Re: [modeller_usage] problem with the separate homo-trimer antiparallel beta sheet

[Modeller Caretaker <>]

On 9/18/14, 9:14 AM, Namita Dube wrote:
> I am really new in the field of building a model and I have a small
> peptide chain comprising of 16 residues and I am trying to build a
> homo-trimer anti-parallel beta sheet for all the 16 residue

Your numbering is a bit off here:

>          rsr.add(secondary_structure.strand(self.residue_range(0,15)))
>          rsr.add(secondary_structure.strand(self.residue_range(16,31)))

vs.

>          rsr.add(secondary_structure.sheet(at['N:1:A'],at['O:31:B'],
>                                        sheet_h_bonds=-16))

Note that in the first you use integers (0, 15, 16, 31) to select 
residues whereas in the second you use strings ('1', '31'). These are 
not the same; see http://salilab.org/modeller/9.14/manual/node326.html
In summary, integers start from zero, while strings are PDB numbers, 
which may not be continuous, or numbers at all, but usually start from 
1. So you're off by one in the second case - should be 'O:32:B'. 
Likewise the second sheet should be between N:17:B and O:48:C.

As was previously mentioned, you have no template for your B and C 
chains. So by default they're going to be floppy, extended chain 
structures. Modeller will do its best to make them beta strands, but 
you're asking quite a lot of it here.

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