Inconsistency in fast table energy vs force

From: Francesco Pietra (chiendarret_at_gmail.com)
Date: Sat Apr 13 2013 - 02:36:32 CDT

I would appreciate very much advice in interpreting the following error
appearing in the log file

Info: INCONSISTENCY IN FAST TABLE ENERGY VS FORCE: 0.000325096 AT 11.9556
Info: INCONSISTENCY IN SCOR TABLE ENERGY VS FORCE: 0.000324844 AT 11.9556
Info: INCONSISTENCY IN VDWA TABLE ENERGY VS FORCE: 0.0040507 AT 0.251946
Info: INCONSISTENCY IN VDWB TABLE ENERGY VS FORCE: 0.00150189 AT 0.251946

on running pressure equilibration (either ts=1.0 or 0.01fs) on a 45,821
atom system (metalloprotein in TIP3 box) with namd2.9-cuda/debian-GNULinux
amd64). No such error on minimizing (to grad 1.0) or heating to 298K. Also,
no such error in other, very similar systems, even along 15ns
equilibration). A couple of ligands is held in place by harmonic colvars,
which worked finely with very similar systems.

At each writing, VdW is compensated, so that the total energy is very
largely negative. No atom moving fast. However, the benchmark (sec/step) is
ca 8 times larger than for systems of similar size where no such error
occurs. At any event, that slowing down makes MD unmanageable. Notice that
params for heme-c are the same used for the other, regularly-behaving
systems.

>From the output of vmd -dispdev I noticed:

psfgen) Warning: failed to guess coordinate due to bad angle CD CG HG2
psfgen) Warning: failed to guess coordinate due to bad angle CD CG HG2
psfgen) Warning: failed to guess coordinate due to bad angle C4D CHA HA
psfgen) Warning: failed to guess coordinate due to bad angle C4C CHD HD
psfgen) Warning: failed to guess coordinate due to bad angle CGA CBA HBA2
psfgen) Warning: failed to guess coordinate due to bad angle CGD CBD HBD2
psfgen) Warning: failed to guess coordinate due to bad angle CD CG HG2
psfgen) Warning: failed to guess coordinate due to bad angle CD CG HG2
psfgen) Warning: failed to guess coordinate due to bad angle C4D CHA HA
psfgen) Warning: failed to guess coordinate due to bad angle C4C CHD HD
psfgen) Warning: failed to guess coordinate due to bad angle CGA CBA HBA2
psfgen) Warning: failed to guess coordinate due to bad angle CGD CBD HBD2
psfgen) Warning: poorly guessed coordinates for 26 atoms (4 non-hydrogen):
psfgen) Warning: poorly guessed coordinate for atom HT1 PHE:2 PRA
psfgen) Warning: poorly guessed coordinate for atom HT2 PHE:2 PRA
psfgen) Warning: poorly guessed coordinate for atom HT3 PHE:2 PRA
psfgen) Warning: poorly guessed coordinate for atom HG2 PRO:28 PRA
psfgen) Warning: poorly guessed coordinate for atom HG2 PRO:55 PRA
psfgen) Warning: poorly guessed coordinate for atom OT1 LYS:127 PRA
psfgen) Warning: poorly guessed coordinate for atom OT2 LYS:127 PRA
psfgen) Warning: poorly guessed coordinate for atom HA HEC:128 HEA
psfgen) Warning: poorly guessed coordinate for atom HB HEC:128 HEA
psfgen) Warning: poorly guessed coordinate for atom HC HEC:128 HEA
psfgen) Warning: poorly guessed coordinate for atom HD HEC:128 HEA
psfgen) Warning: poorly guessed coordinate for atom HBA2 HEC:128
HEA
psfgen) Warning: poorly guessed coordinate for atom HBD2 HEC:128
HEA
psfgen) Warning: poorly guessed coordinate for atom HT1 PHE:2 PRB
psfgen) Warning: poorly guessed coordinate for atom HT2 PHE:2 PRB
psfgen) Warning: poorly guessed coordinate for atom HT3 PHE:2 PRB
psfgen) Warning: poorly guessed coordinate for atom HG2 PRO:28 PRB
psfgen) Warning: poorly guessed coordinate for atom HG2 PRO:55 PRB
psfgen) Warning: poorly guessed coordinate for atom OT1 LYS:127 PRB
psfgen) Warning: poorly guessed coordinate for atom OT2 LYS:127 PRB
psfgen) Warning: poorly guessed coordinate for atom HA HEC:128 HEB
psfgen) Warning: poorly guessed coordinate for atom HB HEC:128 HEB
psfgen) Warning: poorly guessed coordinate for atom HC HEC:128 HEB
psfgen) Warning: poorly guessed coordinate for atom HD HEC:128 HEB
psfgen) Warning: poorly guessed coordinate for atom HBA2 HEC:128
HEB
psfgen) Warning: poorly guessed coordinate for atom HBD2 HEC:128
HEB

which refers to coordinates and angles with H-atoms added by dispdev, i.e.,
what is normally expected for zero occupancy. No error or warning reported.
No grossly detectable anomalies in the structure at the end of the press
run.

I wonder whether the INCONSISTENCY error can suggest where to better look
for bad features in the psf/pdb files.

Thanks
francesco pietra

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:08 CST