Dear Francois:
This is a quick reply just to thank (a lot) you. I never approached
RED before, so that this is an occasion to finally become familiar
with it.
In the meantime, I have taken charges and radii for CONH2 from GLN or
ACN (which are identical) in the Amber 84 DOT2 database, replacing, or
attaching to, the charges and radii for PHE from the same database in
order to get the new residue phenylalanine amide, called PHA. This is
what I got:
PHA N -0.520 0 0 1.515
PHA H 0.248 0 0 0.891
PHA CA 0.214 0 0 1.648
PHA CB 0.038 0 0 1.711
PHA CG 0.011 0 0 1.648
PHA CD1 -0.011 0 0 1.648
PHA CE1 0.004 0 0 1.648
PHA CZ -0.003 0 0 1.648
PHA CE2 0.004 0 0 1.648
PHA CD2 -0.011 0 0 1.648
PHA C 0.675 0 0 1.648
PHA O -0.470 0 0 1.425
PHA N -0.867 0 0 1.515
PHA H1 0.344 0 0 0.891
PHA H2 0.344 0 0 0.891
DOT2 only needs polar hydrogens at this stage. Forget about the 4th
and 5th column. Is this trivial approach sound? The residue should
have total charge zero. Not yet checked and not yet tried with DOT2.
As to Firefly, I have some vague remembrance. Is that a new name for
the diverged GAMESS? If so, we are at GAMESS US because the main use I
made of such programs is for the most complete CASS. GAMESS US was
second choiche (1st choice MOLCAS but too expensive for us), while the
diverged gamess did not offer cass, at least at those times. Also,
extension to Microsoft was of no interest to us at the Accademia
because the cluster is Linux amd64 Debian. Actually, with don't have
any Microsoft driven computer.
Thanks again. I'll keep you informed . Please, give me an opinion on
the trivial method I used above.
Cordialement
francesco
On Thu, Sep 2, 2010 at 8:43 PM, FyD <fyd.q4md-forcefieldtools.org> wrote:
> Dear Francesco,
>
> Sorry, I was not in the lab & did not have any ssh connection...
>
>> I am looking for Amber-type charges and radii for a modified residue:
>> phenylalanine amide (amidated phenylalanine). I have no suitable ab
>> initio software for using RED. Thanks for providing data or literature
>> reference.
>
> You know "Firefly" is fast, provided at no cost, works on
> Mac/Windows/UNIX, does not need to be compiled and is fully interfaced
> by R.E.D.
>
> Anyway, you can find your NH2 fragment @
> http://q4md-forcefieldtools.org/Help/Francesco/
> http://q4md-forcefieldtools.org/Help/Francesco.tgz
>
> I used three models as test cases:
> - Cornell et al. FF:
> http://q4md-forcefieldtools.org/Help/Francesco/R.E.D./RED-vIII.4.log
> - Weiner et al. FF (the one you need if I well understood):
> http://q4md-forcefieldtools.org/Help/Francesco/R.E.D.-2/RED-vIII.4.log
>
> WARNING:
> "ESP-A2" & "ESP-C2" charge models are outdated
> Nowdays, they should _not_ be used in atomic charge derivation
> These models are only available in R.E.D. for compatibility with the past
>
> WARNING:
> "ESP-A2" & "ESP-C2" charge models & multi-molecule ESP charge derivation
> are not supported in R.E.D.-III.x
> Job(s) will be performed as a batch; each molecule being independent
>
> [fyd.lynx R.E.D.-2/Data-RED]$ grep "ESP rel" */*/out*
> Mol_m1/output1_m1: ESP relative RMS (SQRT(chipot/ssvpot)) 0.12678
> Mol_m1/output1_m1.sm: ESP relative RMS (SQRT(chipot/ssvpot)) 0.31761
> Mol_m2/output1_m2: ESP relative RMS (SQRT(chipot/ssvpot)) 0.07340
> Mol_m2/output1_m2.sm: ESP relative RMS (SQRT(chipot/ssvpot)) 0.31851
> Mol_m3/output1_m3: ESP relative RMS (SQRT(chipot/ssvpot)) 0.08973
> Mol_m3/output1_m3.sm: ESP relative RMS (SQRT(chipot/ssvpot)) 0.32505
>
> I would vote for using Mol_m1/Mol_m1-o1-sm.mol2; see
> http://q4md-forcefieldtools.org/Help/Francesco/R.E.D.-2/Data-RED/Mol_m1/punch1_m1.sm
> &
> http://q4md-forcefieldtools.org/Help/Francesco/R.E.D.-2/Data-RED/Mol_m1/punch2_m1.sm (charge equivalencing is carried a posteriori to the fit for the Weiner et al. force
> field)
>
> The FF library:
> http://q4md-forcefieldtools.org/Help/Francesco/R.E.D.-2/Data-RED/Mol_m1/Mol_m1-o1-sm.mol2
>
> You just need to define a head for the nitrogen atom in
> Mol_m1-o1-sm.mol2, to combine PHE & NH2
>
> regards, Francois
>
>
>
> _______________________________________________
> 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 03 2010 - 03:30:04 PDT
