Dear all, I found the solution to include disulfide bonds, while the later problem is how to combine multiple cpu caculation and including disulfide bridges together? they can be performed indenpendently but not together, could you please give me some suggestion?
model.py # Example of changing the default optmization schedule from modeller import * from modeller.automodel import * from modeller.parallel import *
j = job(modeller_path="/usr/lib/modeller9v7/bin/modslave.py") j.append(local_slave()) # 1 Processor j.append(local_slave()) # 2 Processors j.append(local_slave()) # 3 Processors j.append(local_slave()) # 4 Processors
class Mymodel(automodel): def special_patches(self, aln): self.patch(residue_type='DISU', residues=(self.residues['151'],self.residues['181'])) self.patch(residue_type='DISU', residues=(self.residues['422'],self.residues['448'])) self.patch(residue_type='DISU', residues=(self.residues['420'],self.residues['482']))
log.verbose() env = environ()
# Give less weight to all soft-sphere restraints: env.schedule_scale = physical.values(default=1.0, soft_sphere=0.7) env.io.atom_files_directory = ['.', '../atom_files']
a = Mymodel(env, alnfile='GBSS1-mult.ali',knowns=('2r4tA','3copA','3cx4A'), sequence='GBSS1')
a.starting_model =1 a.ending_model = 60
# Very thorough VTFM optimization: a.library_schedule = autosched.slow a.max_var_iterations = 300
# Thorough MD optimization: a.md_level = refine.very_slow
# Repeat the whole cycle 2 times and do not stop unless obj.func. > 1E6 a.repeat_optimization = 8 a.max_molpdf = 1e8
a.use_parallel_job(j) a.make()
error: 'module' object has no attribute 'Mymodel' from <Slave on localhost>
Check this thread: http://salilab.org/archives/modeller_usage/2009/msg00281.html
It means that you have to define your MyModel class OUTSIDE the main script and then import it.
Just open a new script, call it MyModel.py, paste the class MyModel inside. Don't forget to add the necessary imports at the top. Then in your main modelling script just add "from MyModel import MyModel" in the beginning. That should do the trick.
Regards,
João [...] Rodrigues @ http://stanford.edu/~joaor/
participants (2)
-
João Rodrigues
-
leuven