Hello Dr Amy
I finally figured out that the problem
was with the cprestrt files produced
during CpH-REMD performed with my
AMBER22 installation. Running with
AMBER24, on the other hand, went well.
At your suggestion, I used visual
checks to rule out conformational
changes/salt bridges as the cause
of the apparently 'stuck' protonation
state.
Observations from AMBER22 are
as below.
Original cpin file (from cpinutil.py):
#############################
STATENE=0.000000,-2.776410,-6.483630,
PKA_CORR=7.1000,0.6000,0.0000,
#############################
(Two Histidine residues were titrated.)
Cprestrt file (obtained from the initial
(0-1 ns) REMD):
#############################
STATENE= 0.0000000000000000 , -2.7764099999999998 ,
PKA_CORR= 7.0999999999999996 , 0.59999999999999998 ,
#############################
Please note the missing third members
of the STATENE/PKA_CORR arrays.
Cprestrt file (obtained from 1-2 ns
REMD extended from the above):
#############################
STATENE= 0.0000000000000000 , -2.7764099999999998 ,
48*0.0000000000000000 ,
PKA_CORR= 7.0999999999999996 , 0.59999999999999998 ,
48*1000.0000000000000 ,
#############################
Please note the values of the third
members of STATENE/PKA_CORR
arrays.
I am very curious about both
the principle and the mechanism
behind those values.
This partly explains the observed
predominance of the third state i.e.,
HIE, in all the extended runs.
Corresponding observations from
AMBER24 are as below.
Original cpin file (from cpinutil.py):
#############################
STATENE=0.000000,-2.776410,-6.483630,
PKA_CORR=7.1000,0.6000,0.0000,
#############################
Cprestrt file (from the initial REMD):
#############################
STATENE= 0.0000000000000000 , -2.7764099999999998 ,
-6.4836299999999998 ,
PKA_CORR= 7.0999999999999996 , 0.59999999999999998 ,
0.0000000000000000 ,
#############################
Cprestrt file (from extended REMD):
#############################
STATENE= 0.0000000000000000 , -2.7764099999999998 ,
-6.4836299999999998 ,
PKA_CORR= 7.0999999999999996 , 0.59999999999999998 ,
0.0000000000000000 ,
#############################
Thanks and regards
Nitin Kulhar
On Tue, Aug 13, 2024 at 9:17 AM He, Amy <he.1768.buckeyemail.osu.edu> wrote:
> Hi Nitin,
>
>
>
> It sounds like the results in the cpout files (before concatenation and
> any other manipulation) do agree with the outcomes of cphstats.
> Therefore, this might not be a processing problem. Have you considered
> visualizing some snapshots (or maybe the endpoint structures) of the
> subsequent chunks? Do they look substantially different from the first
> segment of your simulation? Sometimes the preferred protonation state
> changes because of some conformational change (formation of a salt bridge
> for example). I feel that some visualization might be helpful for
> determining the cause of the issue.
>
>
>
> Best regards,
>
> Amy
>
>
>
> *From: *Nitin Kulhar <bo18resch11002.iith.ac.in>
> *Date: *Sunday, August 11, 2024 at 10:41 PM
> *To: *He, Amy <he.1768.buckeyemail.osu.edu>
> *Cc: *AMBER Mailing List <amber.ambermd.org>
> *Subject: *Re: [AMBER] cphstats processing of pH-REMD 'chunks'
>
> Hi Dr Amy Sorry for the late response. I can see it in the cpout files (of
> all the replicas in the first chunk of MD) that the protonation is toggling
> amongst all the possible states in accord with the pH value. However, this
> is not so for the
>
> Hi Dr Amy
>
>
>
> Sorry for the late response.
>
>
>
> I can see it in the cpout files (of all the replicas
>
> in the first chunk of MD) that the protonation
>
> is toggling amongst all the possible states in
>
> accord with the pH value.
>
>
>
> However, this is not so for the subsequent
>
> chunks of MD, wherein the deprotonated state
>
> has become seemingly 'fixed' regardless of
>
> the pH value, which is changing with the
>
> replica-exchange attempts (the successful ones).
>
>
>
> Therefore, I would deeply appreciate any help
>
> in pointing out the source of error. Please let
>
> me know if I need to share more information.
>
>
>
> Thanks and regards
>
> Nitin Kulhar
>
>
>
>
>
> On Wed, Aug 7, 2024 at 1:24 AM He, Amy <he.1768.buckeyemail.osu.edu>
> wrote:
>
> Hi Nitin,
>
>
>
> It seems like the fraction of protonation from the concatenated files is
> very low. To confirm if there’s a trend of decreased protonation:
>
>
>
> For the individual chunks of simulations, you can run cphstats for the
> original cpout files. If you know the length of each chunk, the total
> fraction of protonation can be calculated as a weighted average. No
> concatenation needed.
>
>
>
> If it doesn’t look like an expected trend, maybe it’s due to the
> concatenation. You can compared with the weighted average and see if you’re
> concatenating in the way you actually want.
>
>
>
> *From: *Nitin Kulhar <bo18resch11002.iith.ac.in>
> *Date: *Tuesday, August 6, 2024 at 9:37 AM
> *To: *He, Amy <he.1768.buckeyemail.osu.edu>, AMBER Mailing List <
> amber.ambermd.org>
> *Subject: *Re: [AMBER] cphstats processing of pH-REMD 'chunks'
>
> Hi Amy Thanks for the email. I apologise for any confusion caused by the
> term 'chunks'. To clarify, I was NOT referring to the various
> interval-based averages namely, cumulative, running, and chunk, that can be
> obtained from cphstats.
>
> Hi Amy
>
>
> Thanks for the email.
>
>
>
> I apologise for any confusion
>
> caused by the term 'chunks'.
>
> To clarify, I was NOT referring to
>
> the various interval-based
>
> averages namely, cumulative,
>
> running, and chunk, that can be
>
> obtained from cphstats.
>
>
>
> I meant 'chunk' in the context of
>
> Jason Swail's tutorial on pH-REMD:
>
> http://jswails.wikidot.com/ph-remd#toc10
> <https://urldefense.com/v3/__http:/jswails.wikidot.com/ph-remd*toc10__;Iw!!KGKeukY!0QPqWCdXnVdLe8rKHuhDQJUoHGhMynPyu-b40QIv63g2Os1e-DpK5_jKEPGP4bL9_uzoA0pBmpXdBHrl_s3gAVsi_QK48jQAXHEKQHKF$>
>
>
>
> To quote from the tutorial:
>
> If you have run 10 ns of simulation
>
> in 1 ns chunks, you will need to
>
> run this program 10 times!
>
>
>
> PFA two titration curves: one obtained from
>
> by processing just the first segment, the other
>
> obtained by processing concatenated cpouts
>
> from all segments.
>
>
>
> Thanks and regards
>
> Nitin Kulhar
>
>
>
> On Mon, Aug 5, 2024 at 4:35 AM He, Amy <he.1768.buckeyemail.osu.edu>
> wrote:
>
> Hi Nitin,
>
>
>
> Why do you think the results are wrong? Did it cumulate, or not cumulate,
> or not space in the way you want?
>
> The manual will give you a detailed difference with examples for the
> Cumulative, running, and “chunk” options of cphstats.
>
>
>
>
>
> *From: *Nitin Kulhar via AMBER <amber.ambermd.org>
> *Date: *Friday, August 2, 2024 at 9:38 PM
> *To: *AMBER Mailing List <amber.ambermd.org>
> *Subject: *[AMBER] cphstats processing of pH-REMD 'chunks'
>
>
>
> Dear all
>
> I performed pH-REMD of 8
> REPLICAS. 25 ns of REMD
> were run in 1 ns SEGMENTS
> (or chunks) as delineated in
> the tutorial by Jason Swails
> available at link <
> https://urldefense.com/v3/__http://jswails.wikidot.com/ph-remd__;!!KGKeukY!yNeLxboggV6vrz65BXBmJdJtWYdmVZcrdIOV1wBGc78GZvpfjF_-WCNsdRPGRjBJUhH7F7YBvR3VSKHDhd3UMV574E0$
> <https://urldefense.com/v3/__http:/jswails.wikidot.com/ph-remd__;!!KGKeukY!yNeLxboggV6vrz65BXBmJdJtWYdmVZcrdIOV1wBGc78GZvpfjF_-WCNsdRPGRjBJUhH7F7YBvR3VSKHDhd3UMV574E0$>
> >.
>
> *The problem:*
>
> The pKa values obtained by
> processing the cpout files from
> all SEGMENTS (or chunks)
> were questionable, nay, wrong.
> However, the pKa values obtained
> by processing the cpout file from
> the first SEGMENT alone were
> correct.
>
> *Salient details of the REMD:*
>
> The inputs for each SEGMENT
> were generated by the
> genremdinputs.py script.
>
> The cpin and starting coordinates
> for REPLICA 'i' of SEGMENT 'j'
> were respectively the cprestart
> and restart files generated by
> REPLICA 'i' during SEGMENT 'j-1'.
>
> While the seed (ig) value
> differed among REPLICAS
> of the same SEGMENT, it
> was the same for a given
> REPLICA in all SEGMENTS.
> (Is that genremdinputs.py's doing?)
>
> *Salient details of processing cpouts:*
>
> The cpout files from consecutive
> SEGMENTS were concatenated
> with the cat command. This was
> done for each REPLICA separately.
>
> The concatenated cpout files of ALL
> REPLICAS were sorted according to
> pH values by using cphstats program
> with the --fix-remd option.
>
> The pH-sorted cpout files from above
> were processed with cphstats to obtain
> pKa values and population data.
>
> Please help me spot the mistake
> in my logic or execution of the steps.
>
> Thanks and regards
> Nitin Kulhar
>
> --
>
>
> Disclaimer:- This footer text is to convey that this email is sent by one
> of the users of IITH. So, do not mark it as SPAM.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
>
> https://urldefense.com/v3/__http://lists.ambermd.org/mailman/listinfo/amber__;!!KGKeukY!yNeLxboggV6vrz65BXBmJdJtWYdmVZcrdIOV1wBGc78GZvpfjF_-WCNsdRPGRjBJUhH7F7YBvR3VSKHDhd3UQjpcOUQ$
> <https://urldefense.com/v3/__http:/lists.ambermd.org/mailman/listinfo/amber__;!!KGKeukY!yNeLxboggV6vrz65BXBmJdJtWYdmVZcrdIOV1wBGc78GZvpfjF_-WCNsdRPGRjBJUhH7F7YBvR3VSKHDhd3UQjpcOUQ$>
>
>
>
> Disclaimer:- This footer text is to convey that this email is sent by one
> of the users of IITH. So, do not mark it as SPAM.
>
>
>
> Disclaimer:- This footer text is to convey that this email is sent by one
> of the users of IITH. So, do not mark it as SPAM.
>
--
Disclaimer:- This footer text is to convey that this email is sent by one
of the users of IITH. So, do not mark it as SPAM.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 26 2024 - 20:30:02 PDT
