[gmx-users] Fwd: Selecting the temperature distribution
Justin Lemkul
jalemkul at vt.edu
Wed Apr 24 14:16:31 CEST 2013
On 4/23/13 9:43 PM, bharat gupta wrote:
> Dear Justin/Mark,
>
> I have asked this question previously in the forum, I got some reply from
> other members. It will be more useful if you can provide you expert
> comments on the same. I am planning to run REMD for a peptide (406 atoms )+
> solvent system (27639). The temperature range I selected is from 300 to
> 500. I want to select appropriate temp. for 56 replicas (as I have 56
> processors available). I used the t-remd calculator for temp. generation.
> It provided some 200 temp. values. Here are my questions :
> 1. Is it possible to select equally spaced temp. values from those values ??
> 2. I checked that reducing the no. of water mol. decreases the temp.
> values. What if I reduce the no. of water mol., will if affect the
> simulation ??
>
I haven't said anything because I agree with what Massimo has already told you.
If that is comforting in some way to know, then so be it, but I think it is
rather rude to suggest that you would rather someone else answer your question,
even after being given thorough and correct insight. This is a community forum
with many experts who have valuable insight to share.
-Justin
>
>
> On Tue, Apr 23, 2013 at 11:15 PM, massimo sandal <devicerandom at gmail.com>wrote:
>
>> In general, the smaller is your system, the less temperatures you will need
>> (and you'll have better performance).
>>
>> Notice however that implicit solvent, while surely a possibility worth
>> considering, is not usually considered to be very good -take care that if
>> you write a paper from implicit solvent results, reviewers might not be
>> happy. There is a chance that the results coming out of your simulation
>> might be nonsense. Discuss this choice with your supervisor and/or with
>> expert colleagues who know about limitations of implicit solvent. You need
>> to be able to justify your choice scientifically -for example testing it
>> with a known,similar system and observing that implicit solvent reproduces
>> the behaviour of that system in explicit solvent well.
>>
>> About reducing the box size, by all means try it, but always make sure it
>> is large enough to avoid that the periodic copies of your molecule see each
>> other. See
>>
>> http://ringo.ams.sunysb.edu/index.php/MD_Simulation:_Protein_in_Water#Box_Preparationand
>> be sure to understand it.
>>
>>
>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>
>>> Thanks a lot for your prompt responses. By using implicit solvent , I am
>>> getting on 9 temperature values. I think this should work , I will try it
>>> out. Also, i checked that when the no. of water molecules are reduced ,
>> the
>>> no. of temp. values are also reduced. If I reduce the no. of water
>>> molecules or reduce the size of box , will it help. As of now I am using
>>> octahedron box.
>>>
>>>
>>> On Tue, Apr 23, 2013 at 10:43 PM, massimo sandal <devicerandom at gmail.com
>>>> wrote:
>>>
>>>> It depends on what you want to do. Possible it is certainly possible,
>> but
>>>> you can't be guaranteed to observe the conformational changes you
>> desire
>>> to
>>>> observe. Again, it does not depend only on the REMD, but also on the
>>> length
>>>> of it. How long will it be? 10 ns? 100? 1000? 10.000? Plus, it also
>>> depends
>>>> on your system itself -and this you cannot know without trying.
>>>>
>>>> If you want to improve sampling beyond what standard REMD can do, to
>>>> exploit your computational resources at the best, you can look into
>> other
>>>> approaches like solute tempering (
>>>> http://www.pnas.org/content/102/39/13749.abstract ), or metadynamics (
>>>> http://en.wikipedia.org/wiki/Metadynamics ). However I advise you to
>>> study
>>>> *well* this kind of things, talk with experts in these techniques, and
>>>> remember that there is no guarantee any of them will bring the result
>> you
>>>> want. Good luck! :)
>>>>
>>>>
>>>>
>>>>
>>>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>>>
>>>>> So, my final question is whether is possible to do REMD for my
>> system,
>>>>> using the computational resource that I have.
>>>>>
>>>>>
>>>>> On Tue, Apr 23, 2013 at 10:06 PM, massimo sandal <
>>> devicerandom at gmail.com
>>>>>> wrote:
>>>>>
>>>>>> Who knows? It depends on the size of your peptide, on the energy
>>>>> landscape,
>>>>>> on how long is the run you plan to do. I would bet on "no",
>> however.
>>>>>>
>>>>>>
>>>>>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>>>>>
>>>>>>> But if I choose a smaller temperature range , would it be
>> possible
>>> to
>>>>>>> observe any folding event ??
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Apr 23, 2013 at 9:16 PM, massimo sandal <
>>>>> devicerandom at gmail.com
>>>>>>>> wrote:
>>>>>>>
>>>>>>>> Thanks, now it's clearer.
>>>>>>>>
>>>>>>>>> Now, how can I temp. from these, so that the replicas can
>>>> exchange
>>>>>> ...
>>>>>>>>
>>>>>>>> You can't, I would say. The system you have requires so many
>>>> replicas
>>>>>> to
>>>>>>>> exchange properly from the two temperature extremes you set up.
>>> As
>>>>> you
>>>>>>> have
>>>>>>>> seen, if you pick up temperatures in that range randomly, they
>>>> can't
>>>>>>>> exchange anymore. They are too far away.
>>>>>>>>
>>>>>>>> I would choose a smaller temperature range. There is little
>> else
>>>> you
>>>>>> can
>>>>>>>> do, I think.
>>>>>>>>
>>>>>>>>
>>>>>>>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>>>>>>>
>>>>>>>>> Sorry for that, I explain it again. Actually, I used the this
>>>> link
>>>>> to
>>>>>>>>> generate a temp. distribution. But I can do REMD for 56
>>> replicas
>>>>>> only,
>>>>>>>> as I
>>>>>>>>> have 56 processors available. The t-remd calculator provides
>>> 220
>>>>>>>>> temperature values :
>>>>>>>>>
>>>>>>>>> 300.00, 301.01, 302.02, 303.04, 304.06, 305.08, 306.11,
>> 307.14,
>>>>>> 308.17,
>>>>>>>>> 309.20, 310.24, 311.27, 312.32, 313.36, 314.40, 315.45,
>> 316.50,
>>>>>> 317.56,
>>>>>>>>> 318.61, 319.68, 320.74, 321.81, 322.88, 323.95, 325.02,
>> 326.10,
>>>>>> 327.18,
>>>>>>>>> 328.26, 329.35, 330.44, 331.53, 332.63, 333.72, 334.83,
>> 335.93,
>>>>>> 337.04,
>>>>>>>>> 338.15, 339.26, 340.37, 341.49, 342.61, 343.74, 344.87,
>> 346.00,
>>>>>> 347.13,
>>>>>>>>> 348.27, 349.41, 350.55, 351.69, 352.85, 354.00, 355.15,
>> 356.31,
>>>>>> 357.47,
>>>>>>>>> 358.63, 359.80, 360.97, 362.14, 363.32, 364.49, 365.68,
>> 366.86,
>>>>>> 368.05,
>>>>>>>>> 369.24, 370.44, 371.64, 372.84, 374.05, 375.26, 376.47,
>> 377.68,
>>>>>> 378.90,
>>>>>>>>> 380.12, 381.34, 382.57, 383.80, 385.03, 386.27, 387.51,
>> 388.75,
>>>>>> 390.00,
>>>>>>>>> 391.25, 392.51, 393.76, 395.02, 396.29, 397.56, 398.83,
>> 400.10,
>>>>>> 401.38,
>>>>>>>>> 402.68, 403.96, 405.25, 406.54, 407.84, 409.14, 410.44,
>> 411.74,
>>>>>> 413.05,
>>>>>>>>> 414.40, 415.71, 417.03, 418.36, 419.68, 421.01, 422.35,
>> 423.68,
>>>>>> 425.03,
>>>>>>>>> 426.37, 427.72, 429.07, 430.43, 431.79, 433.15, 434.52,
>> 435.89,
>>>>>> 437.26,
>>>>>>>>> 438.64, 440.02, 441.40, 442.79, 444.18, 445.58, 446.98,
>> 448.38,
>>>>>> 449.79,
>>>>>>>>> 451.20, 452.62, 454.03, 455.46, 456.88, 458.31, 459.75,
>> 461.18,
>>>>>> 462.63,
>>>>>>>>> 464.08, 465.53, 466.99, 468.45, 469.91, 471.38, 472.85,
>> 474.32,
>>>>>> 475.80,
>>>>>>>>> 477.28, 478.76, 480.25, 481.74, 483.24, 484.74, 486.25,
>> 487.76,
>>>>>> 489.27,
>>>>>>>>> 490.79, 492.31, 493.83, 495.36, 496.90, 498.43, 499.97,
>> 501.52,
>>>>>> 503.07,
>>>>>>>>> 504.63, 506.18, 507.78, 509.34, 510.91, 512.49, 514.07,
>> 515.65,
>>>>>> 517.24,
>>>>>>>>> 518.83, 520.43, 522.03, 523.64, 525.25, 526.86, 528.48,
>> 530.10,
>>>>>> 531.73,
>>>>>>>>> 533.36, 535.00, 536.63, 538.27, 539.92, 541.58, 543.23,
>> 544.90,
>>>>>> 546.56,
>>>>>>>>> 548.23, 549.90, 551.58, 553.27, 554.96, 556.65, 558.35,
>> 560.06,
>>>>>> 561.76,
>>>>>>>>> 563.48, 565.19, 566.92, 568.65, 570.38, 572.11, 573.85,
>> 575.60,
>>>>>> 577.47,
>>>>>>>>> 579.23, 580.99, 582.76, 584.52, 586.30, 588.08, 589.86,
>> 591.65,
>>>>>> 593.44,
>>>>>>>>> 595.24, 597.04, 598.85, 600.66
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Now, how can I temp. from these, so that the replicas can
>>>> exchange
>>>>>> ...
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Tue, Apr 23, 2013 at 9:04 PM, massimo sandal <
>>>>>>> devicerandom at gmail.com
>>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> I don't understand your question. If you got the
>> temperature
>>>>>>>>> distribution,
>>>>>>>>>> what else do you need?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>>>>>>>>>
>>>>>>>>>>> I have got the temperature distribution from the same
>> link,
>>>> but
>>>>>> how
>>>>>>>> to
>>>>>>>>>>> select evenly spaced temperatures for 56 replicas, I need
>>> to
>>>>> know
>>>>>>>> that
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On Tue, Apr 23, 2013 at 6:21 PM, massimo sandal <
>>>>>>>>> devicerandom at gmail.com
>>>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> Look here: http://folding.bmc.uu.se/remd/
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> 2013/4/23 bharat gupta <bharat.85.monu at gmail.com>
>>>>>>>>>>>>
>>>>>>>>>>>>> Dear gmx users,
>>>>>>>>>>>>>
>>>>>>>>>>>>> I am planning to run REMD for a peptide (406 atoms )+
>>>>> solvent
>>>>>>>>> system
>>>>>>>>>>>>> (27639). The temperature range I selected is from 300
>>> to
>>>>>> 500. I
>>>>>>>>> want
>>>>>>>>>> to
>>>>>>>>>>>>> select appropriate temp. for 56 replicas. I randomly
>>>> chose
>>>>>> some
>>>>>>>>> temp
>>>>>>>>>>>>> distribution and the exchange probabilities was 0.0.
>> I
>>>> know
>>>>>>> that
>>>>>>>> we
>>>>>>>>>> can
>>>>>>>>>>>> use
>>>>>>>>>>>>> the formula Ti=T0*ek*i, but what is the value for i
>>> and
>>>> K
>>>>>> here
>>>>>>>> ??
>>>>>>>>>>>>>
>>>>>>>>>>>>> --------
>>>>>>>>>>>>> BHARAT
>>>>>>>>>>>>> --
>>>>>>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>>>>>>> * Please search the archive at
>>>>>>>>>>>>>
>>> http://www.gromacs.org/Support/Mailing_Lists/Searchbefore
>>>>>>>>> posting!
>>>>>>>>>>>>> * Please don't post (un)subscribe requests to the
>> list.
>>>> Use
>>>>>> the
>>>>>>>>>>>>> www interface or send it to
>>>> gmx-users-request at gromacs.org.
>>>>>>>>>>>>> * Can't post? Read
>>>>>>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>>>>>>
>>>>>>>>>>>> --
>>>>>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>>>>>> * Please search the archive at
>>>>>>>>>>>>
>> http://www.gromacs.org/Support/Mailing_Lists/Searchbefore
>>>>>>>> posting!
>>>>>>>>>>>> * Please don't post (un)subscribe requests to the list.
>>> Use
>>>>> the
>>>>>>>>>>>> www interface or send it to
>>> gmx-users-request at gromacs.org.
>>>>>>>>>>>> * Can't post? Read
>>>>>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> --
>>>>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>>>>> * Please search the archive at
>>>>>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Searchbefore
>>>>>>> posting!
>>>>>>>>>>> * Please don't post (un)subscribe requests to the list.
>> Use
>>>> the
>>>>>>>>>>> www interface or send it to
>> gmx-users-request at gromacs.org.
>>>>>>>>>>> * Can't post? Read
>>>>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>>>> * Please search the archive at
>>>>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before
>>>>>> posting!
>>>>>>>>>> * Please don't post (un)subscribe requests to the list. Use
>>> the
>>>>>>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>>>>>>> * Can't post? Read
>>>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Bharat
>>>>>>>>> Ph.D. Candidate
>>>>>>>>> Biomolecular Engineering Laboratory
>>>>>>>>> Pusan National University
>>>>>>>>> South Korea
>>>>>>>>> Mobile no. - 010-5818-3680
>>>>>>>>> --
>>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>>> * Please search the archive at
>>>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before
>>>>> posting!
>>>>>>>>> * Please don't post (un)subscribe requests to the list. Use
>> the
>>>>>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>>>>>> * Can't post? Read
>>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>>
>>>>>>>> --
>>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>>> * Please search the archive at
>>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before
>>>> posting!
>>>>>>>> * Please don't post (un)subscribe requests to the list. Use the
>>>>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>>>>> * Can't post? Read
>> http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Bharat
>>>>>>> Ph.D. Candidate
>>>>>>> Biomolecular Engineering Laboratory
>>>>>>> Pusan National University
>>>>>>> South Korea
>>>>>>> Mobile no. - 010-5818-3680
>>>>>>> --
>>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>>> * Please search the archive at
>>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before
>>> posting!
>>>>>>> * Please don't post (un)subscribe requests to the list. Use the
>>>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>>>
>>>>>> --
>>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>>> * Please search the archive at
>>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before
>> posting!
>>>>>> * Please don't post (un)subscribe requests to the list. Use the
>>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>>
>>>>> --
>>>>> gmx-users mailing list gmx-users at gromacs.org
>>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>>> * Please search the archive at
>>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>>>> * Please don't post (un)subscribe requests to the list. Use the
>>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>>
>>>> --
>>>> gmx-users mailing list gmx-users at gromacs.org
>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>> * Please search the archive at
>>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>>> * Please don't post (un)subscribe requests to the list. Use the
>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>>
>>> --
>>> gmx-users mailing list gmx-users at gromacs.org
>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>> * Please don't post (un)subscribe requests to the list. Use the
>>> www interface or send it to gmx-users-request at gromacs.org.
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>
>> --
>> gmx-users mailing list gmx-users at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>> * Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-request at gromacs.org.
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>
>
>
--
========================================
Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
========================================
More information about the gromacs.org_gmx-users
mailing list