sch and rsr files) into a subfolder? Scanning the sourcecode I see that redefining modfile.default could work. Thanks & All the best, Ajasja

2 2

Parallel job with modeller on a machine with altinstall python2.3
by ashika torikora 20 Oct '15

20 Oct '15

Hi, I know there have been a million different and redundant questions asked concerning parallel jobs using modeller. However, after thoroughly reading through most of them (and the manual pages concerning parallel jobs) I still cannot run a modeller parallel job. I understand that in order to run parallel modeller one has to have the python2.3 modules. I have installed Python2.3.6 as an alternative install on my RHEL machine (the default Python for my machine is 2.7.5). However even after installing Python2.3.6 I still get the same error (no module named socket, although Python2.3.6 has the socket module --python2.7.5 does not--). Adding a shebang with the python2.3 location in the beginning of the input changes nothing. Here is my input: from modeller import * from modeller.automodel import * from modeller.parallel import * # Use 4 CPUs in a parallel job on this machine j = job() j.append(local_slave()) j.append(local_slave()) j.append(local_slave()) j.append(local_slave()) log.verbose() env = environ() env.io.atom_files_directory = ['.', './template_struc'] # Create a new class based on 'loopmodel' so that we can redefine # select_loop_atoms class MyLoop(dopehr_loopmodel): # This routine picks the residues to be refined by loop modeling def select_loop_atoms(self): # Two residue ranges (both will be refined simultaneously) return selection(self.residue_range('278:', '284:')) a = MyLoop(env, alnfile = 'alignment.ali', # alignment filename knowns = '2cjf_0', # codes of the templates sequence = '2cjf', # code of the target loop_assess_methods=assess.DOPE) # assess each loop with DOPE a.starting_model= 1 # index of the first model a.ending_model = 1 # index of the last model a.loop.starting_model = 1 # First loop model a.loop.ending_model = 1000 # Last loop model a.loop.assess_methods=(assess.DOPE, assess.GA341, assess.normalized_dope) a.use_parallel_job(j) # Use the job for model building a.make() # do modeling and loop refinement And here is my error: Could not find platform independent libraries <prefix> Could not find platform dependent libraries <exec_prefix> Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>] 'import site' failed; use -v for traceback Traceback (most recent call last): File "backup.py", line 4, in ? from modeller.parallel import * File "/usr/lib/modeller9.15/modlib/modeller/parallel/__init__.py", line 1, in ? from modeller.parallel.job import job File "/usr/lib/modeller9.15/modlib/modeller/parallel/job.py", line 1, in ? import socket, random, os, select ImportError: No module named socket

3 2

Directly setting backbone dihedral angles (build an alpha helix)
by Ajasja Ljubetič 12 Oct '15

12 Oct '15

Dear all, I would like to build a perfect helix as an initial structure for some further manipulations. (Without a good initial structure the homology modelling part I apply later does not converge). Right now I'm using mdl.build_sequence(d.entire_sequence) to build the sequence, then I save it to a PDB, load it in VMD, set the backbone angles there (to get an alpha helix) save it to another PDB and load at back into modeller. So I guess my basic question is: *Is it possible to set backbone dihedral angles to a specific value? Or is there any other way to use mdl.build_sequence to get an alpha helix?* Thank you & best regards, Ajasja Ljubetič National Institute of Chemistry Slovenia

3 2

Conformational Modelling
by Parker de Waal 02 Oct '15

02 Oct '15

Hi Modeller users, I have an interesting case where I have two known structures of related proteins A and B. A is in the inactive conformation while B is in the active conformation. Is it possible to use modeller to predict the active conformation of A using structure B as the template? Further, despite being in different conformations, only a handful of residues are important for this conformational change (the rest remain stable and are distal from the active site). Is it possible to hold the correct parts of protein A in place and only perform modelling on residues within the active site (taken from protein B as template)? Best, Parker

2 1

Model Building
by Wong Li Zhe 02 Oct '15

02 Oct '15

Dear all, I am trying to build a protein model. Is it possible to build the model with the generated PDB files (with lowest DOPE score). There are 2 regions I which to build, where the first region consist of 1 template & the second region consists of 7 templates. I have built 100 models each for the 2 regions using single template approach and multiple template approach for the latter. >From the 100 models, I pick the best model with lowest DOPE score. With that, I have 1 model for first region & 1 model for second region. I would like to use the generated 2 models, combine them to build a final model. To my understanding, I can't proceed straight to the 'model building' command as it doesn't append PDB file. And I was trying to generate the alignment file for the 2 models but it won't work either. Could any of you enlighten me on this? Thank you & have a nice day! Best regards, Li Zhe [http://imu.edu.my/e-banner/sew-2015.jpg] Explore the endless possibilities in the field of medicine & health sciences. Register today at http://info.imu.edu.my/sew

2 1

Problem in oligomer modelling
by Soong Jiaxin 29 Sep '15

29 Sep '15

Hi all, I'm modelling a dodecamer. Somehow, there seems to be a problem for some of the monomers to connect to the other monomers. The oligomer template I used has interdimeric contacts. However, the way my model connects among the monomers is different from the template I used. The tail for the monomer in concern goes in and out of the beta strands of the neighbour monomer instead of connecting by touching the C-terminal region interface of the next monomer. Is there a way to fix this or is this normal for a protein model? I have attached a copy of my modelling script as well. Thank you for your help in advance. Regards, Vanessa

2 1

using DSS/BS crosslink information as restrain
by thereal sisterdot 08 Sep '15

08 Sep '15

Dear modeller users and "caretakers", i was using modeller to model a heterohexameric complex from mouse where the yeast orthologous structure is known. to evaluate the models i used the DOPE score and visualized a dozen available intersubunit DSS/BS crosslinks using Xlink Analyzer in chimera. for 2 neighbouring subunit 3 crosslinks indicate that the published heterohexameric complex appears to have a different arrangement in that part than the published one (crosslinked residues between subunits are around 60A apart) in order to obtain a model satisfying those crosslink restrains would one add the 3 crosslinks as below, or can you recommend a better approach: class MyModel(automodel): def special_restraints(self, aln): rsr = self.restraints at = self.atoms # Constrain the distance between alpha carbons in residues 611 chain D # and 224 chain B to be less than 30 angstroms: rsr.add(forms.upper_bound(group=physical.xy_distance, feature=features.distance(a['CA:611:D'], a['CA:224:B']), mean=30., stdev=0.1)) thanks a lot!!!! Maria

2 2

Import error
by GaoKyle 08 Sep '15

08 Sep '15

Hi,I'm running win7 64bit, and have both 32bit python3.4 and modeller9.15 installed. I'm getting this error when I tried to run a simple script:-------------------------------------Traceback (most recent call last): File "D:\学习\论文\Methylation\modeller\fasta2pir.py", line 1, in <module> from modeller import * File "C:\Program Files (x86)\Modeller9.15\modlib\modeller\__init__.py", line 93, in <module> import _modellerImportError: DLL load failed: 操作系统无法运行 %1。-------------------------------------------Please ignore the Chinese. I really have no idea of how to fix this problem.Please advise me.Thank you !

2 1