How should I evaluate protein surfaces in terms of hydrophobicity and surface charge properties of the surface.

Particularly I am looking to compare hydrophobic patches or surface charge between two proteins, from .pdb or Fasta sequences for example.

The goal is to study protein adsorption mechanism in the context of protein purification.

  • $\begingroup$ Have you used pymol? $\endgroup$
    – James
    Feb 15, 2017 at 10:01
  • $\begingroup$ Yes, but does pymol shows a numeric value for total surface hydrophobicity percentage for example? $\endgroup$ Feb 15, 2017 at 12:19
  • $\begingroup$ Yes. You can calculate solvent accessible surface as a fraction of the total surface. Although I'm not exactly sure what "total surface hydrophobicity percentage" is. $\endgroup$
    – James
    Feb 16, 2017 at 3:56
  • $\begingroup$ I've made some edits to your question to make it more about the biological concepts rather than "what tool is best". $\endgroup$
    – James
    Feb 16, 2017 at 4:06

2 Answers 2


Since you have the structures the best option, in my opinion, is Pymol.


Open Pymol. Load up your protein of choice. Download color_h.py which is a script from the University of Osaka that colours the residues according to Eisenberg's scale of hydrophobicity. Load this into Pymol by File->Run->PATH/TO/color_h.py. Then in Pymol run the following commands

  1. show surface

  2. color_h

enter image description here

Note that the script can be edited to use whatever scales you like. @mimat points you toward protscale and I would suggest the same. The Eisenberg scale is a rather old consensus scale. More up to date scales might be more appropriate in your case.

To start quantifying this, Pymol has tools for calculating solvent accessibility for example.

enter image description here

Additionally PDBSum and PDBe-PISA can estimate solvent accesible surface area.

Surface charge

Vacuum electrostatics

The quickest and easiest way to generate electrostatics is with the inbuilt vacuum electrostatics tools. In Pymol, select the action button (A)->generate->vacuum electrostatics->protein contact potential.

enter image description here


Since the Pymol inbuilt electrostatics make a lot of assumptions, for publication it is best to get a much more accurately quantified surface map. However, ABPS is a bit more involved. Read the wiki if this sounds more like what you need.


I think you are after protscale. It gives you a whole bunch of options to predict how your protein will be 'behave' for HPLC purification and hydrophobicity, buried residues, etc.

  • $\begingroup$ Protscale is a great tool for sequence analysis. Since we have the structures I think there are more reliable ways to look for biochemically interesting surface patches that take advantage of the 3D knowledge we have. $\endgroup$
    – James
    Feb 16, 2017 at 4:04
  • 1
    $\begingroup$ I agree with you completely, your answer will definitely give a more reliable and complete picture of hydrophobicity. Then again it also depends on the volume of proteins you need to analyse, since the poster was asking about purification you often want something quick and dirty for a lot of different targets. $\endgroup$
    – mimat
    Feb 16, 2017 at 12:28
  • $\begingroup$ I see. Good point! $\endgroup$
    – James
    Feb 17, 2017 at 3:27

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .