[gmx-users] "Justifying" 4fs production runs after 1fs equilibrations?

Tue Feb 11 21:12:40 CET 2014

Hi,

For your problem of assessing a homology model, I think that there are 
lots of other issues that I would be far more concerned about (e.g. 
force field accuracy, convergence of your simulations, etc.). To be 
fairly sure of anything I see in a simulation, I ideally would like to 
observe the same thing happening in a number of simulations using a 
number of different conditions, be that with different force fields, 
different starting structures, etc. I know that this is not always 
feasible but wherever possible I would try to do this. If you were to 
see different things with your different potential setups (4 fs with 
virtual sites and 2 fs constraining h-bonds or all-bonds), I would be 
more inclined to look for another issue (e.g. convergence) rather than 
these two different setup's.

Regarding the issue of persuading the postdoc of your suggested 
approach, I cannot suggest more than you have been doing. If he is not 
willing to accept the evidence that you are showing him, the only thing 
you can do is to try and find more. There are plenty of papers out there 
that use the 4 fs and virtual sites setup. For what's its worth, I have 
used both types of setups for simulations of membranes (which have some 
explicit hydrogens, so not just united-atom PC membranes) and have seen 
no significant differences in the membrane properties using the two 
different approaches.

Relating to your specific time step issue and publishing, as long as you 
can show that what you have done is sensible (such as through reading, 
following the advice of and referencing the paper and manual that you 
mentioned before), I cannot see how a manuscript would be rejected 
solely upon this matter. That might not be the case for some of the 
other things I have alluded to, such as being able to show convergence, 
reproducibility and so on.

Anyway, these are just some of my personal opinions. Hopefully they help 
a little with your situation. I am sure that there are others on the 
list with more experience of such matters than me who can also help.

Cheers

Tom

On 02/11/2014 03:45 PM, unitALX wrote:
> Yar, I understand. However, because of the differences in software
> experience (NAMD / GROMACS), presenting literature references is not as
> effective as I thought it would be, historical inertia and spirited
> argumentation is having more weight than I thought it would, and in between
> what is published for GROMACS and his best practice methods for NAMD, there
> is a gray area for which is effectively being biased towards NAMD style
> protocols (or what have you) and while you are correct in the long-term
> thermodynamic limit of publishing work, I have kinetic barrier to overcome
> in explaining something to someone who has seniority over me in my place of
> work and whose advice gives the de facto direction of the work I do. Hence
> why I am asking this more gray question of how to "justify" and "explain"
> across the software divide what I think are faster but valid choices for
> protocols. It is with considerable uneasiness that I see explanations being
> made in this gray area that are not consistent with the papers I am
> reading.
>
> To be more blunt about my situation, we are in a gray area on the choice of
> a homology model whose stability is being assessed by dynamics, and I am
> hearing people say things like "well its a gray area, we can just argue it
> this way, or that way, and then we'll submit the results to a journal, and
> if they don't agree, we'll just submit the same work to another journal". I
> face a higher burden of proof than the textbook approach. Please advise.
>
>
> On Tue, Feb 11, 2014 at 4:12 PM, TomPiggot [via GROMACS] <
> ml-node+s5086n5014470h92 at n6.nabble.com> wrote:
>
>> Hi,
>>
>> Personally I would also ask your colleague if he could also provide
>> evidence (e.g. published papers) to back up what he is saying to you (i.e.
>> that it is necessary to use a 1 fs timestep with an NVE ensemble to achieve
>> acceptable results). This way, you can make an informed decision based upon
>> the available evidence you can find, rather than simply upon one persons
>> view point. I would suggest the he will find it difficult to come up with
>> such evidence for your case of a straight forward MD simulation.
>>
>> You should also remember that you are the one that will need to be able to
>> defend (to an examiner or reviewer) what you have done and why. Simply
>> "because the postdoc told me it was the correct thing to do" will
>> (unfortunately) not be good enough!
>>
>> Cheers
>>
>> Tom
>> ________________________________________
>> From: [hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=0>[[hidden
>> email] <http://user/SendEmail.jtp?type=node&node=5014470&i=1>] on behalf
>> of Szilárd Páll [[hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=2>]
>>
>> Sent: 11 February 2014 13:33
>> To: Discussion list for GROMACS users
>> Subject: Re: [gmx-users] "Justifying" 4fs production runs after 1fs
>> equilibrations?
>>
>> On Tue, Feb 11, 2014 at 12:11 PM, unitALX <[hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=3>>
>> wrote:
>>
>>> Helllo all!
>>>
>>> In my general situation, I have a batch of homology models that I would
>> like
>>> to assess for stability by molecular dynamics.
>>>
>>> I am working with a postdoc in my lab who was extensive experience with
>>> NAMD, but does not use GROMACS. We have been developing protocol for
>> these
>>> MD simulations, but we had some different views on equilibration and
>>> production.
>>>
>>> Equlibration
>>> My friend is recommending that I do equilibration for 500ps @ 1fs with
>> no
>>> constraints. I have no real objection to that, but he made it seem like
>> the
>>> results might be quite different than an equilibration for 200ps @ 2fs
>> with
>>> LINCS all-bonds. Would it make a critical difference?
>>>
>>> Production
>>> I was excited by the performance possible with LINCS all-bonds, -vsite
>>> aromatics @ 4fs, but my friend looks at 4fs with disgust.
>> I would you mind asking your friend whether multiple time stepping
>> disgusts him too? :)
>>
>> --
>> Szilárd
>>
>>> I responded with
>>> arguments from the GROMACS 4 paper, and a paper called "Improving
>> efficiency
>>> of large time-scale molecular dynamics simulations of hydrogen-rich
>> systems"
>>> re: the relationship between the period of the fastest vibration and
>> largest
>>> allowable timestep, but he seems quite convinced that for publication,
>> only
>>> 1fs and possibly 2fs production runs (preferably without barostat and
>>> thermostat??) are acceptable. The suggestion to drop the barostat and
>>> thermostat for production follows the philosophy that if your
>> minimization
>>> and equilibration is properly done, then the system should be stable in
>>> production without pressure & temperature control. I'm having a hard
>> time
>>> finding better literature "justifications" for usage of LINCS all-bonds
>> @
>>> 2fs, and certainly for -vsites @ 4fs; P&T control seems quite standard
>> in
>>> publications using GROMACS.  Can you suggest any papers or calculations
>> I
>>> can do to show the validity of these options?
>>>
>>> --
>>> View this message in context:
>> http://gromacs.5086.x6.nabble.com/Justifying-4fs-production-runs-after-1fs-equilibrations-tp5014461.html
>>> Sent from the GROMACS Users Forum mailing list archive at Nabble.com.
>>> --
>>> Gromacs Users mailing list
>>>
>>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to [hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=4>.
>>
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to [hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=5>.
>>
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to [hidden email]<http://user/SendEmail.jtp?type=node&node=5014470&i=6>.
>>
>>
>>
>> ------------------------------
>>   If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://gromacs.5086.x6.nabble.com/Justifying-4fs-production-runs-after-1fs-equilibrations-tp5014461p5014470.html
>>   To unsubscribe from "Justifying" 4fs production runs after 1fs
>> equilibrations?, click here<http://gromacs.5086.x6.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=5014461&code=YWxlYy56YW5kZXJAZ21haWwuY29tfDUwMTQ0NjF8MTcxMDQ5NjUwOA==>
>> .
>> NAML<http://gromacs.5086.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
> --
> View this message in context: http://gromacs.5086.x6.nabble.com/Re-Justifying-4fs-production-runs-after-1fs-equilibrations-tp5014474.html
> Sent from the GROMACS Users Forum mailing list archive at Nabble.com.

-- 
Dr Thomas Piggot
University of Southampton, UK.