Fwd:Re: ABF-Zcoord:

From: ΑυΣύ (supram_at_nankai.edu.cn)
Date: Wed Feb 04 2009 - 22:04:40 CST

    I am very grateful for your help, thank you very much!!!
    I checked my input files for free energy calculation with the ABF method and
ran it for about 1ns.The result seems not reasonable(larger than the real free energy along the recation coordinate-¦Ζ).
    I focus on the method of restraining the L-glutamic acid to make it move along the Z axis. If I make no restraint to the L-glutamic acid, it can not move along the Z axis. In the L-glutamic acid, all the other carbon and nitrogen atoms are connected to hydrogen atoms except the carbon and oxygen atoms in the carboxyl at the two end of the L-glutamic acid. I have made "rigidbonds all". The abf coordinate is zCoord.I define the glycosidic oxygen atoms of ¦Β-CD as the reference atoms and "fixed" them as to restrain their mass center at the origin coordinate(O cooridinate). I define the four oxygen atoms in the carboxyl at the two end of the L-glutamic acid as the RC(recation coordinate)-¦Ζ.I have made the four RC oxygen atoms free of any restraint of freedoms. I also have made harmonic positional restraints to the other atoms of L-glutamic acid(including carbon,oxygen and hydrogen atoms) to make the L-glutamic acid move along the Z axis.
    my question is: is there some clash which affects the free energy calculation with ABF method? I mean the clash is between the restraints(to the other atoms of L-glutamic acid,including carbon,oxygen and hydrogen atoms) which makes the L-glutamic acid move along the Z axis) and the free energy calculation with ABF method.
    I also want to ask another question: is there some other proper methods in restraining the atoms of L-glutamic acid (to make it move along the Z axis)
and not affect the free energy calculation with ABF method?
    Any suggestion will be helpful to me. Thank you very much in advance !!!!!!!

>Yes, your approach seems reasonable (given the information you gave),
>but as in all free energy calculations, and simulations in general,
>certain results are not guaranteed. Specifically, only people who
>know the specific system you're simulating can appreciate your choice
>of reaction coordinate and the set of constraints you included.
>
>If this approach fails, there may be other technical ways to do what
>you're planning. For instance, you may want to have a look at the
>collective variables module in the CVS version, that implements ABF on
>a broader choice of reaction coordinates and can do multidimensional
>sampling as well. If you see that zCoord is insufficient, you may add
>x and y in the sampling rather than constraining them.
>
>But basically, if a given reaction coordinate doesn't work, I'm afraid
>you're mostly on your own to understand why.
>
>Giacomo
>
>
>---- -----
> Giacomo Fiorin
> Center for Molecular Modeling at
> University of Pennsylvania
> 231 S 34th Street, Philadelphia, PA 19104-6323
> phone: (+1)-215-573-4773
> fax: (+1)-215-573-6233
> mobile: (+1)-267-324-7676
> mail: giacomo.fiorin_
_gmail.com
> web: http://www.cmm.upenn.edu/
>---- ----
>
>> Dear Namd developers,
>> I am running a free energy calculation with ABF method for the
>> complex of ¦Β-CD
>> with L-glutamic acid (Glu) in waters. The abf coordinate is zCoord. When
>> running the free energy calculation, the the abf abf1 { 14 35 56 77 98 119
>> 140 }oxygens atoms of the ¦Β-CD are fixed; and I want to constraint the abf
>> abf2 { 149 150 164 165 } oxygens atoms of the L-glutamic acid on the XY plan
>> to make sure the L-glutamic acid (Glu)moving along the Z axis.
>> Is my method reasonable? Is there other methods to make sure that the
>> L-glutamic acid (Glu)moving along the Z axis?
>> I will very be deeply grateful for your help, thank you very much!
>> Here are my config file for the ABF simulation.
>>
>> structure bcdlglu_wb_na.psf
>> coordinates restart.coor
>> velocities restart.vel
>> extendedsystem restart.xsc
>> firsttimestep 1010000
>>
>> paraTypeCharmm on
>> parameters par_all22_prot.inp
>> parameters CSFF_parm.inp
>>
>> exclude scaled1-4
>> 1-4scaling 1.0
>> cutoff 14.0
>> switching on
>> switchdist 12.0
>> pairlistdist 16.0
>>
>> timestep 1.0
>> rigidbonds all
>> nonbondedFreq 1
>> fullElectFrequency 2
>> stepspercycle 20
>>
>> langevin on
>> langevinDamping 1
>> langevinTemp 298.15
>> langevinHydrogen on
>>
>> useGroupPressure yes
>> useFlexibleCell no
>> useConstantArea no
>>
>> langevinPiston on
>> langevinPistonTarget 1
>> langevinPistonPeriod 100
>> langevinPistonDecay 50
>> langevinPistonTemp 298.15
>>
>> wrapwater on
>> wrapall on
>>
>> PME on
>> PMEGridSizeX 36
>> PMEGridSizeY 36
>> PMEGridSizeZ 60
>>
>> fixedatoms on
>> fixedatomsfile bcdlglu_wb_na_f.pdb
>> fixedatomscol B
>>
>> constraints on
>> consref restart.coor
>> conskfile bcdlglu_wb_na_h.pdb
>> conskcol B
>> selectconstraints on
>> selectconstrx on
>> selectconstry on
>> selectconstrz off
>>
>> outputName output-ABF
>> binaryoutput no
>> outputEnergies 1000
>> outputTiming 1000
>> outputPressure 1000
>> dcdfile bcdlglu_wb.dcd
>> dcdfreq 1000
>> dcdunitcell yes
>> xstfreq 1000
>> restartname restart-1
>> binaryrestart no
>> restartfreq 1000
>>
>> source /nfs/s09r2p2/yuliu_nk/NAMD2.6-1/Abf/abf.tcl
>>
>> abf coordinate zCoord
>>
>> abf abf1 { 14 35 56 77 98 119 140 }
>>
>> abf abf2 { 149 150 164 165 }
>>
>> abf ximin -8
>>
>> abf ximax -4
>>
>> abf dxi 0.1
>>
>> abf fullsamples 1000
>>
>> abf dsmooth 0.1
>>
>> abf forceconst 20
>>
>> abf applybias yes
>>
>> abf outputfreq 1000
>>
>> abf writexifreq 1000
>>
>> abf outfile bcdlglu.abf
>>
>> run 4000000
>>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:19 CST