Conserved Waters Program?

Li-Wei Hung lwhung at
Fri Jan 17 20:09:08 EST 1997

Just in case you couldn't find the program that does it all.

Long time ago I also did water clustering but with several steps.

1. Use 'watertidy' (in ccp4) to make sure waters are not symmetry related
2. Use different res-id or chain-id for waters in different pdbs.
3. Superpose all pdbs to a reference pdb.
4. Cut and paste waters all together in one pdb.
5. Use 'contact'(in ccp4), and select distance range of clustering.
   ( in your case 0.0 to 0.5)
6. Check the output file and you'll see what you want.

The results were good when checked on a graphics terminal. I am not quite
sure if
0.5A is a good number for clustering cutoff. It depends on the coordinate
errors and hence the resolution of your structures.

I probably still have most of the input files. Let me know if you 
need help. 


Li-Wei Hung
lwhung at
Phone: (510)486-4338
Fax:   (510)486-6059

More information about the Xtal-log mailing list