[gmx-users] protein / bilayer problems

[Marc Baaden]

Hi ,

I am not sure about your atom name problem, I'd guess something
is wrong with the topology. But as you said you managed to make
it go away by changing the order, so that should be fine.


jxs818 at bham.ac.uk said:
>>  Fatal error: Invalid order for directive atomtypes,  file
>> ""lipid.itp"", line 11 

You should not be using lipid itp. The "proper" way is to incorporate
the additional forcefield parameters into the forcefield that
you are using, eg creating a set ffgmx_lipid.itp, ffgmx_lipidnb.itp
and ffgmx_lipidbon.itp and use that.

We have done that here for ffgmx and ffG43a1.

This works fine.

Now, I know several people worked hard on grompp to make the problems
go away (Berk Hess did a lot of testing/improving), so maybe you can
get it to work somehow even using lipid.itp. But the above method is
I would say the foolproof one.

Check the mailing list archives for more details on what the problems
are with included parameter definitions.

BTW: for the gmx-developers/maintainers .. should we incorporate a
      merged forcefield+lipid itp set into the distribution ? The
      problem seems to come up quite often .. (?)

Cheers,
Marc


-- 
 Dr. Marc Baaden - Laboratory of Molecular Biophysics, Oxford University
 mailto:baaden at smplinux.de  - ICQ# 11466242 -  http://www.marc-baaden.de
 FAX/Voice +49 697912 39550  -  Tel: +44 1865 275380  or  +33 609 843217