AMBER Archive (2004)

Subject: AMBER: nmanal - segmentation fault

From: Sivakolundu, Sivashankar (Sivashankar.Sivakolundu_at_stjude.org)
Date: Fri Mar 26 2004 - 00:11:16 CST

Hi amber users,
I've been trying to do normal mode analysis for my protein which has 1150 atoms using quasih and nmanal. quasih ran the superimposed trajectory without any errors and generated 50 eigenvectors. When I tried to run it through nmanal, it seems to do something for few seconds and then throws a segmentation fault. The output file doesnt have anything in it. It seems to read the topology file fine since when set the ntrun=6 to calculate RMS fluctuations the execution went to completion without error and the atom numbers in the output file were correct. When I set the ntrun to values
1,2, or 3, it gives seg fault. The nmanal command I ran was

nmanal -O -i nmanal2.in -o nmanal.out -v vecs -m mass -p topology -proj proj

The vecs file was generated by quasih; mass file by pdbtomass; and the topology file was in old format. The control file had the following:

#
 &data
  ntrun=1,
  nvect=50,
  ibeg=1, iend=50,
  ihsful=1,
 &end

The computer I use is an SGI with 8GB available memory. I've increased the sizes in sizes2.h based on required sizes(error messages given when I ran with default sizes).
Since it doesnt give any other error messages I dont know if the info I've given above is enough to do any troubleshooting. Please let me know if you need any other information to get an idea of what the problem is.

Any help is greatly appreciated.

Thank you very much

Siva Sivakolundu
Postdoc, Struc. Biol.
SJCRH, Memphis,TN

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu