[gmx-users] calculation of standard deviation of SASA per residue

[Mark Abraham]

On 14/12/2011 4:01 PM, Harpreet Basra wrote:
> Hi
> I want to calculate the standard deviation of SASA of each residue of 
> a peptide. When using -or option for complete peptide in g_sas we get 
> residue vs sasa only.

I don't follow what result you are describing as unsuitable for you needs.

> So i tried by creating an index file for each residue separately 
> (residue1.ndx, residue2.ndx and so forth) and then by calculating by 
> using -o option i got time vs sasa for each residue, thinking that 
> this will give me average sasa per residue as well as its standard 
> deviation. But I see that the average value obtained by this method 
> for each residue is different from the one i got by using -or option 
> (and calculating for complete peptide). Why is it so? Doesnt -or 
> option gives average sasa over entire trajectory per residue? Can 
> anyone suggest any way to compute the standard deviation?
>

It's hard to comment when we don't know which groups you chose for which 
calculation. Using an index file to change *only* the output group and 
using g_sas -or should definitely work, but I don't see why custom 
groups should be needed at all.

Mark