Eleanor Dodson wrote:
Even for NAD I think I would make my own "new" dictionary.
I must say that I would not (in the general case that is, there may be an argument for doing so for NAD). There will be a price to pay at deposition time.
To address your particular problem (and Ignoring the problems of the hydrogens for now) the issue of the explosion in the nicotinamide is due to a mismatch between the dictionary name and the PDB name for the nitrogen atoms, NN1 and NN7. Coot has an algorithm to map the dictionary names to PDB file names, but the NN1 and NN7 are convolutedly (and erroneously) named (IMHO) that it does not match properly. You can get round this by editing the dictionary for NAD (I get mine from LIBCHECK [1]), changing occurrences of NN1 to 'NN1 ' (i.e. include a quoted space at the end of the 3 chars). Do the same for NN7 -> 'NN7 '. Then read in that file to Coot: File -> Import CIF dictionary and Coot will be happy.
This is the current situation - things will change when Coot moves to PDB version 3.x-formatted dictionary.
[1] note that for NAD, LIBCHECK will produce a dictionary that will idealise the riboses to flatness (but without a plane restraints, interestingly enough)
Paul.
