Kristen Mayer wrote:
> I am having 2 major problems: First, I am trying to loop through the SA part
> 30 times, and it always stops at 12. Second, the structures have recently
> stopped being random. Meaning, if I do 2 separate runs with different
> structures as my starting coordinates, the results are always the same- the
> #1 structures are all identical, the #2 structures are identical, etc.

Since I can't run your script without the various ancillary files, I can
only offer 
a few observations:

1.  Why are you reading in starting coords in two places?  One is at the
very top of the
script, and one is just inside the "main" loop.

2.  Your coor read statement inside the main loop could be problematic,
since you're using
a single @ before the filename.  File reads within loops usually require
@@ before the 
filename to force xplor to re-read the file each iteration (see p 11 of
the old xplor 3.1 
manual).  

This kind of construct is asking for trouble.  You should really use the
mechanism that 
the script originally had--read the starting coords once, at the top of
the script, 
copy it to the comparison array with a coor copy end statement, and then
use the 
coor swap end coor copy end lines at the top of each loop iteration to
grab a fresh copy 
of the starting coords.  

3.  Finally, if you want to evaluate reproducability problems (which, I
suppose, is the
opposite of what you're having :-), it's good practice to set the random
number seed 
at the beginning of each loop iteration in some convenient way.  For
example,

eval ($count = 0) 
while ($count < 10) loop main
   coor swap end
   coor copy end
   eval ($newseed = 42 + $count)
   set seed $newseed end

   ...

This way, you can jump ahead to any problematic stuctures by just
changing that first 
eval statement.  

Let me know if this helps.  And please send us your output if it
doesn't.

--JK

-- 
Dr John Kuszewski

301-594-5831 (voice)
301-402-2867 (fax)
[email protected]

Reply via email to