To Tristan:
Yes, I came to the same conclusion.  Again, *we* did not develop these parameters, but got them as part of the package available on Github.

To Zhijie:
The Chimera preset is a very different approach using a lookup table of amino acid values, which are shown in the table I linked at the bottom of my previous reply.  It does not use atomic values, and it is not a “potential”: instead the same amino acid type always just has the same value.  I only mentioned it because I thought you may have been familiar with it or used it in the past, and because it is also related to surface hydrophobicity coloring. I know it has been popular in Chimera, so I thought others might like the custom preset for ChimeraX.

My philosophy is generally that we explain in the documentation and links therein how the calculations are done and the values that were used and where they came from (e.g. you were able to look up these values and draw your own conclusions).   We try to provide tools that will be useful and not misleading, of course, but prefer to leave the judgment calls to the researcher since we cannot be up-to-date experts in every area of modeling.

I do understand your point of view since I’ve also worried about these atomic values, however, so I will put more thought into what could/should be done.

Elaine


> On Jul 12, 2019, at 1:47 PM, Tristan Croll <tic20@cam.ac.uk> wrote:
>
> To me, it looks like these values were calculated for the protonated forms of ASP and GLU. By convention, OD2 and OE2 are the atoms that receive the hydrogen.
>
> On 2019-07-12 21:27, Zhijie Li wrote:
>> Hi Elaine,
>> Thanks for the explanation. I understand the complications.  However
>> this function directly affects a button in the GUI that a lot of user
>> probably will press.  In Chimera it also controls a preset scenario?
>> For membrane proteins it might not be a problem but for soluble
>> proteins the outcome might be quite misleading. I think if the
>> underlying parameters are highly questionable it is probably better
>> not to make it so accessible to the users. Or maybe add a caveat note
>> in the documents?  Just my opinions.
>> Regards,
>> Zhijie
>>> On Jul 12, 2019, at 11:46 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
>>> Hi Zhijie,
>>> We’ve also noticed the same issues with these atomic potentials.  However, so far we have only implemented the MLP method based on pyMLP with these original values (developed by others), not wanting to get into the parametrization business ourselves.
>>> <http://rbvi.ucsf.edu/chimerax/docs/user/commands/mlp.html>

>>> <https://github.com/plantigrade/pyMLP>
>>> There may be different atomic potential sets for use with MLP that were already developed by others, but they would take time to add, so we haven’t thoroughly researched the issue.  If you have opinions or suggestions, let us know.
>>> As an alternative, I’ve provided a ChimeraX command script to color protein surfaces by the Kyte-Doolittle hydropathy values of amino acids (like the preset that is available in Chimera).
>>> <http://rbvi.ucsf.edu/chimerax/docs/user/kyte-doolittle_hydrophobicity.cxc>
>>> However, those Kyte-Doolittle values are simply a constant for each amino acid type, rather than a potential that varies in 3D space.  Still I find that the two approaches (Kyte-Doolittle values vs. MLP) often give a very similar picture of a transmembrane protein.
>>> You can add it as a custom preset in ChimeraX using the instructions here:
>>> <http://rbvi.ucsf.edu/chimerax/docs/user/preferences.html#startup>
>>> It could be easily modified to any amino acid hydrophobicity scales, or you could have multiple presets for the different scales.  Several scales are shown in this table from the Chimera User Guide:
>>> <http://plato.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/hydrophob.html>
>>> I hope this helps,
>>> Elaine
>>> -----
>>> Elaine C. Meng, Ph.D.
>>> UCSF Chimera(X) team
>>> Department of Pharmaceutical Chemistry
>>> University of California, San Francisco
>>>> On Jul 11, 2019, at 9:48 PM, Zhijie Li <zhijie.li@utoronto.ca> wrote:
>>>> Hello,
>>>> I noticed that the lipophilic atomic potentials of OD2 and OE2 are set to very positive values in mlp.py:
>>>> 'ASP': {'C': -0.54,
>>>>        'CA': 0.02,
>>>>        'CB': 0.45,
>>>>        'CG': 0.54,
>>>>        'N': -0.44,
>>>>        'O': -0.68,
>>>>        'OD1': -0.68,
>>>>        'OD2': 0.53}
>>>> 'GLU': {'C': -0.54,
>>>>        'CA': 0.02,
>>>>        'CB': 0.45,
>>>>        'CD': -0.54,
>>>>        'CG': 0.45,
>>>>        'N': -0.44,
>>>>        'O': -0.68,
>>>>        'OE1': -0.68,
>>>>        'OE2': 0.53},
>>>> As as result the hydrophobic surface representation always show hydrophobic patchs on the carboxylates, which look quite odd to me.   Is there a justification for this setting?  In addition, aren't the two oxygens in the carboxylate equal considerting that they are in resonance?
>>>> Zhijie
>> _______________________________________________
>> ChimeraX-users mailing list
>> ChimeraX-users@cgl.ucsf.edu
>> Manage subscription:
>> http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
>
>
> _______________________________________________
> ChimeraX-users mailing list
> ChimeraX-users@cgl.ucsf.edu
> Manage subscription:
> http://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
>