Topic Options
#37606 - 08/15/19 06:45 AM strengthen hydrogen bond
MD4DW Online   confused
Forum Member

Registered: 07/10/19
Posts: 10
Hey,

I would like to strengthen a hydrogen bond between an oxygen atom and a hydrogen atom in my system. I thought RESDistance might do the trick (I'm open to other suggestions though!).
I haven't used this module before though and I find it a bit hard to understand how to use it and how to define its parameters for this sort of use -- can someone help me with that?
Thanks!

Top
#37607 - 08/15/19 09:38 AM Re: strengthen hydrogen bond [Re: MD4DW]
lennart Online   content

Forum Member

Registered: 09/25/03
Posts: 4741
Loc: ~ 59N, 15E
We use the NOE module for this purpose. The following creates three harmonic restraints with a flat bottom from rmin to rmax, and a maximal force fmax at long distances:

NOE
ASSI SELE ATOM DNA2 12 O6 END SELE ATOM DNA1 1 H42 END -
kmin 10.0 rmin 1.6 kmax 10.0 rmax 2.0 fmax 1.0
ASSI SELE ATOM DNA2 12 H1 END SELE ATOM DNA1 1 N3 END -
kmin 10.0 rmin 1.7 kmax 10.0 rmax 2.1 fmax 1.0
ASSI SELE ATOM DNA2 12 H22 END SELE ATOM DNA1 1 O2 END -
kmin 10.0 rmin 1.6 kmax 10.0 rmax 1.9 fmax 1.0
END

]
_________________________
Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden

Top
#37624 - 08/21/19 05:34 AM Re: strengthen hydrogen bond [Re: lennart]
MD4DW Online   confused
Forum Member

Registered: 07/10/19
Posts: 10
I did this with your parameters:
NOE
ASSI SELE (segid @CURRENTSEGID) .and. @Hatom END -
SELE (segid @CURRENTSEGID) .and. @Hacceptor END -
kmin 10.0 rmin 1.6 kmax 10.0 rmax 2.0 fmax 1.0
END

and the atom pairs still kept on breaking away during my dynamics...
so I amped up the the parameters:
kmin 50000.0 rmin 1.3 kmax 50000.0 rmax 1.9 fmax 10000
and also added
NOE
SCALE 3.0
END

and still, the Hbond breaks apart during my dynamics.

What's going on / what am I doing wrong ??

here's the ANALysis output:

NOE> PRINT ANAL ! See what we have so far
NOEPRI: LISTING OF ALL CONSTRAINTS
NUMBER OF CONSTRAINTS= 5 SCALE FACTOR= 3.000

NOE ATOMS
RMIN KMIN RMAX KMAX FMAX TCON REXP
R RDELTA ENERGY FORCE RAVE
RESTRAINT: 1 1 1
FIRST SET ATOM: 4038 PROA 255 HZ1
SECOND SET ATOM: 1806 PROA 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.788 0.000 0.000 0.000 0.146
RESTRAINT: 2 1 1
FIRST SET ATOM: 8447 PROB 255 HZ1
SECOND SET ATOM: 6215 PROB 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.674 0.000 0.000 0.000 0.146
RESTRAINT: 3 1 1
FIRST SET ATOM:12856 PROC 255 HZ1
SECOND SET ATOM:10624 PROC 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.901 0.001 0.144 207.625 0.146
RESTRAINT: 4 1 1
FIRST SET ATOM:17265 PROD 255 HZ1
SECOND SET ATOM:15033 PROD 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.599 0.000 0.000 0.000 0.146
RESTRAINT: 5 1 1
FIRST SET ATOM:21674 PROE 255 HZ1
SECOND SET ATOM:19442 PROE 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.700 0.000 0.000 0.000 0.146
TOTAL Violations are 0.001
TOTAL # of Violations is 1

NOE> PRINT ANAL CUT 2.0 ! list
NOEPRI: LISTING FOR CONSTRAINTS WITH RDELTA > 2.000
NUMBER OF CONSTRAINTS= 5 SCALE FACTOR= 3.000

NOE ATOMS
RMIN KMIN RMAX KMAX FMAX TCON REXP
R RDELTA ENERGY FORCE RAVE
TOTAL Violations are 0.001
TOTAL # of Violations is 0

NOE> END
NOE: CURRENT NUMBER OF CONSTRAINTS= 5

Top
#37625 - 08/21/19 08:32 AM Re: strengthen hydrogen bond [Re: MD4DW]
lennart Online   content

Forum Member

Registered: 09/25/03
Posts: 4741
Loc: ~ 59N, 15E
Your input looks fine to me, so I don't quite know what may be wrong.

What does the NOE analysis tell you after the dynamics run?
Do you see the NOE energy term increase in the logfile?
Can you post a brief, but working, version of your dynamics input file.
_________________________
Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden

Top
#37636 - 08/26/19 06:53 AM Re: strengthen hydrogen bond [Re: lennart]
MD4DW Online   confused
Forum Member

Registered: 07/10/19
Posts: 10
Do I need to somehow loop over the dynamics part in order for force constraints to get the updated positions of the atom pare and therefore update the force during the dynamics or something ?

in any event, here is some of my code:

relevant input:

!---strengthen ASP116-RSB Hbond---
set Hatom (resid 255) .and. (resname LYR) .and. (type HZ1)
set Hacceptor (resid 116) .and. (resname ASP) .and. (type OD2)

set i = 0

goto restrainbond

label restrainbond
incr i by 1
if @i EQ 1 then set CURRENTSEGID PROA
if @i EQ 2 then set CURRENTSEGID PROB
if @i EQ 3 then set CURRENTSEGID PROC
if @i EQ 4 then set CURRENTSEGID PROD
if @i EQ 5 then set CURRENTSEGID PROE

!with RESD -- not tested yet
!RESDistance KVAL 1000.0 RVAL 1.6 -
! 1.0 @CURRENTSEGID 116 OD2 @CURRENTSEGID 255 HZ1

!with NOE
NOE
ASSI SELE (segid @CURRENTSEGID) .and. @Hatom END -
SELE (segid @CURRENTSEGID) .and. @Hacceptor END -
kmin 50000.0 rmin 1.3 kmax 50000.0 rmax 1.9 fmax 10000
END

if @i .lt. 5 then goto restrainbond

OPEN WRITE UNIT 77 CARD NAME NOE_step7_3.DAT
NOE
SCALE 3.0 ! Multiply all energies and forces by 3
WRITE UNIT 77 ANAL
PRINT ANAL ! See what we have so far
PRINT ANAL CUT 2.0 ! list
END
!---------------------------------

DYNA CPT leap restart nstep @nstep timestep 0.001 echeck -1 -
nprint 10 iprfrq 100 ntrfrq 0 isvfrq 50 -
iunrea 11 iunwri 12 iuncrd 13 iunvel -1 kunit -1 -
nsavc 10 nsavv 0 -
omm langevin gamma 10 firstt @temp finalt @temp -
prmc tens 0.00001 przz 1.0 iprsfrq 15

! get NOE analysis
NOE
WRITE UNIT 77 ANAL
PRINT ANAL ! See what we have so far
PRINT ANAL CUT 2.0 ! list
END


The ANALysis printed after the dynamics (the one before the dynamics is in my last post):

CHARMM> ! get NOE analysis
CHARMM> NOE

NOE> WRITE UNIT 77 ANAL
RDTITL> PRINT ANAL
RDTITL> No title read.

NOE> PRINT ANAL ! See what we have so far
NOEPRI: LISTING OF ALL CONSTRAINTS
NUMBER OF CONSTRAINTS= 5 SCALE FACTOR= 3.000

NOE ATOMS
RMIN KMIN RMAX KMAX FMAX TCON REXP
R RDELTA ENERGY FORCE RAVE
RESTRAINT: 1 1 1
FIRST SET ATOM: 4038 PROA 255 HZ1
SECOND SET ATOM: 1806 PROA 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.627 0.000 0.000 0.000 0.146
RESTRAINT: 2 1 1
FIRST SET ATOM: 8447 PROB 255 HZ1
SECOND SET ATOM: 6215 PROB 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.597 0.000 0.000 0.000 0.146
RESTRAINT: 3 1 1
FIRST SET ATOM:12856 PROC 255 HZ1
SECOND SET ATOM:10624 PROC 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
4.394 2.494 71830.292 30000.000 0.146
RESTRAINT: 4 1 1
FIRST SET ATOM:17265 PROD 255 HZ1
SECOND SET ATOM:15033 PROD 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
1.566 0.000 0.000 0.000 0.146
RESTRAINT: 5 1 1
FIRST SET ATOM:21674 PROE 255 HZ1
SECOND SET ATOM:19442 PROE 116 OD2
1.300 50000.000 1.900 50000.000 10000.000 0.000 1.000
5.984 4.084 119522.538 30000.000 0.146
TOTAL Violations are 6.578
TOTAL # of Violations is 2


As you can see, some bonds are well over the NOE flat energy distance range (5.98 A and 4.39 A)


Edited by MD4DW (08/26/19 06:57 AM)

Top
#37637 - 08/26/19 12:23 PM Re: strengthen hydrogen bond [Re: MD4DW]
rmv Online   content

Forum Member

Registered: 09/17/03
Posts: 8373
Loc: 39 03 48 N, 77 06 54 W
Check your output log, esp. for the results of atom selections.
_________________________
Rick Venable
computational chemist


Top
#37638 - 08/27/19 03:26 AM Re: strengthen hydrogen bond [Re: MD4DW]
lennart Online   content

Forum Member

Registered: 09/25/03
Posts: 4741
Loc: ~ 59N, 15E
You OMMitted an important piece of information - you are running this on a GPU using OpenMM, which does not support the NOE restraint.

Then I believe that RESD should be used. Something like this should introduce a harmonic restraint between atoms PROA 255 HZ1 and PROA 116 OD2, with an equilibrium distance of 1.8 and a force constant 50.0 kcal/mol/^2:


resd kval 50.0 Rval 1.8 Eval 2 -
1.0 PROA 255 HZ1 PROA 116 OD2

The corresponding energy term is
E(Rij)=50/2 *(Rij -1.8)^2
_________________________
Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden

Top
#37639 - 08/27/19 04:46 AM Re: strengthen hydrogen bond [Re: lennart]
MD4DW Online   confused
Forum Member

Registered: 07/10/19
Posts: 10
I didn't realize that some modules just do not work on OMM... Does the PULL module also not work on OMM? I was struggling with that one too in the past.

In any event, thanks for your help thus far.

And by the way, that's a nice pun you did there wink

Top
#37640 - 08/27/19 04:48 AM Re: strengthen hydrogen bond [Re: MD4DW]
lennart Online   content

Forum Member

Registered: 09/25/03
Posts: 4741
Loc: ~ 59N, 15E
openmm.doc and testcases test/c*test/omm*inp show what works in omm.
_________________________
Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden

Top

Moderator:  BRBrooks, John Legato