Dear Maral,
I finally looked at the P9171 job:
First in your input PDB:
ATOM 9 HA2 GLY 1 -1.743 1.370 -1.489 C
You define a carbon atom with an hydrogen atom name: I am glad to see
that PyRED does handle that case (because the element is defined as
"C") but this is error prone - to be avoided I think: better using:
ATOM 9 CB GLY 1 -1.743 1.370 -1.489 C
thus if there is a beta-carbone in your molecule it might mean a
N3-substituted Alanine residue?
Thus you could end up with the following names (I also renamed other
atoms to get names easy to be used at the _end_ of your MD simulations
using the 'ptraj' tools):
ATOM 1 C1 ACE 0 -2.444 -2.719 -0.324 C
ATOM 2 C2 ACE 0 -2.429 -1.189 -0.324 C
ATOM 3 O2 ACE 0 -3.483 -0.555 -0.324 O
ATOM 4 H11 ACE 0 -3.465 -3.102 -0.324 H
ATOM 5 H12 ACE 0 -1.923 -3.072 -1.214 H
ATOM 6 H13 ACE 0 -1.923 -3.072 0.566 H
ATOM 7 N ALA 1 -1.251 -0.609 -0.324 N
ATOM 8 CA ALA 1 -1.117 0.833 -0.324 C
ATOM 9 CB ALA 1 -1.743 1.370 -1.489
C !!!
ATOM 10 N31 ALA 1 -1.629 2.793 -1.478 N
ATOM 11 N32 ALA 1 -1.514 4.216 -1.467
N1+
ATOM 12 N33 ALA 1 -1.400 5.640 -1.456
N1-
ATOM 13 C ALA 1 0.352 1.230 -0.324 C
ATOM 14 O ALA 1 1.230 0.370 -0.324 O
ATOM 15 H ALA 1 -0.419 -1.182 -0.324 H
ATOM 16 HA ALA 1 -1.595 1.243 0.566 H
ATOM 17 HB1 ALA 1 -2.796 1.090 -1.498 H
ATOM 18 HB2 ALA 1 -1.255 0.973 -2.379 H
ATOM 19 N1 NME 2 0.625 2.534 -0.324 N
ATOM 20 C3 NME 2 1.981 3.041 -0.324 C
ATOM 21 H1 NME 2 -0.141 3.192 -0.324 H
ATOM 22 H31 NME 2 1.962 4.132 -0.324 H
ATOM 23 H32 NME 2 2.502 2.688 0.566 H
ATOM 24 H33 NME 2 2.502 2.688 -1.214 H
Then using P9171, i.e. the outputs in:
P9171/Data-R.E.D.Server/Data-Default-Proj/ I define new inputs for a new run:
cp P9171/Data-R.E.D.Server/Data-Default-Proj/Mol_red1-out.ent
./Mol_red1.pdb (geometry already optimized)
cp P9171/Data-R.E.D.Server/Data-Default-Proj/Project-out.config
./Project.config
and keep only:
MOLECULE1-TITLE = ACE-N3_ALA-NME
MOLECULE1-INTRA-MCC1 = 0.0 | 1 2 3 4 5 6 | R
MOLECULE1-INTRA-MCC1 = 0.0 | 19 20 21 22 23 24 | R
modify:
MOLECULE1-ATMTYPE = CT HC HC HC C O N H CX H1 CT H1 H1 NC N* N? C
O N H CT H1 H1 H1
into (with NX NY NZ are the new types for that N3 azido group):
MOLECULE1-ATMTYPE = CT HC HC HC C O N H CX H1 CT H1 H1 NX NZ NY C
O N H CT H1 H1 H1
and modify (optional; PyRED auto. corrects if duplicates are found):
MOLECULE1-ATMNAME = C1 H3 H4 H5 C2 O1 N1 H6 CA HA HA2 H1 H2 N3
N32 N322 C O N H CH3 HH31 HH32 HH33
into - far more understandable ;-)
MOLECULE1-ATMNAME = C1 H11 H12 H13 C2 O2 N H CA HA CB HB1 HB2 N31
N32 N33 C O N1 H1 C3 H31 H32 H33
See $AMBERHOME/dat/leap/parm/parm99.dat
NY 14.01 0.530 nitrile N (Howard et al.JCC,16,243,1995)
You are going to use that NY atom type (terminal nitrogen in nitrile
and azido by now), while NX and NZ are two new types to be defined:
i.e.:
---N=N=N -> types: ---NX-NZ-NY
the frcmod.unknown file generated in a PyRED job will list the
corresponding missing FF parameters; so you have to create a
'frcmod.user' input file (below) manually according to that
'frcmod.unknown' file
then do a new archive, and resubmit:
tar zcvf archive.tgz Mol_red1.pdb Project.config frcmod.user
See also
http://q4md-forcefieldtools.org/Tutorial/Tutorial-4-demo1.pdf
for more information about "MOLECULE1-ATMTYPE" and far more...
You could submit that data in REDDB once you would have validate the
FF: this is an interesting analog...
regards, Francois
frcmod.user file you could use as a starting point:
FRCMOD input file to be used by PyRED - q4md-forcefieldtools.org
MASS mass pol Source
NX 14.01 0.000 Same as nitrile NY (Howard
et al.JCC,16,243,1995)
NZ 14.01 0.000 Same as nitrile NY
BOND K(kcal.mol-1.ang-2) Dist0(ang) Source
NX-NZ 600 1.23 parm99/QM optimized geometry(*)
NY-NZ 600 1.10 parm99/QM optimized geometry(*)
CT-NX 370 1.47 adapted parm99/QM optimized
geometry(*)
ANGLE K(kcal.mol-1.rad-2) Theta0(deg) Source
CX-CT-NX 50.0 109.50 parm99 here CX is CT
ff99SB/ffXXSB_?
H1-CT-NX 50.0 109.50 parm99
CT-NX-NZ 70.0 113.70 guess parm99/QM optimized
geometry(*)
NX-NZ-NY 80.0 180.00 parm99/set to 180
DIHEDRAL Path V(kcal.mol-1.rad-1) Phase(deg.) Period Source
CX-CT-NX-NZ 1 0.00 0.0 3.0 V3 set to zero
H1-CT-NX-NZ 1 0.00 0.0 3.0 V3 set to zero
CT-NX-NZ-NY 1 0.00 0.0 1.0 V1 set to zero
IMPROPER V(kcal.mol-1.rad-1) Phase(deg.) Period Source
NONBON R*(ang) Eps(kcal.mol-1) Source
NX 1.8240 0.1700 OPLS - like NY (Howard et
al.JCC,16,243,1995)
NZ 1.8240 0.1700 OPLS - like NY
(*) better running a QM job with the following "System.config" file to
measure more accurate values:
MEPCHR_Calc = OFF
CHR_TYP = RESP-X1
METHOD_OPTCALC = B3LYP
BASSET_OPTCALC = 6-311+G(d,p)
> I am trying to make a non standard residue (an azide group is
> attaching to a GLY residue) for AMBER using RED. The job was PROJECT
> P9171.
>
> 1. AZA nonstandard residue: It seems that RED is not able to assign
> the right atom types to Nitrogens in azide group two nitrogens with
> double bonds (one with +1 charge and the other with -1 charge. I
> attached their figure to this email.
>
> These are the unknown parameters in frcmod.unknown
> N?
> CT-NC
> N*-N?
> N*-NC
> CX-CT-NC
> H1-CT-NC
> N?-N*-NC
> CT-NC-N*
> CX-CT-NC-N*
> H1-CT-NC-N*
> N?-N*-NC-CT
>
> How can I get the right atom types for these atoms? I already tried
> “http://www.quimica.urv.es/~bo/MOLMOD/General/Forcefields/AMBER.html#N
> <http://www.quimica.urv.es/~bo/MOLMOD/General/Forcefields/AMBER.html#N>” but
> I can not find these exact atoms types.
> How can I get the unknown parameters?
>
> I also attached the mol2, leaprc.q4mdfft (that I modified),
> frcmod.unknown and frcmod.known for both AZA to this email.
F.-Y. Dupradeau
---
http://q4md-forcefieldtools.org/FyD/
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Feb 26 2017 - 02:30:02 PST
