The attached files are from a workflow I used with CHARMM to count water transits through a bilayer. It assumes that any drift in z has been removed by some alignment procedure that ensures the bilayer is intact and centered at a known position. Briefly,

• watchk.inp :: identifies waters that visit the bilayer midplane region, creates a log file
• BldWatEnt.csh :: processes the log file to get a unique list of RESIDs, and creates a stream file of ENTER commands for use with CORREL
• wtrans.inp :: creates time series of z coord for identified waters using the stream file from BldWatEnt.csh, one file per water in a subdir (env var $ZWAT)
• wtransits.f90 :: Fortran90 program that analyzes the water z time series produced, and reports on true transits

The idea is to greatly reduce the number of water molecules to consider explicitly by determining which ones actually visit the hydrophobic midplane region in the initial screening (watchk.inp).

The Fortran program expects command line args, and gives a brief help message when none are provided.

(The .txt extensions should be removed before use; make sure the .csh file has execute permission.)

The BldWatEnt.csh program could not handle more than 500 water molecules in one run. Is there a way to increase its size limit? Thanks.

There's nothing in the BldWatEnt.csh script that limits the number of waters.

Hi, Rick, when I run this program, it gave me error message as "Found 1201 unique water RESIDs, foreach: Too many arguments". When I cut the trajectory into two halves, and ran each piece one after another, it worked. That is why I suspected some kind of size limitation involved.

It's possible the 'foreach' command has a limit of 1200 for the number of items inside the parentheses; I hadn't run into that.