On Wed, Jun 25, 2008, In Hee Park wrote:
>
>
> $ ambpdb -p prmtop < inpcrd > pdb
> | New format PARM file being parsed.
> | Version = 1.000 Date = 05/22/06 Time = 12:10:21
> At line 349 of file _ambpdb.f
> Fortran runtime error: Bad value during integer read
Can you check to see if the prmtop file has a "RADII" section? It looks like
ambpdb is getting confused because the lack of a radii section is causing it
to think this is an oldstyle prmtop file, but it's not. Or, maybe there is
something else that is leading to this confusion.
Anyway, please try the following patch to amber10/src/etc/ambpdb.f:
--- ambpdb.f 16 Mar 2008 20:58:47 -0000 9.4
+++ ambpdb.f 25 Jun 2008 20:44:57 -0000
.. -227,7 +227,7 ..
SUBROUTINE GETNAM(NATOM,NRES,IGRAPH,IPRES,LBRES,ITITL,NF,KPF,
1 X,IX,IH,ib,jb,nbond,chg,radius,lastat,ter,hasradii)
character(len=4) igraph,lbres,ititl
- logical ter,hasradii
+ logical ter,hasradii,oldstyle
DIMENSION IGRAPH(*),IPRES(*),LBRES(*),ITITL(20)
DIMENSION X(*),IX(*),IH(*),ib(*),jb(*),chg(*),lastat(*),radius(*)
CHARACTER*80 FMT
.. -236,6 +236,7 ..
AFMT = '(20A4)'
RFMT = '(5E16.8)'
hasradii = .false.
+ oldstyle = .false.
C
C ----- READ THE MOLECULAR TOPOLOGY -----
C
.. -243,6 +244,7 ..
TYPE = 'TITLE'
CALL NXTSEC(NF, 0, 0,FMTIN, TYPE, FMT, IOK)
READ(NF,FMT) (ITITL(I),I=1,20)
+ if( iok.eq.-1 ) oldstyle = .true.
c
FMTIN = IFMT
TYPE = 'POINTERS'
.. -269,7 +271,7 ..
chg(i) = chg(i)/18.2223
end do
c
- if( iok.eq.-1 ) then ! this is an old-style prmtop file
+ if( oldstyle ) then
c
READ(NF,40) (X(I),I = 1,NATOM)
READ(NF,30) (IX(I),I = 1,NATOM)
.. -341,7 +343,7 ..
c --- prmtop file will have information needed to generate where the
c TER cards in the output should be:
c
- if( iok.eq.-1 ) then ! old-style prmtop:
+ if( oldstyle ) then
read(NF,30) (idummy,jdummy,kdummy,ldummy,i=1,ntheth)
read(NF,30) (idummy,jdummy,kdummy,ldummy,i=1,ntheta)
read(NF,30) (idummy,jdummy,kdummy,ldummy,mdummy,i=1,nphih)
If this doesn't work (or even if it does) please consider posting the prmtop
and inpcrd files that you gave to ambpdb: we need to track down the real
origin of this problem.
...dac
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
to majordomo.scripps.edu
Received on Sun Jun 29 2008 - 06:07:15 PDT
