Previous Thread
Next Thread
Print Thread
Page 1 of 2 1 2
NMA parameters
#7989 08/23/05 11:39 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear CHARMMers,

I have a question to parametrize N-methylacetamide (NMA).
In the CHARMM topology file (top_all22_model.inp), the structure of NMA is designed to optimize the trans-NMA (cis-cis) structure according to the terminology of Nandini and Sathyanarayana (J. Mol. Struct. Theochem 579 (2002) 1-9). According to their study, there are several stable structures for NMA even in vacuum. I am wondering whether it is acceptable to change the dihedral angle in the IC table to have different NMA conformers or not.

I appreciate your comments.

Best wishes,

Re: NMA parameters
fujisaki #7990 08/24/05 12:12 AM
Joined: Sep 2003
Posts: 8,504
rmv Online Content
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 8,504
Not a problem-- but the IC table is used mostly for coordinate building and analysis, and has no impact on energetics.


Rick Venable
computational chemist

Re: NMA parameters
rmv #7991 08/24/05 12:57 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

note that the peptide bond in NMA (and the peptide bond in general) was optimized to be planar in charmm. however, the peptide bond around the nitrogen is pyramidal in the gas phase. the reason the structure is planar in charmm is due to hydrogen bonding of the peptide bond leading to a change in hybridization, such that the planar structure is more stable.

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #7992 08/24/05 06:58 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Rick and Alex,

Thanks for your replies and useful information.
I have a further problem now. I am trying to rotate
the methyl groups in NMA by modifying the topology file
(top_all22_model.inp), but it does not seems to work.
Which file and which parameters should I modify to rotate the methyl group, say, by 60 degrees around the CN or CC axis?

I appreciate your help.

Best wishes,

Re: NMA parameters
fujisaki #7993 08/24/05 07:44 PM
Joined: Sep 2003
Posts: 8,504
rmv Online Content
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 8,504
That depends somewhat on how you are obtaining Cartesian coords, and what you're trying to do. Note that changing the IC data in the toplogy file has no effect unless you're using IC BUILD to create Cartesian coords.

Assuming you have coords, one way is to define an axis along the bond with COOR AXIS (and a double atom selection), and then use COOR ROTA AXIS to rotate the methyl group (via an atom selection) around that axis.

corman.doc
select.doc


Rick Venable
computational chemist

Re: NMA parameters
rmv #7994 08/25/05 12:34 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

it is not clear what you are trying to do. do you want the minimized structure to have the methyl hydrogens rotated by 60 deg.?

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #7995 08/25/05 03:42 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Rick and Alex,

What I want to do is to have a stable structure of NMA with (one of or both of) methyl group rotated by 60 degees. So I need to modify the parameter file, but I can't understand which parameters I have to choose. I thought I need to modify the following dihedral angles

HA CT3 NH1 C
O C CT3 HA

but the parameter file assumes that their potential barriers are zero! So I am wondering why NMA become a stable structure ...

I appreciate your comments.

Best wishes,

Re: NMA parameters
fujisaki #7996 08/25/05 06:16 PM
Joined: Sep 2003
Posts: 8,504
rmv Online Content
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 8,504
Rotation of methyl groups hasn't been a dominant focus of CHARMM parameter development; it's not as easy as it might seem, as there are quantum effects involved in that fairly rapid rotation. For NMA itself, the principal torsional term is likely that for CT3-C-NH1-CT3, and an improper torsion for the planar C=O group. The rotational barriers for the CH3 group are probably mostly based on VDW interactions; you should write a script to compute the energy as function of CH3 rotation.


Rick Venable
computational chemist

Re: NMA parameters
rmv #7997 08/26/05 02:25 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Rick,

Thanks for your help.

I calculated the potential energy as a function of the methyl group rotation angle (while other atoms fixed), and found that the structure I found is the only minimum. The barrier height is about 0.8 kcal/mol, and it comes from, as Rick said, the VDW interaction (attraction) because there is no dihedral interaction as I mentioned. On the other hand, Guo and Karplus (J.Phys.Chem.96(1992)7273) found that the barrier height is about 0.1 kcal/mol for NMA in vacuum by the ab initio method, so this means that the CHARMM potential is too high for the methyl group rotation. That is, CHARMM is not so reliable when we consider the methyl group rotation!

Re: NMA parameters
rmv #7998 08/27/05 02:26 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

electrostatics will also play a role, but that's besides the point. if the orienation of the methyl groups is that important, do QM surface of the rotation and then adjust the dihedral parameters for the terms you listed above to reproduce the QM surfaces. A single 3 fold term with a relatively small force constant should be adequate. switching the phase between 0 and 180 will allow you to control the orientation of the methyl hydrogens.

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #7999 08/29/05 02:53 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Alex,

Thanks for your reply.
Our group is now working on QM calculations for NMA. I will report something on this issue later.

Re: NMA parameters
alex #8000 09/01/05 03:47 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Alex,

I have a question to modify the dihedral force parameters. As you know, I want to change the dihedral parameters for the methyl group in NMA (or any peptide), and I changed the parameter file from

HA CT3 NH1 H 0.00 3 0.0

to

HA CT3 NH1 H 0.10 3 0.0

and I calculated the dihedral energy as a function of the methyl group rotation angle (while other degrees of freedom fixed). I got the three fold dihedral interaction as expected. The barrier height is 0.6 kcal/mol because there are three hydrogens.

Then I changed the delta=0.0 to, say, 180.0, that is, from

HA CT3 NH1 H 0.10 3 0.0

to

HA CT3 NH1 H 0.10 3 180.0

and I expected 60 degree shift for the potential curve, but the result is about 12.5 degree shift. How can I understand this?

I appreciate your comments.

Re: NMA parameters
fujisaki #8001 09/03/05 06:28 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

1) be sure that there is not an additional dihedral parameter with a different multiplicity acting on the dihedral you are studying.

2) the combination of electrostatic, LJ and dihedral terms are contributing to the energy surface, that the obtained value of the minimum is due to the combination of these terms. do an analysis of the individual energy terms as a function of the dihedral angle to understand in better detail what is going on. the substitution options can write out the individual terms as a function of the dihedral.

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #8002 09/04/05 12:57 AM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Alex,

I checked what you said, but I still can't solve this problem.
Could you look at my input code, and give me some advice?

------------------------------------------------------------

* NMA in vacuum
*

BOMBLEV -1

! Topology and parameters

open unit 1 card read name top_all22_model.inp
read RTF card unit 1
close unit 1

open unit 1 card read name par_all22_prot.inp
read PARA card unit 1
close unit 1

!
! read in sequence
read sequ card
* sequence for NMA
*
1
NMA

generate nma setup

! build in missing coordinates using values in
! the parameter set
ic para
ic seed 1 CR 1 N 1 C
ic build

coor init sele type H* end
hbuild

print coor

mini conj nstep 1000 Inbfrq 10 CUTNB 13.0 CTONNB 8.0 CTOFNB 12.0 -
group vdw fswitch vswitch

coor axis sele type CR end sele type N end

set j 1
label loop

coor rotate axis phi @j sele type HR1 .or. type HR2 .or. type HR3 end
energy

!open write formatted unit 10 name scratch/NMA-rot@j.pdb
!write coor pdb unit 10
!* Rotated structure
!*
!close unit 10

coor rotate axis phi -@j sele type HR1 .or. type HR2 .or. type HR3 end

incr j by 1
if j lt 181 goto loop

stop

------------------------------------------------------------

I just changed the parameter files as follows:

------------------------------------------------------------

!HA CT3 NH1 C 0.0000 3 0.00 ! ALLOW PEP
HA CT3 NH1 C 0.1000 3 120.00 ! ALLOW PEP
! LK for autogenerate dihe, sp2-methyl, no dihedral potential

------------------------------------------------------------

I expected that the dihedral interaction energy displayed in "ENER INTERN>" as a function of the dihedral angle should shift by 60 degrees, but it results in 16.5 degrees shift.

I appreciate your reply.

Re: NMA parameters
fujisaki #8003 09/04/05 07:02 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

below is an example for calculating the surface; note the use of "cons dihe" to sample the dihedral surface. keep the phase of the dihedral at 0 or 180.

alex

set n 999.
set 3 @n ! cutim
decr n by 1
set 4 @n ! cutnb
decr n by 3
set 5 @n ! ctonnb
incr n by 2
set 6 @n ! ctofnb
set 7 switch
set 8 atom
set 9 vatom

update inbfrq 20 ihbfrq 0 imgfrq 20 -
@7 @8 vswitch @9 cutim @3 cutnb @4 ctonnb @5 ctofnb @6

mini abnr nstep 100 nprint 10
mini nraph nstep 50 nprint 5 tolgrd 0.0

! set minimimum values to offset surface to 0.0
set a ?ener
set b ?elec
set c ?vdw
set d ?dihe
set e ?angl
set f ?bond
set g ?urey

open unit 99 write form name butane_cccc_@id2.surf

set dihe 180.

label loop

cons dihe but 1 c1 but 1 c2 but 1 c3 but 1 c4 forc 10000.0 min @dihe

mini abnr nstep 100 nprint 10 inbfrq 100
mini nraph nstep 20 nprint 10 tolgrd 0.0
cons cldh

energy
! offset by global minimum energy
set 5 ?ener
decr 5 by @a
set 6 ?elec
decr 6 by @b
set 7 ?vdw
decr 7 by @c
set 8 ?dihe
decr 8 by @d
set 9 ?angl
decr 9 by @e
set 0 ?bond
decr 0 by @f
set 4 ?urey
decr 4 by @g

write title unit 99
* @dihe @5 @6 @7 @8 @9 @0 @4
*

decr dihe by 5.

if dihe ge 0.0 goto loop


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #8004 09/05/05 12:05 AM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Alex,

Thanks for your reply and information.

First off I have to say that I misunderstood in the previous post. I looked at the minimized structure with the following parameter file change

!HA CT3 NH1 C 0.0000 3 0.00 ! ALLOW PEP
HA CT3 NH1 C 0.1000 3 120.00 ! ALLOW PEP
! LK for autogenerate dihe, sp2-methyl, no dihedral potential

and the methyl group in NMA certainly shifts by 120/3=40 degrees. But the problem is if I shift further the methyl group rotation angle, this is not the stable structure with respect to the dihedral interaction. I think this is strange. This does not occur if I choose 0 degree or 180 degrees, but it does for other dihedral shift angles. How can I understand this?

Re: NMA parameters
fujisaki #8005 09/05/05 12:29 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

look at the individual energy components and you will figure it out. note that you can use the INTER command with a selection of 4 atoms to output the dihedral energy for those 4 atoms.

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #8006 09/05/05 06:39 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Dear Alex,

THanks for your reply.

I used

INTER sele type N .or. type CR end

instead of just "energy" command in the code of post #7789, and I looked at INTE INTERN lines, but the result turns out to be the same as the ENER INTERN lines with the energy command. So this minimized structure is not stable when I rotate the methyl group. I would like to know what happens when the dihedral parameter in the CHARMM parameter file changes from 0 or 180. It seems that only 0 and 180 degrees are allowed in CHARMM, and if we use others, some error might occur ...

I appreciate your comment.

Re: NMA parameters
fujisaki #8007 09/06/05 12:12 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

at this stage I don't think I can be of additional help. you have to go carefully through the documentation and really think about what you are trying to do. our contribution to the forum is to point you in the right direction to solve your problem; we can't solve all the details of the problem.

alex


School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201
Re: NMA parameters
alex #8008 09/08/05 05:39 PM
Joined: Oct 2004
Posts: 71
F
Forum Member
OP Offline
Forum Member
F
Joined: Oct 2004
Posts: 71
Hi Alex,

Thanks for your reply.
I think I have solved my problem! Thanks for your guidance!

Page 1 of 2 1 2

Moderated by  alex, lennart, rmv 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.4
(Release build 20200307)
Responsive Width:

PHP: 5.6.33-0+deb8u1 Page Time: 0.018s Queries: 54 (0.010s) Memory: 1.0895 MB (Peak: 1.3030 MB) Data Comp: Off Server Time: 2020-10-24 17:40:18 UTC
Valid HTML 5 and Valid CSS