VMD-L Mailing List

From: Ashar Malik (asharjm_at_gmail.com)
Date: Wed Dec 13 2017 - 01:34:08 CST

Hi Jrhau,

That sounds OK.

A couple of things to consider when working with selections is to avoid
duplication as I hinted at earlier and also mentioned by Peter. The two
selections must be mutually exclusive.

Secondly, not sure about the plugin you are using -- but its always
important to be consistent with your cutoffs.
Assuming you run a simulation and then compute energies -- the non-bonded
cutoff should be consistent between the two.

If you don't have a simulation and aim to run one --- remain consistent
with your cutoffs.
If you choose to not run a simulation - then ignore what I just said.

Best,
/A

On Wed, Dec 13, 2017 at 5:04 PM, jrhau lung <jrhaulung_at_gmail.com> wrote:

> Dear Ashar:
> My protein complex has 3 parts, and are subunit A, subunit B and a
> ligand (name XXX). The three parts are not covalent linked to each other
> and would like to calculate the nonbond energy between the three.
>
> Since Nondond energy only calculate the energy between selection 1 and
> selection 2.
> To calculate the nonbond energy between the three. It would need to
> collect the nonbond energy between them piece by piece and sum them up.
> Hope this is right. Please tell me if I were wrong. Thanks.
>
> sincerely,
>
> Jrhau
>
> 2017-12-13 11:41 GMT+08:00 Ashar Malik <asharjm_at_gmail.com>:
>
>> I am not sure what you want to achieve.
>>
>> chain A and resname XXX --- will only select resname XXX
>> chain A or resname XXX --- select all of chain A and all residues
>> resnames XXX (not sure if this would double your selection if resname XXX
>> is already included in the selection of chain A)
>>
>> is resname XXX covalantly linked to chain A?
>>
>> if you want to measure a quantity (say energy) between everything in
>> chain A and resname XXX and if resname XXX is also labelled as chain A then
>> you should use
>>
>> selection - 1:: chain A and not resname XXX
>> # This will select everything in chain A excluding
>> resname XXX
>> selection - 2:: resname XXX
>> # Will only select resname XXX
>>
>> You should only use the above if you are absolutely sure there is only
>> one residue labelled resname XXX cause if you have multiple you will still
>> get an answer - but I am not sure what I would make of it.
>>
>>
>>
>>
>>
>>
>> On Wed, Dec 13, 2017 at 4:04 PM, jrhau lung <jrhaulung_at_gmail.com> wrote:
>>
>>> Dear Ashar:
>>> Thanks for your hint. Yes, the complex contains only one ligand.
>>> And I realize that using "and" in selection need both selected parts
>>> are true.
>>>
>>> Instead of using "and", input "chain A or resname XXX" in selection
>>> 1 and "chain B" in selection 2 works.
>>> But I guess the nonbond energy between chain A and ligand XXX is still
>>> missing.
>>>
>>> sincerely
>>>
>>> Jrhau
>>>
>>> 2017-12-12 22:43 GMT+08:00 Ashar Malik <asharjm_at_gmail.com>:
>>>
>>>> Just to add to this, your original question reads:
>>>>
>>>> ... which contains two protein subunits (chain A and chain B. A and B
>>>> also interact each other) and *"**a"* ligand (resname 711 and located
>>>> in the middle of the complex and *contact with both chain A and B*)
>>>> ...
>>>>
>>>> This to me sounds like you have just one thing by the name of resname
>>>> of 711.
>>>>
>>>> However when you type in:
>>>>
>>>> "chain A and resname 711" in selection 1 and "chain B and resname 711"
>>>>
>>>> you are looking for two 711s one in each chain A and B. The error of no
>>>> atom in selection appears to suggest that while you have a 711 :: protein
>>>> ~ 711 = -40 kcal/mol there is just one of this (as understood from your
>>>> original question.
>>>>
>>>> set sel [atomselect top "resname 711"]
>>>> $sel get chain
>>>>
>>>> Only if the second line returns A and B should you expect an answer
>>>> from ::
>>>>
>>>> Input "chain A and resname 711" in selection 1 and "chain B and resname
>>>> 711" in selection 2
>>>>
>>>> On Wed, Dec 13, 2017 at 2:25 AM, Peter Freddolino <petefred_at_umich.edu>
>>>> wrote:
>>>>
>>>>> I'm sorry, but this doesn't seem to make sense:
>>>>> " If chain A and chain B do not both have residues with this name, input
>>>>> "chain A" in selection 1 and "chain B" in selection 2 should get same error
>>>>> message as
>>>>> when input "chain A and resname 711" in selection 1 and "chain B and
>>>>> resname 711". "
>>>>>
>>>>> Why would that be? If chain A and chain B each have many atoms, but
>>>>> none are resname 711 (which is a very odd residue name), then you would get
>>>>> normal results for the first case, but things would not work for the
>>>>> second. Perhaps you could double check what the chain names are for all of
>>>>> the residues in your system? Perhaps
>>>>> set sel [atomselect top all]
>>>>> lsort -unique [$sel get {chain resname resid}]
>>>>>
>>>>>
>>>>> On Tue, Dec 12, 2017 at 2:13 AM, jrhau lung <jrhaulung_at_gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Dear Giocomo:
>>>>>> Thanks for your advice.
>>>>>> The protein complex only contains 250 residues, including the
>>>>>> ligand 711. The "resname 711" should be correct for the analysis.
>>>>>> If chain A and chain B do not both have residues with this
>>>>>> name, input "chain A" in selection 1 and "chain B" in selection 2
>>>>>> should get same error message as
>>>>>> when input "chain A and resname 711" in selection 1 and "chain B and
>>>>>> resname 711".
>>>>>>
>>>>>> Continue struggling.
>>>>>>
>>>>>> sincerely,
>>>>>> Jrhau
>>>>>>
>>>>>> 2017-12-12 11:51 GMT+08:00 Giacomo Fiorin <giacomo.fiorin_at_gmail.com>:
>>>>>>
>>>>>>> Perhaps chain A and chain B do not both have residues with this
>>>>>>> name? Also, are you sure that you are not interchanging "resname" with
>>>>>>> "resid"?
>>>>>>>
>>>>>>> On Mon, Dec 11, 2017 at 9:52 PM, jrhau lung <jrhaulung_at_gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Dear João and VMD friends:
>>>>>>>>
>>>>>>>> I would like to calculate the nonbond energy of a protein
>>>>>>>> complex, which contains two protein subunits (chain A and chain B.
>>>>>>>> A and B also interact each other) and a ligand (resname 711 and
>>>>>>>> located in the middle of the complex and contact with both chain A and B)
>>>>>>>> using NAMDenergy plugin. What should I input in selection 1 and selection 2
>>>>>>>> to let result return in one calculation run. Thanks for any help.
>>>>>>>>
>>>>>>>> Input "protein" in selection 1 and "resname 711" in selection 2 got
>>>>>>>> ~ -40 kcal/mol
>>>>>>>> Input "chain A" in selection 1 and "chain B" in selection 2 got ~
>>>>>>>> -300 kcal/mol
>>>>>>>> Input "chain A and resname 711" in selection 1 and "chain B and
>>>>>>>> resname 711" in selection 2 returned a error message "both
>>>>>>>> selection must contain at least one atom"
>>>>>>>>
>>>>>>>>
>>>>>>>> sincerely,
>>>>>>>>
>>>>>>>> Jrhau
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Giacomo Fiorin
>>>>>>> Associate Professor of Research, Temple University, Philadelphia, PA
>>>>>>> Contractor, National Institutes of Health, Bethesda, MD
>>>>>>> http://goo.gl/Q3TBQU
>>>>>>> https://github.com/giacomofiorin
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Best,
>>>> /A
>>>>
>>>
>>>
>>
>>
>> --
>> Best,
>> /A
>>
>
> 

-- 
Best,
/A