Huiying Li wrote:
I used Coot:findwaters facility in REFMAC (CCP4 6.1.1) to add waters to the refined structure. Often for the well ordered water sites the routine added two water molecules in single water site, with their distance (~1 Angs) way shorter than allowed hydrogen bonding distance. I have to remove the extra water molecules manually.
Sounds like an old version. Things have improved.
Is this a intended feature? It seems to only create extra work to the user.
:-(
What is the purpose of having different chain IDs for waters output from this routine?
I thought (at the time) that it was sensible to add waters to a different chain ID to the protein atoms (I still do, in fact). There are others (somewhat less involved in model-building I suspect) that think otherwise.
Paul.
