SASA values different for full protein and sum of per residuewise SASA

GROMACS version: 5.0.7
GROMACS modification: No

Hi all,

I am analysing my protein-in-water type simulations for SASA (using gmx SASA). I wanted to compute the SASA values for each conformation. I was comparing for a conformation if the values of SASA for each residue if summed add equally when I calculate for the whole protein. Apparently there was a difference in both the values.
For example, for a 1 conformation, the total protein SASA = 44.271nm^2 whereas the sum of residuewise SASA is 117.828nm^2 Both calculations have same parameter/agruments

My concern is should I expect different values?

Regards,
Krushna

Hi,
Can you report the command line that you use to to calculate residuewise one?
Thank you
Alessandra

Hi Alessandra,
Command:
gmx sasa -surface ‘group “Protein_GLU_11”’ -output ‘“Hydrophobic” group “Protein_GLU_11” and charge {-0.2 to 0.2}; “Hydrophilic” group “Protein_GLU_11” and not charge {-0.2 to 0.2}; “Total” group “Protein_GLU_11”’ -f 2mxu-monomer.gro -s monomer-e.tpr -n index.ndx -o 2mxuProtein_GLU_11.xvg

Regards,
Krushan

The selection for -surface should always be the entire protein, otherwise you’re going to generate a surface for each residue as if the rest of the protein isn’t there, overestimating the actual SASA of that residue.

Hi Justin,

This clears my doubt for proteins. However, I also want to calculate contribution of each residue for a selected conformations and I tried computing SASA with and without surface, I get the same values which are overestimated.

For example:

Without surface:
Total = 2.768
Using Surface:
Total = 2.768
Hydrophobic = 0.603
Hydrophilic = 2.165

I alried computing the SASA for residues using Pymol and the SASA of whole protein. Apparently, sum total SASA of esch residue and the total sasa of whole protein does match.

Regards,
Krusha

Please provide the actual commands you used to obtain each output. I don’t understand what “without surface” means. You always need a surface group for the calculation.

Sorry for the delay.
I have tabulated detailed analysis as to show what my problem is. I took PDB: 1IYT which is a complete helical conformation of a peptide amyloid beta.
Tools used to calculate SASA (probe size 0.14nm:

  1. gmx sasa
  2. pymol
  3. mdTraj

As per your previous question.The following command is to compute the SASA for the whole protein

gmx sasa -surface 'group "Protein"' -output '"Hydrophobic" group "Protein" and charge {-0.2 to 0.2}; "Hydrophilic" group "Protein" and not charge {-0.2 to 0.2}; "Total" group "Protein"' -f 1iyt.gro -s 1iyt.tpr -n sasa.ndx -o FullProtein.xvg

For individual residues, I replace 'Protein' in the above command with the residues i.e.:
'Protein_ASP_1' as per ndx file and the total SASA I get is summed to get the total SASA for the whole protein.

When performing same task without -surface, the command is as follows:

printf “1\n” |gmx sasa -f 1iyt.gro -s 1iyt.tpr -n sasa.ndx -o nosurfaceFullProtein.xvg

Residuewise total SASA of protein (nm^2)
|mdTraj|________GMX_______|Pymol_|
|______|No Surface|Surface|______|
|42.15_|__117.49__|117.49_|47.64_|

Total SASA of Protein (nm^2)
|mdTraj|________GMX_______|Pymol_|
|______|No Surface|Surface|______|
|42.15_|__41.982__|41.982_|47.64_|

Regards,
Krushna

I don’t know what to expect when you don’t provide a selection for -surface but this has always been a key element of the calculation since you will get different results based on different selections. Choosing a single residue for a surface calculation is, as I said before, incorrect. The results you get are physically unrealistic.