[gmx-users] Problems

[David]

On Mon, 2004-05-03 at 15:58, Henrik Rundgren wrote:
> Hello again. Yes I did change the number of entries at the top of the 
> aminoacids.dat file. I figured that the simulation should run anyway and 
> that there shouldn't be problems due to the fact that the topology was 
> ok (looks like it). I added the combinations that wasn't made by 
> pdb2gmx..... and started to simulate.. I have done EM and a couple of 
> short MD's (2 X 10000 steps) to stabalize the box size and density 
> a.s.o.... and to see that it works... and it did. I only collect the 
> protein coords in the .XTC file, and it only records the atoms that are 
> included as "Protein" Residues by Grompp.... I really must get grompp to 
> understand that all atoms in the *.top should be included as 

you can force it by manually making an index file.

> "Protein".... I have tried to change name in the topology file, for 
> instance from MELE to MELEU (I shortened the length of the res.names in 
> *.pdb so that the "chain identifier" wouldn't go on)... but this didn't 
> change the outcome.. so apparently grompp doesn't need to have the same 
> names in *.top as in *.rtp files... Apparently it only cares that the 
> residues are in aminoacids.dat to include that residue as a "Protein".. 
Are you sure it is using the correct aminoacids.dat file? (if you have
one in your local directory it will be preferred over the one in the
library directory.

> But this didn't work...
> 
> There is a problem when mdrun was ran, following error occured.. after about 81000 steps the following happened.
>  
> 1.	Warning: 1-4 interaction at distance larger than 2.56
> 	These are ignored for the rest of the simulation
> 	turn on -debug for more information
> 	Segmentation fault
> 
> It's cyclic... atleast when I check the *.gro file.... and the short simulations ..... Maybe just something went baserk in it...
> Does this segmentation fault have anything to do with the above mentioned problem? 

If you have problems due to large forces the program can SEGV. Do you
update the neighborlist often enough?


-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++