En, quickly run through a few papers about chemical shift prediction. Now I
think what I found in my case is pretty normal. SHIFTX, SHIFTS and PROSHIFT,
all have slightly larger deviation of shift values of N15 from experimental
values. Figure 11 of "Biopolymers, Vol. 65, 408–423 (2002)" shows a similar
shift of the fitted diagonal line. Whether this is related to the hydrogen
bond between water and N-H group remains a question to me though. Gotta
read the paper carefully tomorrow.
Sorry to bomb your mail boxes. Have a good night sleep. :)
Mingfeng
On 5/4/07, Mingfeng Yang <mfyang.gmail.com> wrote:
>
>
> Dear Dr. Roitberg,
>
> Sorry for the confusion.
>
> The protein has ~40 residues, and the experimental shift values for the 40
> N (of amide group) are roughly from 115 to 125. The predicted values are
> roughly from 110 to 120. In average, shifts program under-estimate the N
> shift values by 5. The correlation between simulation and experiment is
> strong.
>
> Doing the simulation in explicit is impossible ( can't reach meaningful
> time scale and sample enough conformational space.), which is why I chose
> implicit solvent model.
>
> The under-estimation mentioned above makes me a little nervous. I am a
> totally layman on NMR thing, really in need of a reasonable explanation.
>
> I jumped into a wet lab after graduate study, which is really good because
> I can finally test my simulation with benchwork. But one thing bad is that I
> can't find someone to talk about the theoretical thing. :(
>
> Thanks,
> Mingfeng
>
> On 5/3/07, Adrian Roitberg <roitberg.qtp.ufl.edu> wrote:
> >
> >
> > Dear Mingfeng,
> > N (of Amide) chemical shifts should be around 120, but some range of
> > maybe 20 to 30. Do you mean that your N shifts from the program are
> > around 25 ? That would make them 5 times smaller then experiment.
> >
> > Please clarify.
> > We can N shifts around 120 for amide N using the same setup. As
> > mentioned H-amide chemical shifts are much harder. Doing the simulation
> > in real water does not help either, because shifts does not really look
> > at the waters when doing the calculation, so you are still missing the
> > effect of charges.
> >
> > Cheers
> > Adrian
> >
> >
> > Mingfeng Yang wrote:
> > > Emmm. I am gotta reply to myself.
> > >
> > > I guess the under-estimate of backbone N chemical shift must due to
> > the
> > > fact that N-H can't form hydrogen bond with water in implicit solvent.
> > > Am I
> > > right?
> > >
> > > Anyway, it's pretty exciting to see that simulation can reproduce
> > > experimental values so well.
> > >
> > >
> >
> > >> I took an ensemble average of the chemical shift values of N, CA, CB
> > >> which were calculated by Dr. Case's SHIFTS program, and compared with
> > >> the corresponding chemical shifts determined by solution state NMR.
> > >>
> > but the N chemical shifts
> > >> are smaller than experimental values by 5.
> > >>
> >
> > --
> > Dr. Adrian E. Roitberg
> > Associate Professor
> > Quantum Theory Project and Department of Chemistry
> >
> > University of Florida PHONE 352 392-6972
> > P.O. Box 118435 FAX 352 392-8722
> > Gainesville, FL 32611-8435 Email adrian.qtp.ufl.edu
> >
> > ============================================================================
> >
> > To announce that there must be no criticism of the president,
> > or that we are to stand by the president right or wrong,
> > is not only unpatriotic and servile, but is morally treasonable
> > to the American public."
> > -- Theodore Roosevelt
> > -----------------------------------------------------------------------
> > The AMBER Mail Reflector
> > To post, send mail to amber.scripps.edu
> > To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
> >
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun May 06 2007 - 06:07:25 PDT
