From: Udaya Dahal (dahal.udaya_at_gmail.com)
Date: Mon Feb 25 2019 - 16:28:18 CST

The problem seems to be resolved in the newer version of VMD. The version I
was using that showed error was (1.9.1) and when I used 1.9.3, the problem
seems resolved. Anyone using new version should be fine :)

-Udaya

On Mon, Feb 25, 2019 at 4:44 PM Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov>
wrote:

> Yes, I've seen this, and it has driven me crazy looking to debug this on
> occasion. What are the flags at the top of the PSF? If I remember
> correctly, it is read correctly if the NAMD-flag is there (space-delimited
> reads), but not if it only uses the EXT flag (width-delimited reads).
> Unfortunately, VMD doesn't always get the logic right when writing out psf
> files with long atomtypes or atomnames, particularly if you write out psf
> files that have longer than usual atomtypes.
>
> -Josh
>
>
>
> On 2019-02-25 11:58:49-07:00 owner-vmd-l_at_ks.uiuc.edu wrote:
>
> Hi,
>
> Is there any restriction on PSF file format for the atom definitions and charge that is being extracted by VMD? What I figured out is that when the atomtype is longer (for. eg. in CHARMM-cgenff some atomtypes are HGPAM2), VMD is extracting wrong value of the charge. In the following line in atom section of psf:
>
> 986 PH1 2 PAHN HZ1 HGPAM2 0.340000 1.0080 0
>
> Instead of reading charge as 0.34 it is reading as 2. Since, for very big systems when we are trying to see charge distribution it is very likely that we would be overlooking these sort of error. Has anyone encountered this sort of error before or is there any specific character space for psf file? I generated this psf file using VMD.
>
> Thank you
> Udaya Dahal
>
>
>