From: Gumbart, JC (gumbart_at_physics.gatech.edu)
Date: Thu Nov 12 2020 - 22:54:08 CST

 So does the problem mainly arise from these modifications that are probably being reflected in the log files ?

Indeed - one of the water molecules has impossible coordinates and at some point, clashes with an atom of your molecule, crashing the energy calculation. If it’s not converging you should investigate why - whether it’s too slow or some other cause. As much as we tried to automate tedious steps in FFTK, there’s no avoiding careful examination, e.g., of the individual water interaction calculations.

Best,
JC

On Nov 12, 2020, at 11:36 PM, Nibedita Ray Chaudhuri <nivirc.143_at_gmail.com<mailto:nivirc.143_at_gmail.com>> wrote:

I have used FFTK to generate the input gau files but I had to make slight modifications in the input format before submitting the gaussian jobs since they were not initially converging. For example:
1. Modified the coordinate input format from Z-matrix to the standard Redundant IC format of Gaussian calculation.
2. Used opt=calcfc to calculate the force constants once at the beginning of the calculation.
The charge optimization step in FFTK requires the gaussian log files. So does the problem mainly arise from these modifications that are probably being reflected in the log files ?

Nibedita

On Fri, Nov 13, 2020 at 3:16 AM Gumbart, JC <gumbart_at_physics.gatech.edu<mailto:gumbart_at_physics.gatech.edu>> wrote:
Thanks for sharing your files. Something is wrong with your gaussian input file, namely NG2-ACC-N2.gjf. It does not appear you used FFTK to generate it.

Best,
JC

On Nov 12, 2020, at 12:48 AM, Gumbart, JC <gumbart_at_physics.gatech.edu<mailto:gumbart_at_physics.gatech.edu>> wrote:

Hi Nibedita,

Did the tutorial work for you? Can you try to narrow down what is different? If you can’t figure it out, you can send me files off list to try to reproduce it.

Best,
JC

On Nov 6, 2020, at 6:22 AM, Nibedita Ray Chaudhuri <nivirc.143_at_gmail.com<mailto:nivirc.143_at_gmail.com>> wrote:

Hello,
I am trying to parameterize a small molecule following the fftk tutorial using the fftk plugin of VMD version1.9.3, but I am getting the following error in the Opt. Charges step:

domain error: argument not in valid range
domain error: argument not in valid range
    while executing
"expr { sqrt($epsA * $epsH) * ( pow(($rminA+$rminH)/$dH1,12) - 2.0*pow(($rminA+$rminH)/$dH1,6) ) + sqrt($epsA * $epsH) * ( pow(($rminA+$rminH)/$dH2..."
    (procedure "::ForceFieldToolKit::ChargeOpt::computeIntE" line 46)
    invoked from within
"::ForceFieldToolKit::ChargeOpt::computeIntE $i $j"
    ......................................
   ......................................

I have been following the earlier threads but I am not sure what is the specific problem in my case. I tried to do the same in the new version of VMD 1.9.4a48 but this time it shows:

ERROR) No molecules loaded.mol addfile operates on one molecule only
ERROR) No molecules loaded.mol addfile operates on one molecule only
    while executing
"mol addfile $pdbPath waitfor all $cmpdMolID"
    (procedure "::ForceFieldToolKit::ChargeOpt::optimize" line 88)
    invoked from within
"::ForceFieldToolKit::ChargeOpt::optimize"
    (procedure "::ForceFieldToolKit::gui::coptRunOpt" line 55)
    invoked from within
"::ForceFieldToolKit::gui::coptRunOpt "
    invoked from within
".fftk_gui.hlf.nb.chargeopt.runOpt invoke "
    invoked from within
".fftk_gui.hlf.nb.chargeopt.runOpt instate {pressed !disabled} { .fftk_gui.hlf.nb.chargeopt.runOpt state !pressed; .fftk_gui.hlf.nb.chargeopt.runOpt in.."
    (command bound to event)

I checked my PSF file and apparently, everything seems fine. Is there any way this problem can be resolved without having to edit the plugin code?

I am pretty lost here. So, any suggestions will be highly appreciated.

I am attaching the debugged log file. Please let me know if any more information is needed

Nibedita
<Charge-Opt.debug.log>