Dear All,
I've been trying to undergo a multiple conformation one stage resp
fitting of four different geometries of 99 atom heme model I'm
studying. I've done several single conformation resp procedures to
each geometry with success. I've also done some multiple conformation
resp fittings to smaller models with success.
I always write the .in and .qin by hand and I'm assured they respect
the strict format required. The .dat (.esp) file is done with cat
command. I simply concatenate by order the esp containing .dat files
for each geometry, which are obtained with espgen over the .log
gaussian 03 output.
When I run the command $AMBERHOME/bin/resp -O -i xxx.in -o xxx.out -q
xxx.qin -e xxx.dat I get the following error:
At line 837 of file _resp.f (unit = 10, file = 'xxx.dat')
Fortran runtime error: Bad value during integer read
I've checked the _resp.f code line and it contains:
read(10,'(2i5)') inat,nesp
Since this line is from the following program section:
----------------------------------------------------------------------
c read in the electrostatic potential points used in the fitting,
c building up as we go the matrices for LU decomposition
c
c called from Main
----------------------------------------------------------------------
I can only suspect the problem is with the format of the esp contained
in the .dat file. The thing that amazes me the most is that when esp
in this format is used for a single conformation resp it works!! When
it's concatenated it doesn't. Smaller molecules work just fine when
concatenated... It's really confusing. This leads me to suspect that
the file size is prohibitive, but if that's the case why isn't there
anything written about this on the manual or something. Since I have
99 atoms and I'm calculating the esp with B3LYP/Gen (6-31G(d) and
LANL2TZ(f)) with SCF=Tight, Pop(MK,ReadRadii,Full) and
IOp(6/33=2,6/41=10,6/42=17) I might have too much points too compute
with resp... But again I'm just guessing!
Any feedback would be greatly appreciated. I'm really going insane
with this problem. Programs within the AmberTools package should
seriously include better error reporting.
Thanks for the patience. If you need any files to evaluate the problem
with more detail, I'll gladly supply them. I just want this problem
solved ASAP.
Best Regards,
João Henriques
Faculdade de Ciências da Universidade de Lisboa
Departamento de Química e Bioquímica,
Campo Grande, 1749-016 Lisboa, Portugal
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 28 2010 - 09:30:03 PDT
