Re: [gmx-users] mobile ligand with intermolecular interactions

[Justin Lemkul]

On 11/13/17 3:00 AM, Ahmet Yildirim wrote:

Sorry. The pymol and GROMACS give the same results for every three target
dihedrals. The updated results are here:
https://www.dropbox.com/s/7qgn7qa4rf2yc14/dih.pdf?dl=0

The results show that the average dihedrals are almost the same in both the
state A and state B for both [dihedrals] or [dihedral_restraints]. I think
this is normal?


Yes, because the purpose of selectively turning on the additional bond 
or restraint potential is to enforce the original geometry. Your results 
look fine to me, probably you just need a slightly larger force constant.


-Justin


On 13 November 2017 at 09:49, Ahmet Yildirim  wrote:


Hi,

Let me take that in reverse order:

Note that choosing appropriate atoms to apply restraints is not a trivial
matter, and results can vary depending on the choices made.


I know.

I doubt there's a bug. I tested this feature a lot myself and have used

it. How different are the dihedrals in the B-state vs. the A-state, and how
much do they differ from the target values?


The dihedrals are almost the same in both the state A and state B (except
for the first one) when the [dihedrals] or [dihedral_restraints] is used.
Please see the results https://www.dropbox.com/s/rdw6
lg0c4xviufk/dih.pdf?dl=0

I think there is bug?

On the other hand, the pymol and GROMACS calculate differently the first
target dihedral angle between 1587, 1581, 1579 and 14 atoms, respectively
-155 and 66. I don't know why? That is strange? Note that I used -155
degrees obtained from pymol as a target value of the first dihedral in the
simulation. The other two target dihedrals are the same for both pymol and
GROMACS.

Thanks.

On 13 November 2017 at 01:55, Justin Lemkul  wrote:



On 11/10/17 11:59 PM, Ahmet Yildirim wrote:


Hi,

Just tested it. The [ dihedral_restraints ] didn't help. The RMSD of the
ligand is the same when the [ dihedral_restraints ] or [ dihedrals ] is
used at end of the complex topology file. This is a bug?


I doubt there's a bug. I tested this feature a lot myself and have used
it. How different are the dihedrals in the B-state vs. the A-state, and how
much do they differ from the target values? Note that choosing appropriate
atoms to apply restraints is not a trivial matter, and results can vary
depending on the choices made.

-Justin

  From my first mail:

I think [ dihedral_restraints ] will be the same as [ dihedrals ] because

the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
taken as 0 (zero), right?


Checked also the position of the ligand during the simulation on the VMD.
It is very mobile in the decoupled state as I said in the previous mail.

*First test:*
[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
   13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
   1393   1391   2615   1   88.8 0.088.8 41.84
   1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedrals ]
; ai ajakaltype thA  fcA   thB  fcB
   1410  1393  1391  26152   -159.70.0-159.741.84
   1393  1391  2615  26142122.60.0 122.641.84
   1391  2615  2614  26102 12.80.0  12.841.84

*Second test:*

[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
   13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
   1393   1391   2615   1   88.8 0.088.8 41.84
   1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedral_restraints ]
;   aiajakal  typephiA dphiA  fcAphiB  dphiB
fcB
   1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
41.84
   1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
41.84
   1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
41.84

On 10 November 2017 at 22:36, Justin Lemkul  wrote:



On 11/10/17 2:21 PM, Ahmet Yildirim wrote:

Dear users,

I add the following [ intermolecular_interactions ] section into the
topology file of protein-ligand complex for free energy calculations.
Unfortunately this setting doesn't provide me what I want in the
decoupled
state. The ligand in the decoupled state is very mobile compared to the
coupled state. It means that the conformational sampling of the ligand
in
the decoupled state is quite different from the coupled state. That is
not
expected of course. Any suggestions as to why this might be?

Do you think that there is something wrong with the following settings,
e.g. [ dihedrals ] part?

Yes, because you're setting [dihedrals] as if they were restraints.

Dihedrals have a multiplicity, phase angle (not the angle you want it to
be, it's the shape and position of the 

Re: [gmx-users] mobile ligand with intermolecular interactions

[Ahmet Yildirim]

Sorry. The pymol and GROMACS give the same results for every three target
dihedrals. The updated results are here:
https://www.dropbox.com/s/7qgn7qa4rf2yc14/dih.pdf?dl=0

The results show that the average dihedrals are almost the same in both the
state A and state B for both [dihedrals] or [dihedral_restraints]. I think
this is normal?

On 13 November 2017 at 09:49, Ahmet Yildirim  wrote:

> Hi,
>
> Let me take that in reverse order:
>>
>> Note that choosing appropriate atoms to apply restraints is not a trivial
>> matter, and results can vary depending on the choices made.
>>
> I know.
>
> I doubt there's a bug. I tested this feature a lot myself and have used
>> it. How different are the dihedrals in the B-state vs. the A-state, and how
>> much do they differ from the target values?
>>
>
> The dihedrals are almost the same in both the state A and state B (except
> for the first one) when the [dihedrals] or [dihedral_restraints] is used.
> Please see the results https://www.dropbox.com/s/rdw6
> lg0c4xviufk/dih.pdf?dl=0
>
> I think there is bug?
>
> On the other hand, the pymol and GROMACS calculate differently the first
> target dihedral angle between 1587, 1581, 1579 and 14 atoms, respectively
> -155 and 66. I don't know why? That is strange? Note that I used -155
> degrees obtained from pymol as a target value of the first dihedral in the
> simulation. The other two target dihedrals are the same for both pymol and
> GROMACS.
>
> Thanks.
>
> On 13 November 2017 at 01:55, Justin Lemkul  wrote:
>
>>
>>
>> On 11/10/17 11:59 PM, Ahmet Yildirim wrote:
>>
>>> Hi,
>>>
>>> Just tested it. The [ dihedral_restraints ] didn't help. The RMSD of the
>>> ligand is the same when the [ dihedral_restraints ] or [ dihedrals ] is
>>> used at end of the complex topology file. This is a bug?
>>>
>>
>> I doubt there's a bug. I tested this feature a lot myself and have used
>> it. How different are the dihedrals in the B-state vs. the A-state, and how
>> much do they differ from the target values? Note that choosing appropriate
>> atoms to apply restraints is not a trivial matter, and results can vary
>> depending on the choices made.
>>
>> -Justin
>>
>>  From my first mail:
>>>
>>> I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
 the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
 taken as 0 (zero), right?

>>>
>>> Checked also the position of the ligand during the simulation on the VMD.
>>> It is very mobile in the decoupled state as I said in the previous mail.
>>>
>>> *First test:*
>>> [ intermolecular_interactions ]
>>> [ bonds ]
>>> ; ai ajtype   bA  kA bB  kB
>>>   13912615  6  0.654   0.00.654   4184.0
>>>
>>> [ angles ]
>>> ; ai ajak typethA  fcAthB  fcB
>>>   1393   1391   2615   1   88.8 0.088.8 41.84
>>>   1391   2615   2614   1   32.9 0.032.9 41.84
>>>
>>> [ dihedrals ]
>>> ; ai ajakaltype thA  fcA   thB  fcB
>>>   1410  1393  1391  26152   -159.70.0-159.741.84
>>>   1393  1391  2615  26142122.60.0 122.641.84
>>>   1391  2615  2614  26102 12.80.0  12.841.84
>>>
>>> *Second test:*
>>>
>>> [ intermolecular_interactions ]
>>> [ bonds ]
>>> ; ai ajtype   bA  kA bB  kB
>>>   13912615  6  0.654   0.00.654   4184.0
>>>
>>> [ angles ]
>>> ; ai ajak typethA  fcAthB  fcB
>>>   1393   1391   2615   1   88.8 0.088.8 41.84
>>>   1391   2615   2614   1   32.9 0.032.9 41.84
>>>
>>> [ dihedral_restraints ]
>>> ;   aiajakal  typephiA dphiA  fcAphiB  dphiB
>>> fcB
>>>   1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
>>> 41.84
>>>   1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
>>> 41.84
>>>   1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
>>> 41.84
>>>
>>> On 10 November 2017 at 22:36, Justin Lemkul  wrote:
>>>
>>>
 On 11/10/17 2:21 PM, Ahmet Yildirim wrote:

 Dear users,
>
> I add the following [ intermolecular_interactions ] section into the
> topology file of protein-ligand complex for free energy calculations.
> Unfortunately this setting doesn't provide me what I want in the
> decoupled
> state. The ligand in the decoupled state is very mobile compared to the
> coupled state. It means that the conformational sampling of the ligand
> in
> the decoupled state is quite different from the coupled state. That is
> not
> expected of course. Any suggestions as to why this might be?
>
> Do you think that there is something wrong with the following settings,
> e.g. [ dihedrals ] part?
>
> Yes, because you're setting [dihedrals] as if they were 

Re: [gmx-users] mobile ligand with intermolecular interactions

[Ahmet Yildirim]

Hi,

Let me take that in reverse order:
>
> Note that choosing appropriate atoms to apply restraints is not a trivial
> matter, and results can vary depending on the choices made.
>
I know.

I doubt there's a bug. I tested this feature a lot myself and have used it.
> How different are the dihedrals in the B-state vs. the A-state, and how
> much do they differ from the target values?
>

The dihedrals are almost the same in both the state A and state B (except
for the first one) when the [dihedrals] or [dihedral_restraints] is used.
Please see the results https://www.dropbox.com/s/rdw6
lg0c4xviufk/dih.pdf?dl=0

I think there is bug?

On the other hand, the pymol and GROMACS calculate differently the first
target dihedral angle between 1587, 1581, 1579 and 14 atoms, respectively
-155 and 66. I don't know why? That is strange? Note that I used -155
degrees obtained from pymol as a target value of the first dihedral in the
simulation. The other two target dihedrals are the same for both pymol and
GROMACS.

Thanks.

On 13 November 2017 at 01:55, Justin Lemkul  wrote:

>
>
> On 11/10/17 11:59 PM, Ahmet Yildirim wrote:
>
>> Hi,
>>
>> Just tested it. The [ dihedral_restraints ] didn't help. The RMSD of the
>> ligand is the same when the [ dihedral_restraints ] or [ dihedrals ] is
>> used at end of the complex topology file. This is a bug?
>>
>
> I doubt there's a bug. I tested this feature a lot myself and have used
> it. How different are the dihedrals in the B-state vs. the A-state, and how
> much do they differ from the target values? Note that choosing appropriate
> atoms to apply restraints is not a trivial matter, and results can vary
> depending on the choices made.
>
> -Justin
>
>  From my first mail:
>>
>> I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
>>> the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
>>> taken as 0 (zero), right?
>>>
>>
>> Checked also the position of the ligand during the simulation on the VMD.
>> It is very mobile in the decoupled state as I said in the previous mail.
>>
>> *First test:*
>> [ intermolecular_interactions ]
>> [ bonds ]
>> ; ai ajtype   bA  kA bB  kB
>>   13912615  6  0.654   0.00.654   4184.0
>>
>> [ angles ]
>> ; ai ajak typethA  fcAthB  fcB
>>   1393   1391   2615   1   88.8 0.088.8 41.84
>>   1391   2615   2614   1   32.9 0.032.9 41.84
>>
>> [ dihedrals ]
>> ; ai ajakaltype thA  fcA   thB  fcB
>>   1410  1393  1391  26152   -159.70.0-159.741.84
>>   1393  1391  2615  26142122.60.0 122.641.84
>>   1391  2615  2614  26102 12.80.0  12.841.84
>>
>> *Second test:*
>>
>> [ intermolecular_interactions ]
>> [ bonds ]
>> ; ai ajtype   bA  kA bB  kB
>>   13912615  6  0.654   0.00.654   4184.0
>>
>> [ angles ]
>> ; ai ajak typethA  fcAthB  fcB
>>   1393   1391   2615   1   88.8 0.088.8 41.84
>>   1391   2615   2614   1   32.9 0.032.9 41.84
>>
>> [ dihedral_restraints ]
>> ;   aiajakal  typephiA dphiA  fcAphiB  dphiB
>> fcB
>>   1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
>> 41.84
>>   1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
>> 41.84
>>   1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
>> 41.84
>>
>> On 10 November 2017 at 22:36, Justin Lemkul  wrote:
>>
>>
>>> On 11/10/17 2:21 PM, Ahmet Yildirim wrote:
>>>
>>> Dear users,

 I add the following [ intermolecular_interactions ] section into the
 topology file of protein-ligand complex for free energy calculations.
 Unfortunately this setting doesn't provide me what I want in the
 decoupled
 state. The ligand in the decoupled state is very mobile compared to the
 coupled state. It means that the conformational sampling of the ligand
 in
 the decoupled state is quite different from the coupled state. That is
 not
 expected of course. Any suggestions as to why this might be?

 Do you think that there is something wrong with the following settings,
 e.g. [ dihedrals ] part?

 Yes, because you're setting [dihedrals] as if they were restraints.
>>> Dihedrals have a multiplicity, phase angle (not the angle you want it to
>>> be, it's the shape and position of the minima in the function) and a
>>> force
>>> constant. So mdrun is doing what you're telling it, but what you're
>>> telling
>>> it doesn't make physical sense.
>>>
>>> [ intermolecular_interactions ]
>>>
 [ bonds ]
 ; ai ajtype   bA  kA bB  kB
13912615  6  0.654   0.00.654   4184.0

 [ angles ]
 ; ai ajak typethA  fcAthB  fcB

Re: [gmx-users] mobile ligand with intermolecular interactions

[Justin Lemkul]

On 11/10/17 11:59 PM, Ahmet Yildirim wrote:

Hi,

Just tested it. The [ dihedral_restraints ] didn't help. The RMSD of the
ligand is the same when the [ dihedral_restraints ] or [ dihedrals ] is
used at end of the complex topology file. This is a bug?


I doubt there's a bug. I tested this feature a lot myself and have used 
it. How different are the dihedrals in the B-state vs. the A-state, and 
how much do they differ from the target values? Note that choosing 
appropriate atoms to apply restraints is not a trivial matter, and 
results can vary depending on the choices made.


-Justin


 From my first mail:


I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
taken as 0 (zero), right?


Checked also the position of the ligand during the simulation on the VMD.
It is very mobile in the decoupled state as I said in the previous mail.

*First test:*
[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
  13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
  1393   1391   2615   1   88.8 0.088.8 41.84
  1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedrals ]
; ai ajakaltype thA  fcA   thB  fcB
  1410  1393  1391  26152   -159.70.0-159.741.84
  1393  1391  2615  26142122.60.0 122.641.84
  1391  2615  2614  26102 12.80.0  12.841.84

*Second test:*
[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
  13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
  1393   1391   2615   1   88.8 0.088.8 41.84
  1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedral_restraints ]
;   aiajakal  typephiA dphiA  fcAphiB  dphiB
fcB
  1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
41.84
  1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
41.84
  1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
41.84

On 10 November 2017 at 22:36, Justin Lemkul  wrote:



On 11/10/17 2:21 PM, Ahmet Yildirim wrote:


Dear users,

I add the following [ intermolecular_interactions ] section into the
topology file of protein-ligand complex for free energy calculations.
Unfortunately this setting doesn't provide me what I want in the decoupled
state. The ligand in the decoupled state is very mobile compared to the
coupled state. It means that the conformational sampling of the ligand in
the decoupled state is quite different from the coupled state. That is not
expected of course. Any suggestions as to why this might be?

Do you think that there is something wrong with the following settings,
e.g. [ dihedrals ] part?


Yes, because you're setting [dihedrals] as if they were restraints.
Dihedrals have a multiplicity, phase angle (not the angle you want it to
be, it's the shape and position of the minima in the function) and a force
constant. So mdrun is doing what you're telling it, but what you're telling
it doesn't make physical sense.

[ intermolecular_interactions ]

[ bonds ]
; ai ajtype   bA  kA bB  kB
   13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
   1393   1391   2615   1   88.8 0.088.8 41.84
   1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedrals ]
; ai ajakaltype thA  fcA   thB  fcB
   1410  1393  1391  26152   -159.70.0-159.741.84
   1393  1391  2615  26142122.60.0 122.641.84
   1391  2615  2614  26102 12.80.0  12.841.84


Should I use the following [ dihedral_restraints ] part instead of the
above [ dihedrals ] part?


Yes, this would probably accomplish what you want.

-Justin

[ dihedral_restraints ]

;   aiajakal  typephiA dphiA  fcAphiB  dphiB
fcB
   1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
41.84
   1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
41.84
   1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
41.84

I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
taken as 0 (zero), right?


Thanks in advance



--
==

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalem...@vt.edu | (540) 231-3129
http://www.biochem.vt.edu/people/faculty/JustinLemkul.html

Re: [gmx-users] mobile ligand with intermolecular interactions

[Ahmet Yildirim]

Hi,

Just tested it. The [ dihedral_restraints ] didn't help. The RMSD of the
ligand is the same when the [ dihedral_restraints ] or [ dihedrals ] is
used at end of the complex topology file. This is a bug?
>From my first mail:

> I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
> the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
> taken as 0 (zero), right?


Checked also the position of the ligand during the simulation on the VMD.
It is very mobile in the decoupled state as I said in the previous mail.

*First test:*
[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
 13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
 1393   1391   2615   1   88.8 0.088.8 41.84
 1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedrals ]
; ai ajakaltype thA  fcA   thB  fcB
 1410  1393  1391  26152   -159.70.0-159.741.84
 1393  1391  2615  26142122.60.0 122.641.84
 1391  2615  2614  26102 12.80.0  12.841.84

*Second test:*
[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
 13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
 1393   1391   2615   1   88.8 0.088.8 41.84
 1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedral_restraints ]
;   aiajakal  typephiA dphiA  fcAphiB  dphiB
fcB
 1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
41.84
 1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
41.84
 1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
41.84

On 10 November 2017 at 22:36, Justin Lemkul  wrote:

>
>
> On 11/10/17 2:21 PM, Ahmet Yildirim wrote:
>
>> Dear users,
>>
>> I add the following [ intermolecular_interactions ] section into the
>> topology file of protein-ligand complex for free energy calculations.
>> Unfortunately this setting doesn't provide me what I want in the decoupled
>> state. The ligand in the decoupled state is very mobile compared to the
>> coupled state. It means that the conformational sampling of the ligand in
>> the decoupled state is quite different from the coupled state. That is not
>> expected of course. Any suggestions as to why this might be?
>>
>> Do you think that there is something wrong with the following settings,
>> e.g. [ dihedrals ] part?
>>
>
> Yes, because you're setting [dihedrals] as if they were restraints.
> Dihedrals have a multiplicity, phase angle (not the angle you want it to
> be, it's the shape and position of the minima in the function) and a force
> constant. So mdrun is doing what you're telling it, but what you're telling
> it doesn't make physical sense.
>
> [ intermolecular_interactions ]
>> [ bonds ]
>> ; ai ajtype   bA  kA bB  kB
>>   13912615  6  0.654   0.00.654   4184.0
>>
>> [ angles ]
>> ; ai ajak typethA  fcAthB  fcB
>>   1393   1391   2615   1   88.8 0.088.8 41.84
>>   1391   2615   2614   1   32.9 0.032.9 41.84
>>
>> [ dihedrals ]
>> ; ai ajakaltype thA  fcA   thB  fcB
>>   1410  1393  1391  26152   -159.70.0-159.741.84
>>   1393  1391  2615  26142122.60.0 122.641.84
>>   1391  2615  2614  26102 12.80.0  12.841.84
>>
>>
>> Should I use the following [ dihedral_restraints ] part instead of the
>> above [ dihedrals ] part?
>>
>
> Yes, this would probably accomplish what you want.
>
> -Justin
>
> [ dihedral_restraints ]
>> ;   aiajakal  typephiA dphiA  fcAphiB  dphiB
>> fcB
>>   1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
>> 41.84
>>   1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
>> 41.84
>>   1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
>> 41.84
>>
>> I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
>> the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
>> taken as 0 (zero), right?
>>
>>
>> Thanks in advance
>>
>>
> --
> ==
>
> Justin A. Lemkul, Ph.D.
> Assistant Professor
> Virginia Tech Department of Biochemistry
>
> 303 Engel Hall
> 340 West Campus Dr.
> Blacksburg, VA 24061
>
> jalem...@vt.edu | (540) 231-3129
> http://www.biochem.vt.edu/people/faculty/JustinLemkul.html
>
> ==
>
> --
> 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 

Re: [gmx-users] mobile ligand with intermolecular interactions

[Justin Lemkul]

On 11/10/17 2:21 PM, Ahmet Yildirim wrote:

Dear users,

I add the following [ intermolecular_interactions ] section into the
topology file of protein-ligand complex for free energy calculations.
Unfortunately this setting doesn't provide me what I want in the decoupled
state. The ligand in the decoupled state is very mobile compared to the
coupled state. It means that the conformational sampling of the ligand in
the decoupled state is quite different from the coupled state. That is not
expected of course. Any suggestions as to why this might be?

Do you think that there is something wrong with the following settings,
e.g. [ dihedrals ] part?


Yes, because you're setting [dihedrals] as if they were restraints. 
Dihedrals have a multiplicity, phase angle (not the angle you want it to 
be, it's the shape and position of the minima in the function) and a 
force constant. So mdrun is doing what you're telling it, but what 
you're telling it doesn't make physical sense.



[ intermolecular_interactions ]
[ bonds ]
; ai ajtype   bA  kA bB  kB
  13912615  6  0.654   0.00.654   4184.0

[ angles ]
; ai ajak typethA  fcAthB  fcB
  1393   1391   2615   1   88.8 0.088.8 41.84
  1391   2615   2614   1   32.9 0.032.9 41.84

[ dihedrals ]
; ai ajakaltype thA  fcA   thB  fcB
  1410  1393  1391  26152   -159.70.0-159.741.84
  1393  1391  2615  26142122.60.0 122.641.84
  1391  2615  2614  26102 12.80.0  12.841.84


Should I use the following [ dihedral_restraints ] part instead of the
above [ dihedrals ] part?


Yes, this would probably accomplish what you want.

-Justin


[ dihedral_restraints ]
;   aiajakal  typephiA dphiA  fcAphiB  dphiB
fcB
  1410  1393  1391  2615 1-159.7   0.00.0-159.70.0
41.84
  1393  1391  2615  2614 1 122.6   0.00.0 122.60.0
41.84
  1393  1391  2615  2614 1  12.8   0.00.0 12.8 0.0
41.84

I think [ dihedral_restraints ] will be the same as [ dihedrals ] because
the eq. 4.84 in the manual will be equal to 4.61 if dphiA and dphiB are
taken as 0 (zero), right?


Thanks in advance



--
==

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalem...@vt.edu | (540) 231-3129
http://www.biochem.vt.edu/people/faculty/JustinLemkul.html

==

--
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.