Re: [modeller_usage] How to write a dope score for selected region of optimized atoms along with whole model?

[Modeller Caretaker <>]

On 7/20/16 2:01 PM, Mahesh Velusamy wrote:
> I thing modeler generate DOPE score only for selected region of optimized
> atoms but I want to write the dope score for optimized region of atoms
> along with whole structure. Is it possible to do the same?

Sure, it is really easy to write your own assessment method. I assume 
you're instantiating your MyModel class with something like

a = MyModel(..., assess_methods=assess.DOPE)

assess.DOPE is a simple Python function, defined in 
modlib/modeller/automodel/assess.py:

def DOPE(atmsel):
    """Returns the DOPE score of the given model."""
    return ('DOPE score', atmsel.assess_dope())

i.e. given the partial-model selection (atmsel) return the name of the 
score and its value. It's easy to write your own function that does 
something different, e.g.

def my_all_model_dope(atmsel):
    # get model
    m = atmsel.get_model()
    # select all atoms in model
    s = selection(m)
    # return all-model score
    return ('my score', s.assess_dope())

then instantiate your class with

a = MyModel(..., assess_methods=(assess.DOPE, my_all_model_dope))

to get both scores.

Alternatively, you can do whatever analysis you like (including 
calculating scores, of course) in automodel.user_after_single_model(), a 
method which gets called for each generated model:
https://salilab.org/modeller/9.17/manual/node74.html

	Ben Webb, Modeller Caretaker
--
             https://salilab.org/modeller/
Modeller mail list: https://salilab.org/mailman/listinfo/modeller_usage