Re: Protein-ligand affinity

From: Thomas Evangelidis (tevang3_at_gmail.com)
Date: Tue Oct 15 2013 - 14:43:39 CDT

You neglect completely the entropic effect of ligand binding:
DeltaG=DeltaH-T*DeltaS
Judging from your interaction energy DeltaH ~ 0 so in order to have a
favourable Free Energy of Binding (DeltaG<0) then DeltaS>0, which means the
removal of waters and/or ions from the binding cavity by the ligand
increases the entropy of the system.

As Aron & Kenno suggested, MM/PBSA calculations are most suitable for your
case. FEP and TI are cumbersome and used more for ligand optimization.

On 15 October 2013 21:43, James Starlight <jmsstarlight_at_gmail.com> wrote:

> Aron,
>
> thanks for reference.
>
> By the way how the Namd-energy plugin could be effectively used ? In case
> of small tutorial I choce two selections- all atoms of my ligands and 5
> polar residues surrounded ligand within the protein cavity. Than I
> calculate electrostatic energy using cutoff and switch values from the conf
> file of MD production run. In the resulting graph I've obtain Zero-line of
> the energy during of all simulation run (so if this is POTENTIAL energy it
> indicate that no interactions between ligand and selected residues have
> been occurred during MD that is incorect). What I did wrong?
>
> James
>
>
> 2013/10/15 Aron Broom <broomsday_at_gmail.com>
>
>> there was in fact one recently posted on the NAMD website and mentioned
>> on the mailing list.
>>
>>
>> On Tue, Oct 15, 2013 at 1:42 PM, James Starlight <jmsstarlight_at_gmail.com>wrote:
>>
>>> So in that case just ussage of VMD's NAMD energy module would give wrong
>>> results of the G-bonding estimation.
>>> Also I've found that Free energy perturbation and Thermodynamic
>>> integration could be usefull for affinity measurements. I've seen tutorials
>>> for implementation of such methods for gromacs and charm but is there any
>>> for NAMD?
>>>
>>> James
>>>
>>>
>>> 2013/10/15 Aron Broom <broomsday_at_gmail.com>
>>>
>>>> As an FYI, that energy is going to be border-line useless without
>>>> having an unbound control to subtract.
>>>>
>>>> You may want to read up on the literature surrounding this, it is a
>>>> very large topic, with many nice reviews, and the problem itself is fairly
>>>> nuanced. What you seem to be talking about is an MM/PBSA style of
>>>> calculation, so I'd start by looking into that.
>>>>
>>>>
>>>> On Mon, Oct 14, 2013 at 2:20 PM, James Starlight <
>>>> jmsstarlight_at_gmail.com> wrote:
>>>>
>>>>> Dear NAMD users,
>>>>>
>>>>>
>>>>> I have trajectory of the protein complexed with the ligand (burried in
>>>>> the protrein interiour from the start of simulation). Now I want to find
>>>>> out how I can calculate affinity of my ligand to the protein. The one
>>>>> possible way of do such task is the interaction energy estimation of my
>>>>> complex (based on the occurence of the non-covalent contacts between both
>>>>> partners during production run). Could someone provide me with some
>>>>> tutorial or explain me how I could perform such analysis with the VMD tools
>>>>> ?
>>>>>
>>>>> Thanks for help,
>>>>>
>>>>> James
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Aron Broom M.Sc
>>>> PhD Student
>>>> Department of Chemistry
>>>> University of Waterloo
>>>>
>>>
>>>
>>
>>
>> --
>> Aron Broom M.Sc
>> PhD Student
>> Department of Chemistry
>> University of Waterloo
>>
>
>

-- 
======================================================================
Thomas Evangelidis
PhD student
University of Athens
Faculty of Pharmacy
Department of Pharmaceutical Chemistry
Panepistimioupoli-Zografou
157 71 Athens
GREECE
email: tevang_at_pharm.uoa.gr
          tevang3_at_gmail.com
website: https://sites.google.com/site/thomasevangelidishomepage/

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:50 CST