On Wed, Apr 20, 2011 at 11:45 AM, Edward A. Berry
There was discussion a few years back about an extended pdb format in which the fourth digit would go not just 0-9 but 0123456789ZBCDE...XYZabcdefghijkl....z (and the other digits would also if the fourth digit was greater than 9) So maybe the code that handles that isn't kicking in where it should?
This is Ralf's "hybrid36" format. I think Phaser will use that encoding for atom records, but not residues. In any case, using the hybrid36 format is unnecessary in this case, since the residue number field is officially (according to the PDB) four characters long.
The folks who want to deposit structures of whole bacteria will have to wait a little longer.
The PDB won't even handle ribosome structures without splitting them into multiple files, so this has been a problem for at least a decade. (Phenix does not have this limitation, of course, nor do Coot, Refmac, or CNS, as far as I know.) Residue numbers are less likely to overrun the official format, but I think there are structures for fragments of proteins where the actual residue numbering starts above 9999. I'm not sure what the PDB does to these. -Nat