Re: [modeller_usage] build_profile problem

[Modeller Caretaker <>]

On 9/30/14, 7:53 AM, Vinicius Seus wrote:
> I have a problem with build_profile.py. It runs fine but the output file
> only appear one line about my target sequence. It doesn´t appear about
> the molecules from my database (seqdump.pir). There is an warning in the
> midlle of the process that say this:
>
> regress_656W> Number of bins for histogram too small - results maybe
> unreliable, nbins: 4
> regress_656W> Number of bins for histogram too small - results maybe
> unreliable, nbins: 4
> regress_656W> Number of bins for histogram too small - results maybe
> unreliable, nbins: 4

Your database is too small, so it can't calculate e-values. Normally 
build_profile is used to scan against a large database (PDB or UniProt).

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