On 9/18/16 2:26 PM, Sameer Edirisinghe wrote:
Dear Dr. Justin,

Now i can understand what what i have mistaken there. How can i derive
 [bondtypes]
for bonds i have used in .rtp file ?


Determine if you need to in the first place. If grompp gives you errors about missing parameters, first verify that your choice of atom types is correct. Then, if there are in fact interactions that the force field doesn't cover:

1. Make sure your choice of force field was appropriate
2. Parametrize them in a manner consistent with the parent force field

-Justin

Regards

On Sun, Sep 18, 2016 at 11:38 PM, Justin Lemkul <jalem...@vt.edu> wrote:



On 9/18/16 2:06 PM, Sameer Edirisinghe wrote:

Dear Dr. Justin,

If atom names names cant use in ffbonded.itp file how should i add bond
interaction used in .rtp file  to ff..itp file ?  this is the part i dont
understand. Do i need to modify ffnonbonded.itp file to correct the
"Unknown
bond_atomtype' error  which occurring while runing grompp ?  If so how i
should modify ffnonbonded.itp (I have attached my .rtp file herewith)


You do not need to modify ffnonbonded.itp.

You added an atom name in ffbonded.itp - this makes no sense and you
cannot do it.  Bonded and nonbonded interactions are defined by type.

You must use atom names only in .rtp files, because the .rtp entry tells
pdb2gmx which atoms are bonded, and how.  You can't do that by type because
that is not a unique way to identify connectivity.  The list of [bonds] in
the .rtp just tells pdb2gmx which atoms are connect, it is separate from a
[bondtypes] directive, which says what parameters are assigned to each
interaction type.

-Justin



Regards
Sameera

On Sun, Sep 18, 2016 at 10:45 PM, Justin Lemkul <jalem...@vt.edu> wrote:



On 9/18/16 1:12 PM, Sameer Edirisinghe wrote:

Dear Dr. Justin,

Thanks for the reply. Following is the whole .rtp file i used. ( As you
said it was generated by using PRODRG)


You shouldn't use the charges from PRODRG.  They assume a united-atom
force field, and are of very low quality.  You can't force them into
working with OPLS-AA.  It's fundamentally wrong.

If you want free help with a problem, you have to provide full
information. Fragments of .rtp files are not helpful and it's a waste of
time for those who are trying to give you advice.


[ bondedtypes ]

; bonds  angles  dihedrals  impropers all_dihedrals nrexcl HH14
RemoveDih
  1      1       3          1

; MDI
[ MD ]
 [ atoms ]
CAY opls_135 0 1
CAX opls_135 0.048 2
CAW opls_135 0.047 2
CAV opls_135 0.144 2
OAU opls_154 -0.112 2
CAS opls_131 0.219 2
OAT opls_236 -0.41 2
NAO opls_900 0.073 2
HAO opls_250 -0.009 2
CAE opls_260 0.056 3
CAF opls_145 -0.018 3
HAF opls_140 0.004 3
CAA opls_145 -0.018 3
HAA opls_140 0.004 3
CAD opls_145 -0.018 3
HAD opls_140 0.004 3
CAC opls_145 -0.018 3
HAC opls_140 0.004 3
CAB opls_145 -0.022 4
CAG opls_071 0.044 4
CAH opls_145 -0.022 4
CAM opls_145 -0.008 5
HAM opls_140 0.008 5
CAL opls_145 -0.008 6
HAL opls_140 0.008 6
CAI opls_145 -0.008 7
HAI opls_140 0.008 7
CAJ opls_145 -0.008 8
HAJ opls_140 0.008 8
CAK opls_145 0.072 9
NAN opls_900 0.072 9
HAN opls_250 -0.009 9
CAP opls_131 0.214 9
OAQ opls_236 -0.419 9
OAR opls_154 -0.115 9
CAZ opls_135 0.14 9
CBA opls_135 0.045 9
CBB opls_135 0.01 10
CBC opls_135 -0.01 10
HA1 opls_140 -0.013 1
HA2 opls_140 -0.013 1
HA3 opls_140 -0.014 1
HAZ opls_140 -0.013 1
HA0 opls_140 -0.013 1
HAX opls_140 -0.013 1
HAY opls_140 -0.013 1
HAV opls_140 -0.013 1
HAW opls_140 -0.013 1
HAG opls_140 -0.013 1
HAH opls_140 -0.013 1
HA4 opls_140 -0.013 1
HA5 opls_140 -0.013 1
HBA opls_140 -0.013 1
HBB opls_140 -0.013 1
HBC opls_140 -0.013 1
HBD opls_140 -0.013 1
HBF opls_140 -0.013 1
HBG opls_140 -0.013 1
HBE opls_140 -0.013 1
 [ bonds ]
     CAY HA1
     CAY HA2
     CAY HA3
     CAX CAY
     CAX HAZ
     CAX HA0
     CAX CAW
     CAW HAX
     CAW HAY
     CAW CAV
     CAV HAV
     CAV HAW
     CAV OAU
     CAS OAU
     CAS OAT
     CAS NAO
     NAO HAO
     CAE NAO
     CAE CAF
     CAE CAD
     CAF HAF
     CAF CAA
     CAA HAA
     CAB CAA
     CAD HAD
     CAD CAC
     CAC HAC
     CAB CAC
     CAB CAG
     CAG HAG
     CAG HAH
     CAH CAG
     CAH CAM
     CAH CAI
     CAM HAM
     CAM CAL
     CAL HAL
     CAK CAL
     CAI HAI
     CAI CAJ
     CAJ HAJ
     CAK CAJ
     CAK NAN
     NAN HAN
     CAP NAN
     CAP OAQ
     CAP OAR
     CAZ OAR
     CAZ HA4
     CAZ HA5
     CAZ CBA
     CBA HBA
     CBA HBB
     CBA CBB
     CBB HBC
     CBB HBD
     CBB CBC
     CBC HBF
     CBC HBG
     CBC HBE

ffbonded.itp file modified with the relevant bonds as below (part has
mentioned below)

CAY   HA1     1    0.10900   284512.0   ;
  CAY   HA2     1    0.10900   284512.0   ;
  CAY   HA3     1    0.10900   284512.0   ;
  CAX   CAY     1    0.15220   265265.6   ;
  CAX   HAZ     1    0.10900   284512.0   ;



As I said before, you cannot use atom names in ffbonded.itp.

I know atom name is irrelevant but im bit confused how to write .rtp

without names for my 59 atom molecule.


The .rtp file *does* use atom names.  See the manual.  That part is
fine.
What you cannot do is try to assign parameters using names.  Those
require
types.

-Justin



regards

On Sun, Sep 18, 2016 at 10:09 PM, Justin Lemkul <jalem...@vt.edu>
wrote:



On 9/18/16 12:30 PM, Sameer Edirisinghe wrote:

Dear users,

There is one thing i cannot clarify from gromacs manual which is about
following error occurred while running grompp

Fatal error:
Unknown bond_atomtype CAY

 I used to create .rtp file from atom types which are already defined
in
 atomtype.atp file. But still above error occurring while running
grompp.
Not sure should i add again thes atoms to atp file and if so how to
add
them. Part of my rtp file is as below, I have modified the
ffbonded.itp
(only bond types added) successfully.


What did you add to ffbonded.itp?  Whatever it was is causing an
error.

Based on the .rtp entry below, it looks like you tried to use atom
names
in
ffbonded.itp.  That's not right.  All interactions (bonded and
nonbonded)
are defined by atom type.  The name is irrelevant.  If you're re-using
only
OPLS-AA atom types, most of the bonded interactions should be present
already.

[ bondedtypes ]

; bonds  angles  dihedrals  impropers all_dihedrals nrexcl HH14
RemoveDih
  1      1       3          1


You're missing entries here if you're trying to create an OPLS-derived

force field.  Look at oplsaa.ff/aminoacids.rtp

; MDI

[ MD ]
 [ atoms ]
CAY opls_135  0.000  1
CAX opls_135  0.048 2
CAW opls_135  0.047 2
CAV opls_135  0.144 2
OAU opls_154 -0.112 2


This can't possibly be right.  You've defined only heavy atoms
(OPLS-AA

is
an all-atom force field, so there should be H) and the net charge of
this
residue is +0.117, which is physical nonsense.  Based on the names and
suspect charges, it looks like you're just porting over a PRODRG
united-atom topology and trying to call it OPLS-AA.  That's not going
to
work.

Note that you also need [bonds] if you want the residue to actually be
chemically reasonable, otherwise it's just a collection of atoms
sitting
near one another, which will immediately explode if you try to run a
simulation.

-Justin

--
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================
--
Gromacs Users mailing list

* Please search the archive at http://www.gromacs.org/Support
/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
send a mail to gmx-users-requ...@gromacs.org.


--
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================
--
Gromacs Users mailing list

* Please search the archive at http://www.gromacs.org/Support
/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
send a mail to gmx-users-requ...@gromacs.org.




--
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================
--
Gromacs Users mailing list

* Please search the archive at http://www.gromacs.org/Support
/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
send a mail to gmx-users-requ...@gromacs.org.


--
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Reply via email to