[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ccp4bb]: ab initio



***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Original message from Bernhard Rupp:

> DA,
> 
> I notice that the phrase 'ab initio phasing' is used
> sometimes to describe experimental phasing (Mad etc)
> vs MR phases, or does it apply to direct methods 
> only (like in Sheldrick's work)?
> 
> Is there a consensus or preference what 'ab inito phasing'
> exactly means?
> 
> Thx, br
> ------------------------------------
> Bernhard Rupp
> 

--
Dear Bernhard,

     Thank you for bringing this up. There is some history behind
this, as the confusion between 'ab initio' and 'de novo' has been used
in the past to make falsely spectacular claims about such things as
solving a protein structure 'ab initio' by some concoction of direct
methods formulae, when in fact the small print showed that two
isomorphous derivatives had been used. When I challenged this claim in
Como and then in Erice, I remarked that 'ab initio' means from the
beginning' in Latin (and hence perhaps in a closely related modern
language) so that you could always claim to have done something 'ab
initio' if you had done it ... from wherever you started from!

     More seriously, there seems to be a problem with using more than
one Latin expression these days, as more people cannot tell the
difference between them. The other expression is 'de novo', which is
the legitimate expression to describe an experimentally phased
structure, determined without Molecular Replacement, and hence
presumably showing structure which has a chance of being new.

     'Ab initio' should be reserved for solutions in direct methods
style, i.e. from native amplitudes only, without any preconceived idea
of stereochemistry. Perhaps this is becoming an overly purist
viewpoint, since nowadays you would almost have to stop yourself from
measuring anomalous scattering data in order to solve something in a
truly ab initio manner. The same blurring of boundaries will occur
when probabilistic or dual-space methods become smart enough to use a
lot of a priori stereochemical knowledge. However, as long as the
expression 'ab initio' retains its air of mystery as the ultimate
achievement of theoretical macromolecular crystallography, with whiffs
of Nobel incense wafting in the background, its abuse should be
vigorously discouraged and picked out by all 'bona fide' (oh dear, more
Latin) crystallographers.


     With best wishes,

          Gerard.

-- 

     ===============================================================
     *                                                             *
     * Gerard Bricogne                     gb10@GlobalPhasing.com  *
     *                                                             *
     * Global Phasing Ltd.                                         *
     * Sheraton House, Castle Park         Tel: +44-(0)1223-353033 *
     * Cambridge CB3 0AX, UK               Fax: +44-(0)1223-366889 *
     *                                                             *
     ===============================================================