While it is certainly a simple coding job to convert back and forth between orthogonal and fractional coordinates there is perhaps something lost in the former and that is the notion of a molecule embedded in a crystal structure. Generally speaking, it's the latter that we determine. George may well be right that orthogonalization made it easier for those non-crystallographers who use structures but I wonder if those folk are aware of all the information about intermolecular contacts that you lose by looking at the molecular structure only. Sure, you can recover that information from orthogonalized coordinates (please consult a crystallographer if you are in need of guidance) but will you? Sent via BlackBerry by AT&T
-----Original Message----- From: George Sheldrick <[email protected]> Sender: CCP4 bulletin board <[email protected]> Date: Sun, 6 Jan 2013 09:57:28 To: <[email protected]> Reply-To: George Sheldrick <[email protected]> Subject: Re: [ccp4bb] Who invented PDB format? Chemical crystallographers have always used fractional coordinates, it makes it so much easier to handle symmetry, special positions etc. But if the PDB hadn't used orthogonal coordinates, bioinformatics might never have taken off. George On 01/06/2013 09:34 AM, Eleanor Dodson wrote: > Some of us resisted using an orthogonal format for coordinates, > arguing that the output from a crystal structure should refer to > crystal axes. > And since symmetry was a crystal property it was important that we > could "see" it easily. The PDB format won out, but I still use > *coordconv* a lot > to turn back the orthogonalised PDB style to fractional coordinates - > to see if this heavy atom solution is the same as that one, given an > origin shift, etc etc. > Eleanor > > On 4 Jan 2013, at 20:44, Soisson, Stephen M wrote: > -- Prof. George M. Sheldrick FRS Dept. Structural Chemistry, University of Goettingen, Tammannstr. 4, D37077 Goettingen, Germany Tel. +49-551-39-3021 or -3068 Fax. +49-551-39-22582
