VMD-L Mailing List

From: Kevin C Chan (cchan2242-c_at_my.cityu.edu.hk)
Date: Tue Sep 15 2015 - 08:53:57 CDT

Dear Users,

I am struggling with the autopsf plugin.

Sometimes we are in need of feeding some user-defined topologies into
Autopsf. I often work with the command-line interface and the -top command
bothered me. I have read through the tcl code and I fed topologies
separated by "space"

autopsf -top "top1.str top2.str top3.str"

But it still gives me things like

psfgen) building segment O1
psfgen) reading residues from pdb file system_psfgen_autopsf-temp.pdb_O1.pdb
psfgen) unknown residue type DOPC

And of course residue DOPC has been well defined in top1/2/3.str.

In addition to solving -top problem of autopsf, here is my very original
intention:

I have a rather complicated system: a protein consists of S-S bond and
membrane composed of some lipids with self-defined topology.

I have three problems for different approaches on smoothly generating a psf
for the system:
1. My autopsf could not read too many topology files (described above)
2. I could write psfgen tcl codes for the membrane but I found it very
difficult to integrate the automatic S-S bond detection part into a simple
psfgen code
3. GUI version of autopsf is rather buggy, I feed the topology files and it
crashes when reading.
4. I would like to try autopsf reading one single topology file instead of
multiple but I do not know how to correctly combine them. I am using
CHARMM36ff and I would like to combine rtf and some str files.

Thanks in advance for any experience sharing. Reply to any of the above
question would be very helpful.

Regards,
Kevin