Hi Brian,
Some updates.
In your file 1, the O-H bond is in the QM region. When qmshake=0,
sander will exclude the O-H bond from the hbond list. However, in your
file 2, the O-H bond cross the QM/MM boundary, so O is QM and H is MM.
In this case, sander will include the O-H bond in the hbond list no
matter qmshake=0 or not. This difference thus give a difference of 1
in number of degrees of freedom between two simulations.
I have a fix, but still under test. If you are interested, I can
commit it to the york directory in the repositery sometime soon.
Regarding the other problem (NPT simulation crashes) experienced in
your group. Was it a TI calculation as well? If not, could you give
more details about the problem please?
Best,
Qiantao
On Wed, Sep 21, 2011 at 4:44 PM, Brian Radak <radak004.umn.edu> wrote:
> Hi Qiantao,
>
> That's very interesting. I should mention that in the second parm7, there
> is a proton with zero charge that is not part of the quantum region.
> Possibly related?
>
> Another person in our group has been experiencing a problem with different
> symptoms (NVT simulations work but NpT simulations crash). The conclusion
> was also that the number of degrees of freedom was being differently
> calculated in each.
>
> Brian
>
> On Wed, Sep 21, 2011 at 3:43 PM, Qiantao Wang <qiantao.wang.gmail.com>wrote:
>
>> Hi Brian,
>>
>> I think the problem you experienced has something to do with the way
>> that sander counts the number of degrees of freedom, on which
>> temperature calculation depends.
>>
>> Taking your input file as an example, ME input generates a N.D.F. of
>> 9397, but MEO input gives a N.D.F. of 9396. I think this difference is
>> the reason of a constant 0.03~0.04K difference in TEMP in your output.
>>
>> Now one thing I have not fully understood is the way of counting the
>> number of degrees of freedom in the code, so cannot give you a
>> solution yet. Trying to get a bugfix soon, hopefully.
>>
>> Thanks,
>> Qiantao
>>
>> On Wed, Sep 21, 2011 at 9:40 AM, Brian Radak <radak004.umn.edu> wrote:
>> > Sorry, attachments were too big, here are dropbox links instead:
>> >
>> > http://dl.dropbox.com/u/40984968/ME_NVT_0.50_1.inp
>> > http://dl.dropbox.com/u/40984968/MEO_NVT_0.50_1.inp
>> > http://dl.dropbox.com/u/40984968/ME_NpT_0.50_1.crd
>> > http://dl.dropbox.com/u/40984968/ME.parm7
>> > http://dl.dropbox.com/u/40984968/MEO.parm7
>> >
>> > On Wed, Sep 21, 2011 at 8:54 AM, Brian Radak <radak004.umn.edu> wrote:
>> >
>> >> Jason - It is possible that this is an older version of Amber 11. It is
>> >> the development code within our group, although I thought it safe to
>> assume
>> >> that all of the proper git updates had been performed.
>> >>
>> >> Qiantao and Prof. Case - Attached are some input files (.inp, .parm7,
>> and
>> >> .rst7/.crd) for one window of a TI simulation for the pKa of methanol.
>> Note
>> >> that this is a QM - QM TI where the system charge changes at the
>> endpoints
>> >> leaving an uncharged dummy proton in one.
>> >>
>> >> Brian
>> >>
>> >>
>> >> On Wed, Sep 21, 2011 at 7:52 AM, David A Case <case.biomaps.rutgers.edu
>> >wrote:
>> >>
>> >>> On Tue, Sep 20, 2011, Brian Radak wrote:
>> >>>
>> >>> > I encountered the same behavior with 1000 steps of ntt=0 and ntt=3 (I
>> >>> can't
>> >>> > imagine that more steps would suddenly make it better). Judging by
>> the
>> >>> fact
>> >>> > that all of the other fields are identical, I'm going to agree with
>> your
>> >>> > speculation of a computing error or else a rounding error in the
>> >>> printing.
>> >>>
>> >>> Brian: I think we should see your inputs, because something funny is
>> going
>> >>> on.
>> >>> My experience is to see perfect matches, including with the
>> temperature.
>> >>> But
>> >>> there are all kinds of ways in which your inputs might be different
>> than
>> >>> mine.
>> >>>
>> >>> ...thx...dac
>> >>>
>> >>>
>> >>> _______________________________________________
>> >>> AMBER mailing list
>> >>> AMBER.ambermd.org
>> >>> http://lists.ambermd.org/mailman/listinfo/amber
>> >>>
>> >>
>> >>
>> >>
>> >> --
>> >> ================================ Current Address =======================
>> >> Brian Radak : BioMaPS
>> >> Institute for Quantitative Biology
>> >> PhD candidate - York Research Group : Rutgers, The State
>> >> University of New Jersey
>> >> University of Minnesota - Twin Cities : Wright-Rieman Hall
>> 101
>> >> Graduate Program in Chemical Physics : 610 Taylor Road,
>> >> Department of Chemistry : Piscataway, NJ
>> >> 08854-8066
>> >> radak004.umn.edu :
>> >> radakb.biomaps.rutgers.edu
>> >> ====================================================================
>> >> Sorry for the multiple e-mail addresses, just use the institute
>> appropriate
>> >> address.
>> >>
>> >>
>> >
>> >
>> > --
>> > ================================ Current Address =======================
>> > Brian Radak : BioMaPS
>> > Institute for Quantitative Biology
>> > PhD candidate - York Research Group : Rutgers, The State
>> > University of New Jersey
>> > University of Minnesota - Twin Cities : Wright-Rieman Hall
>> 101
>> > Graduate Program in Chemical Physics : 610 Taylor Road,
>> > Department of Chemistry : Piscataway, NJ
>> > 08854-8066
>> > radak004.umn.edu :
>> > radakb.biomaps.rutgers.edu
>> > ====================================================================
>> > Sorry for the multiple e-mail addresses, just use the institute
>> appropriate
>> > address.
>> > _______________________________________________
>> > AMBER mailing list
>> > AMBER.ambermd.org
>> > http://lists.ambermd.org/mailman/listinfo/amber
>> >
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>
>
>
> --
> ================================ Current Address =======================
> Brian Radak : BioMaPS
> Institute for Quantitative Biology
> PhD candidate - York Research Group : Rutgers, The State
> University of New Jersey
> University of Minnesota - Twin Cities : Wright-Rieman Hall 101
> Graduate Program in Chemical Physics : 610 Taylor Road,
> Department of Chemistry : Piscataway, NJ
> 08854-8066
> radak004.umn.edu :
> radakb.biomaps.rutgers.edu
> ====================================================================
> Sorry for the multiple e-mail addresses, just use the institute appropriate
> address.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 23 2011 - 12:30:04 PDT
