Hello Liu Wei--
>
> In ensemble calculation, is it possible that the ensemble members
> have different initial structures rather than share the same one?
>
Certainly. Here's a simple example:
startStructures=["file1.pdb",
"file2.pdb"]
ensembleSize= len(startStructures)
from ensembleSimulation import EnsembleSimulation
esim=EnsembleSimulation('ens',ensembleSize)
esim.setAveType('sum')
ensIndex=esim.member().memberIndex()
protocol.loadPDB(startStructures[ensIndex])
Note that the structures in the different ensemble members need not
have the same PSF.
>
> By the way, is there any trick to make the ensemble members more
> diverse? In my case, they always don't differ enough as I expect.
>
You could try this at the beginning of calcOneStructure:
from monteCarlo import randomizeTorsions
randomizeTorsions(dyn,range=10)
This would randomize all torsion angles in an IVM called dyn by +/-10
degrees.
best regards--
Charles
_______________________________________________
Xplor-nih mailing list
[email protected]
https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
