The real test is "are the forces and energies you get different". If I
recall correctly, in this case the differences you refer to are from the
addition of "placeholder" end dihedrals (i.e. Dihedrals with 0.0 force
constant) which were previously missing. I think the reason was to prevent
so-called "wild card" dihedrals from being improperly applied (others
please correct me if I'm wrong here). The bottom line is that the energies
obtained with the old topology and new should be identical for the same
coordinates. Please let us know if this is not the case.
-Dan
On Tue, Jan 31, 2017 at 3:34 PM Joao Damas <j.damas.acellera.com> wrote:
> Hi guys,
>
> We wanted to change from AmberTools15 to AmberTools16, and we saw in the
> changelog (http://ambermd.org/changelog.html) about the new specifications
> for choosing force-fields.
>
> For now, we were at least testing if oldff/leaprc.ff14SB in AmberTools16
> gave the same results as the leaprc.ff14SB in AmberTools15.
>
> As a matter of fact, it does not. The initial difference that caught my
> attention was:
>
> 591c596
> < AMBER General Force Field for organic molecules (Version 1.7, Nov 2013)
> ---
> > AMBER General Force Field for organic molecules (Version 1.8, Mar 2015)
>
> So I changed to use the gaff.dat from AmberTools15.
>
> After that change, the entire diff between leap.log files using
> AmberTools15 and AmberTools16 is:
>
> 1c1
> < log started: Tue Jan 31 20:21:26 2017
> ---
> > log started: Tue Jan 31 21:15:45 2017
> 48c48
> < Loading parameters:
> /home/joao/maindisk/software/amber14/dat/leap/parm/lipid14.dat
> ---
> > Loading parameters:
> /home/joao/maindisk/software/amber16/dat/leap/parm/lipid14.dat
> 55c55
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/lipid14.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/lipid14.lib
> 65,67c65,67
> < > source leaprc.ff14SB
> < ----- Source:
> /home/joao/maindisk/software/amber14/dat/leap/cmd/leaprc.ff14SB
> < ----- Source of
> /home/joao/maindisk/software/amber14/dat/leap/cmd/leaprc.ff14SB done
> ---
> > > source oldff/leaprc.ff14SB
> > ----- Source:
> /home/joao/maindisk/software/amber16/dat/leap/cmd/oldff/leaprc.ff14SB
> > ----- Source of
> /home/joao/maindisk/software/amber16/dat/leap/cmd/oldff/leaprc.ff14SB done
> 69c69
> < log started: Tue Jan 31 20:21:26 2017
> ---
> > log started: Tue Jan 31 21:15:45 2017
> 169c169
> < Loading parameters:
> /home/joao/maindisk/software/amber14/dat/leap/parm/parm10.dat
> ---
> > Loading parameters:
> /home/joao/maindisk/software/amber16/dat/leap/parm/parm10.dat
> 175c175
> < Loading parameters:
> /home/joao/maindisk/software/amber14/dat/leap/parm/frcmod.ff14SB
> ---
> > Loading parameters:
> /home/joao/maindisk/software/amber16/dat/leap/parm/frcmod.ff14SB
> 183c183
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/amino12.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/amino12.lib
> 213c213
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/aminoct12.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/aminoct12.lib
> 241c241
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/aminont12.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/aminont12.lib
> 267c267
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/nucleic12.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/nucleic12.lib
> 305c305
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/atomic_ions.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/atomic_ions.lib
> 374c374
> < Loading library:
> /home/joao/maindisk/software/amber14/dat/leap/lib/solvents.lib
> ---
> > Loading library:
> /home/joao/maindisk/software/amber16/dat/leap/lib/solvents.lib
> 376a377,380
> > Loading: FB3
> > Loading: FB3BOX
> > Loading: FB4
> > Loading: FB4BOX
> 495a500
> > >>
> 498,499c503,504
> < ----- Source:
> /home/joao/maindisk/software/amber14/dat/leap/cmd/leaprc.gaff
> < ----- Source of
> /home/joao/maindisk/software/amber14/dat/leap/cmd/leaprc.gaff done
> ---
> > ----- Source:
> /home/joao/maindisk/software/amber16/dat/leap/cmd/leaprc.gaff
> > ----- Source of
> /home/joao/maindisk/software/amber16/dat/leap/cmd/leaprc.gaff done
> 501c506
> < log started: Tue Jan 31 20:21:27 2017
> ---
> > log started: Tue Jan 31 21:15:45 2017
> 589c594
> < Loading parameters:
> /home/joao/maindisk/software/amber14/dat/leap/parm/gaff.dat
> ---
> > Loading parameters: ./gaff.dat
> 596c601
> < Loading parameters:
> /home/joao/maindisk/software/amber14/dat/leap/parm/frcmod.ionsjc_tip3p
> ---
> > Loading parameters:
> /home/joao/maindisk/software/amber16/dat/leap/parm/frcmod.ionsjc_tip3p
>
> If one excludes the differences in the paths, the only difference is in the
> loading of news solvents (I think? from solvents.lib). These shouldn’t
> affect.
>
> The output .crd is the same, no differences. But the output .prmtop files
> are different.
>
> Namely,
>
> - there are differences in the POINTERS section, in the 8th and 15th
> entry, which correspond to the MPHIA (number of dihedrals not containing
> hydrogen) and NPHIA (MPHIA + number of constraint dihedrals). The
> difference is between 7371 and 7373, with AmberTools16 presenting two
> more:
>
> 7,8c7,8
> < 7196 17 5555 1670 3625 2269 7454 7371
> 0 0
> < 23114 1552 1670 2269 7371 68 160 213
> 35 1
> ---
> > 7196 17 5555 1670 3625 2269 7454 7373
> 0 0
> > 23114 1552 1670 2269 7373 68 160 213
> 35 1
>
>
> - the previously mentioned entry differences also show in
> DIHEDRALS_WITHOUT_HYDROGEN:
>
> 2097 2106 -2115 2121 44
> 4383 4392 -4401 4407 44
>
>
> - lastly, the RADII section, there are multiple differences, where one
> radius that is 1.2 in AmberTools15, shows as 1.3 in AmberTools16.
> Sample of
> differences:
>
> 22541,22542c22542,22543
> < 1.70000000E+00 1.20000000E+00 1.70000000E+00 1.30000000E+00
> 1.30000000E+00
> < 1.30000000E+00 1.70000000E+00 1.20000000E+00 1.20000000E+00
> 1.70000000E+00
> ---
> > 1.70000000E+00 1.30000000E+00 1.70000000E+00 1.30000000E+00
> 1.30000000E+00
> > 1.30000000E+00 1.70000000E+00 1.30000000E+00 1.30000000E+00
> 1.70000000E+00
>
> The tleap.in file used was this one:
>
> # tleap file generated by amber.build
> source leaprc.lipid14
> source leaprc.ff14SB (CHANGED TO oldff/leaprc.ff14SB for AmberTools16)
> source leaprc.gaff
>
> # Loading parameter files
> loadamberparams frcmod.ionsjc_tip3p
>
> # Loading prepi topologies
>
> # Loading the system
> mol = loadpdb input.pdb
>
> # Adding disulfide bonds
> bond mol.22.SG mol.152.SG
> bond mol.40.SG mol.56.SG
> bond mol.124.SG mol.225.SG
> bond mol.131.SG mol.198.SG
> bond mol.163.SG mol.177.SG
> bond mol.188.SG mol.212.SG
>
> # Writing out the results
> saveamberparm mol structure.prmtop structure.crd
>
> The input.pdb from built from PDB 3PTB.
>
> I can share the files that I used and the outputs. Let me know the platform
> that you prefer for receiving the files.
>
> Is there any reason for these differences? We are trying to have a smooth
> transition from AmberTools15 to AmberTools16 and it would be very helpful
> to have your insight on these. Thank you.
>
> Best regards,
> João
>
> --
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
-------------------------
Daniel R. Roe
Laboratory of Computational Biology
National Institutes of Health, NHLBI
5635 Fishers Ln, Rm T900
Rockville MD, 20852
https://www.lobos.nih.gov/lcb
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jan 31 2017 - 17:30:02 PST
