[gmx-users] Problems with g_anaeig!

[Semen Esilevsky]

Dear gmx users,
I've found a strange problem in g_anaeig.
I'm trying to project the trajectory into few first
eigenvectors. For this I first do:

g_covar -f md_all.xtc -s md0.tpr -n index.ndx -fit
-ref -pbc -tu ns 

Chosing CA group for matrix construction.
Then I pick the resulting file eigenvec.trr and do:

g_anaeig -v eigenvec.trr -f md_all.xtc -s md0.tpr
-filt -n index.ndx -first 1 -last 10

Everything works fine and the output file filtered.xtc
should contain nice filtered trajectory. Ok, let's
test:

g_covar -f filtered.xtc -s md0.tpr -n index.ndx -fit
-ref -pbc -tu ns 
...
...
...
Calculating the average structure ...
Reading frame       0 time    0.000  
-------------------------------------------------------
Program g_covar, VERSION 3.3.1
Source code file: mshift.c, line: 118

Fatal error:
Molecule in topology has atom numbers below and above
natoms (238).
You are probably trying to use a trajectory which does
not match the first 238 atoms of the run input file.
You can make a matching run input file with tpbconv.
-------------------------------------------------------

What is wrong? It should be the same trajectory as
initial md_all.xtc but with different numbers.
Any ideas?

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com