Hi,
The error with shake (ntc=ntf=3) is not present when I make the CN- system
with gaussian charges. But, it gives me error when I use the CN- with two
extra points.
To add the two extra points, I had extended the extra_pts.f code by adding
a new frame (frtype(n)==4) for CN- ion and adding the way it calculated the
forces due to the EPs.
By doing so, I reproduced some results of a paper, and so the extra_pts.f
code worked fine (with shake on).
Now, I have added 2 CN- ions with extra points and 2 K+ ions, and SHAKE is
failing at the 2nd CN-. I also noticed that the usiz (bond length) of the
second CN- ions is read as 0.200 instead of 1.17 (0.200 is the C-EP
length).
My question is ::: is it possible that since I added an extra frame type
for CN-, SHAKE works fine for the 1st CN- and breaks for the second one?
The way the molecules are lined up in prmtop file is
C1 N1 P1 P2 C1 N1 P1 P2
So, is it possible that after working fine with the 1st C1 N1, SHAKE
encounters P1 P2 C1 N1 and messes up, because I added the P1 P2 by adding a
new frame type in extra_pts.f that shake.f does not recognise?
~Surma Talapatra
Graduate Student
Geva Group
University of Michigan
On Sat, 3 Sep 2011 08:50:13 -0400, case <case.biomaps.rutgers.edu> wrote:
> On Fri, Sep 02, 2011, tsurma.umich.edu wrote:
>>
>> I am running a simple MD equilibration with CN- and K+ ions in ~500 H2O.
>> When I have 1 CN-K+ in the system, it works perfectly. But, the moment I
>> have 2K+CN- in H2O the SHAKE (ntc=3, ntf=3) is giving me this error:
>>
>>
>> Coordinate resetting (SHAKE) cannot be accomplished,
>> deviation is too large
>> NITER, NIT, LL, I and J are : 0 0 2966 5 6
>
> This means that atoms 5 and 6 are the ones where shake fails. Finding
out
> what those are (C and N?) might help. Make sure you have done some
> minimization on the system; visualize where the K+ and CN- ions are; try
> a short run with ntc=ntf=2. (Do you have any real need for ntc=3?)
>
> ...dac
>
>
> _______________________________________________
> 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 09 2011 - 13:00:03 PDT
