Re: [modeller_usage] Model evaluation

[Modeller Caretaker <modeller-care@ucsf.edu>]

vidhya ganesh wrote:
>     I have questions on "Model evaluation" step in "Basic modeling". As 
> far as i know, if it is a best model,the dope score profile shows region 
> of low energy.But, the plot in my model with template looks as if some 
> residue position have relatively high energy and in some showing of low 
> energy.How can i judge whether it is a good model or bad?

The only qualitative score is the total score - not the profile. The 
most reliable method in Modeller is to calculate the normalized DOPE 
score, which is a z-score. Good models are likely to have z-scores of 
-1, -2 or better.

	Ben Webb, Modeller Caretaker
--
modeller-care@ucsf.edu             http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage