Re: Need help to generate ubq-get-energy.conf

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu May 14 2009 - 11:33:57 CDT

Hi Mert,
an old group member has just pointed out to me that you appear to be
working from an old version of the namd tutorial. The way that version
worked was to fix all of the atoms except the protein atoms, which
prevents namd from calculating their energies (this is why a new psf/dcd
were not needed). The newer versions of the namd tutorial instead use
namdenergy. Either solution should be fine for your purposes.
Best,
Peter

Mert Gür wrote:
> Dear Eric,
> I got a little bit confused because this is not mentioned in the tutorial.
> According to the tutorial I thought it was only enough to generate the
> protein dcd file. Now you are telling me that I also have to generate a
> protein psf and pdb file. This means that I have to write a conf file
> for a simulation in vacuum am I right? Would that give me the correct
> protein energies which fit the one I obtained in windows via namd's
> namdenergy visual plugin.
> Because in that case I wont have for full-system periodic electrostatics
> which means that PME's are not included in my results. Hence the result
> will be the same with MDenergy. If I am wrong please feel free to
> correct me.
> But when I use vmd's mdenergy extension in windows I think these are
> included.
> I have 1500 trajectories so I need to evaluate energies using a script
> that is why I am not doing calculations in windows.
> I tought that by doing the way I did, it would include PME and periodic
> BC and give me the energy of the protein (This is I think what vmd does
> in windows)
> So actually I could reformulate my problem such that the energies for
> the protein I obtained via linux tutorial and windows tutorial are
> different.
> Does anybody have a example ubq-get-energy.conf file (It is not included
> in the tutorial files)
> Best,
> Mert
>
>
> On Thu, May 14, 2009 at 10:52 AM, Eric H. Lee <ericlee_at_ks.uiuc.edu
> <mailto:ericlee_at_ks.uiuc.edu>> wrote:
>
> As Peter stated, you will have to generate a separate psf file
> without waters as input for the new energy calculations when you
> reprocess the dcd through namd.
>
> -Eric
>
> On May 14, 2009, at 2:01 AM, Mert Gür wrote:
>
>> Dear Peter,
>> As you can see below (for the first frame ) bond-angle-dihedrals
>> and impropers are relatively close values. But when I compare the
>> electrical and vdw energies I see that they are at least one order
>> different.
>> But they seem to be close to the energy of the whole system.
>> The only explanation I can think of is that my script includes the
>> positions of the water molecules (In its coordinates at the
>> beginning ) to all frames
>>
>> My 1EJG-prot.dcd can't include water molecules cause it is only
>> 10mb whereas 1EJG.dcd is 100mb.
>> Therefore I am guessing that I am doing something wrong in the script.
>>
>> Best,
>> Mert
>>
>>
>>
>> For example in linux I obtained
>>
>> ETITLE: TS BOND ANGLE
>> DIHED IMPRP ELECT VDW
>> BOUNDARY MISC KINETIC
>> TOTAL TEMP TOTAL2 TOTAL3
>> TEMPAVG PRESSURE
>> ENERGY: 3039000 126.4302 379.9730
>> 174.6389 23.0378 -57317.3079 5216.9948
>> 0.0000 0.0000 10013.9117 -41382.3216
>> 301.2902 -41358.1629 -41358.1629 301.2902
>> -976.4120 -943.0767 178158.8790 -976.4120 -943.0767
>>
>> whereas in windows I have obtained (which is also close to
>> mdenergy's result)
>>
>> Frame Time Bond Angle
>> Dihed Impr Elec VdW
>> Conf Nonbond Total
>> 0 0 +126.308 +405.648
>> +188.96 +17.5462 -476.427 -122.028
>> +738.463 -598.456
>>
>>
>> On Wed, May 13, 2009 at 11:55 PM, Peter Freddolino
>> <petefred_at_ks.uiuc.edu <mailto:petefred_at_ks.uiuc.edu>> wrote:
>>
>> Hi Mert,
>> I see... I was confused by the fact that your psf file was named
>> "1EJG_wb.psf", which seemed to indicate that it included the
>> water box.
>> How different are your energies from those of namdenergy/mdenergy?
>> Best,
>> Peter
>>
>> Mert Gür wrote:
>> > Dear Peter,
>> > I generated 1EJG-prot.dcd (Which contains only the protein
>> coordinates)
>> > using catdcd from my dcd file as it was told in the tutorial.
>> > So according to the tutorial (section 2.5) I should be able
>> to read this
>> > coordinates of protein only, for each frame and then run it
>> for 0 time
>> > steps. Hence I will end up with each frames protein energy.
>> > That should be what tells namd that I want the energy of the
>> protein.
>> > That is what I have understand .I may be mistaken.
>> > I will take a look at the pairinteractions command but there
>> was nothing
>> > mentioned about it in that section.
>> >
>> > Best,
>> > Mert.
>> >
>> > On Wed, May 13, 2009 at 8:03 PM, Peter Freddolino
>> <petefred_at_ks.uiuc.edu <mailto:petefred_at_ks.uiuc.edu>
>> > <mailto:petefred_at_ks.uiuc.edu <mailto:petefred_at_ks.uiuc.edu>>>
>> wrote:
>> >
>> > Hi Mert,
>> > there doesn't appear to be anything in your config file
>> that *tells*
>> > namd that you want only the energy of the protein; how
>> is it supposed to
>> > know? Have a look at the pairInteractions command...
>> > Best,
>> > Peter
>> >
>> > Mert Gür wrote:
>> > > Hi everyone,
>> > > I am a quite new user of Linux. I have performed a
>> number of MD
>> > > simulations. What I am searching for is the energy of
>> only the
>> > protein.
>> > > I have read the linux tutorial of namd and also read
>> the corresponding
>> > > e-mails in this list.
>> > > My problem is that the way I have written my
>> get-energy.conf file
>> > either
>> > > does not work properly or does not give me the energy
>> of only the
>> > protein.
>> > > I came to this conclusion by comparing the energies
>> both with the
>> > ones
>> > > I have obtained via vmd's namdenergy plugin (in
>> windows) and via
>> > mdenergy
>> > >
>> > > For example in linux I obtained
>> > >
>> > > ETITLE: TS BOND ANGLE
>> DIHED
>> > > IMPRP ELECT VDW BOUNDARY
>> > > MISC KINETIC TOTAL TEMP
>> > > TOTAL2 TOTAL3 TEMPAVG PRESSURE
>> > > ENERGY: 3039000 126.4302 379.9730
>> 174.6389
>> > > 23.0378 -57317.3079 5216.9948 0.0000
>> > > 0.0000 10013.9117 -41382.3216 301.2902
>> > > -41358.1629 -41358.1629 301.2902
>> -976.4120
>> > > -943.0767 178158.8790 -976.4120 -943.0767
>> > >
>> > > whereas in windows I have obtained (which is also
>> close to mdenergy's
>> > > result)
>> > >
>> > > Frame Time Bond Angle
>> Dihed
>> > > Impr Elec VdW Conf
>> Nonbond
>> > > Total
>> > > 0 0 +126.308 +405.648
>> +188.96
>> > > +17.5462 -476.427 -122.028 +738.463
>> -598.456
>> > >
>> > > I attached my file in case you are interested. I used
>> 1EJG crambin
>> > in my
>> > > simulation so the name of the file is correspondingly
>> different.
>> > > Any help or even an example ..-get-energy.conf file
>> would be
>> > appreciated.
>> > > Mert
>> > >
>> > >
>> > >
>> > > structure ../common/1EJG_wb.psf
>> > > coordinates ../common/1EJG_wb.pdb
>> > > binCoordinates ../run4/res.coor
>> > > binVelocities ../run4/res.vel
>> > > extendedSystem ../run4/res.xsc
>> > >
>> > >
>> > > set temperature 300
>> > > set outputname 1EJG
>> > > set restartname res
>> > >
>> > >
>> > > firsttimestep 3039000
>> > > # Input
>> > > paraTypeCharmm on
>> > > parameters ../common/par_all27_prot_lipid.inp
>> > >
>> > > # Force-Field Parameters
>> > > exclude scaled1-4
>> > > 1-4scaling 1.0
>> > > cutoff 12.
>> > > switching on
>> > > switchdist 10.
>> > > pairlistdist 13.5
>> > >
>> > >
>> > > # Integrator Parameters
>> > > timestep 2.0
>> > > rigidBonds all
>> > > nonbondedFreq 1
>> > > fullElectFrequency 2
>> > > stepspercycle 10
>> > >
>> > >
>> > > # Constant Temperature Control
>> > > langevin on
>> > > langevinDamping 5
>> > > langevinTemp $temperature
>> > > langevinHydrogen off
>> > >
>> > >
>> > > ## Periodic Boundary Conditions
>> > > cellBasisVector1 57.9 0. 0.
>> > > cellBasisVector2 0. 53.7 0.
>> > > cellBasisVector3 0. 0. 57.3
>> > > cellOrigin 11.2 10.0 6.3
>> > > wrapAll on
>> > >
>> > > # PME (for full-system periodic electrostatics)
>> > > PME yes
>> > > PMEGridSizeX 60
>> > > PMEGridSizeY 54
>> > > PMEGridSizeZ 60
>> > >
>> > >
>> > >
>> > > # Output
>> > > outputName $outputname
>> > > restartName $restartname
>> > > outputEnergies 500
>> > > outputPressure 500
>> > > restartfreq 500
>> > > dcdfreq 500
>> > > xstFreq 500
>> > >
>> > > # runs 0 time steps for each frame in the dcd
>> > > # opens the dcd file to read the coordinates
>> > > coorfile open dcd 1EJG-prot.dcd
>> > > set i 0
>> > > while { ![coorfile read] } {
>> > > incr i 500
>> > > firsttimestep $i
>> > > run 0
>> > > }
>> > > coorfile close
>> >
>> >
>>
>>
>
> Eric H. Lee
>
> Medical Scholars Program
>
> Theoretical and Computational Biophysics Group, UIUC
>
> ericlee_at_ks.uiuc.edu <mailto:ericlee_at_ks.uiuc.edu>
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:58 CST