> 
> Ian,
> 
> I agree with your description, but not quite with your 
> conclusion. In the 
> original poster's problem one of the cell axes is so short 
> that at most a  < 20 
> Angstrom Patterson outer radius can be used in a 
> Patterson-based rotation 
> function. This will inevitably lead to loss of many 
> interesting Patterson 
> vectors, and probably a low signal-to-noise ratio.
> 
> So in this particular situation the "direct rotation function" 
> (http://dx.doi.org/10.1107/S0907444995001284) might have a 
> distinct advantage 
> compared to Patterson methods, which is why I suggested it. 
> Whether this is 
> enough to solve Pietro's problem remains to be seen. For me 
> it's a case of "know 
> your options, and try them".
> 
> best,
> 
> Kay
> 
> -- 
> Kay Diederichs              http://strucbio.biologie.uni-konstanz.de
> email: [log in to unmask]  Tel +49 7531 88 4049 Fax 3183
> Fachbereich Biologie, Universität Konstanz, Box M647, D-78457 Konstanz
> 

Kay,

But there's nothing in say AMoRe to stop you increasing the radius cutoff to whatever is required to optimise the signal/noise ratio, apart from this silly IF statement which I always used to delete (I just commented out the CALL CCPERR statement):

      ELSE IF (RMAX.GT. AOBS .OR. RMAX .GT. BOBS.OR. RMAX.GT.COBS)
     +         THEN
        WRITE (6,*) 'Radius: ',RMAX, '  Cell: ', AOBS,BOBS,COBS
        CALL CCPERR(1,'Sphere radius > cell edge')

There's also a program limit in AMoRe on the ratio radius-cutoff/high-res-cutoff due to max Bessel order but since this is currently 500 there will be few cases where it's a limit in practice (it's ~ 80, so max 200 Ang radius for 2.5 Ang data).  There's obviously no point in having a radius cutoff bigger than the greatest molecular diameter (there are no intra-vectors longer than this).

Some people object to including the transform of the origin peak vectors in adjacent cells, but the direct rotation function is effectively doing just that and no-one complains (as long as it solves their structure)!  It seems to me it's particularly important to perform this radius optimisation for highly ellipsoidal molecules, because as you point out a limit of 20 Ang with a molecule 100 Ang long will only include a small fraction of the intra-vectors.  There will always be a trade-off between increasing the number the intra-vectors thus improving the signal, and increasing the number of inter-vectors thus increasing the noise.  It's just a question of finding where the optimal radius cutoff giving the maximum signal/noise ratio lies.

The direct rotation function sometimes used to work better than the Crowther-type function not so much due to the lack of radius cutoff, but more because 1) it uses E's not F's; 2) it uses a correlation coefficient function rather than a product function (the latter is much easier to code using FFT's); 3) the radius cutoff for the Crowther function was often not optimised; and 4) care wasn't taken not to lose strong reflections at low res which several people have demonstrated has a disastrous effect on F-based functions, but is less critical with E-based functions.  However use of a correlation coefficient function based on E's is now a feature of AMoRe (if not the CCP4 version then Jorge Navaza's version).

Anyway this is all academic now, I always use Phaser which does all this properly (I have no experience using it for highly ellipsoidal models though).

Cheers

-- Ian


Disclaimer
This communication is confidential and may contain privileged information intended solely for the named addressee(s). It may not be used or disclosed except for the purpose for which it has been sent. If you are not the intended recipient you must not review, use, disclose, copy, distribute or take any action in reliance upon it. If you have received this communication in error, please notify Astex Therapeutics Ltd by emailing [log in to unmask] and destroy all copies of the message and any attached documents. 
Astex Therapeutics Ltd monitors, controls and protects all its messaging traffic in compliance with its corporate email policy. The Company accepts no liability or responsibility for any onward transmission or use of emails and attachments having left the Astex Therapeutics domain.  Unless expressly stated, opinions in this message are those of the individual sender and not of Astex Therapeutics Ltd. The recipient should check this email and any attachments for the presence of computer viruses. Astex Therapeutics Ltd accepts no liability for damage caused by any virus transmitted by this email. E-mail is susceptible to data corruption, interception, unauthorized amendment, and tampering, Astex Therapeutics Ltd only send and receive e-mails on the basis that the Company is not liable for any such alteration or any consequences thereof.
Astex Therapeutics Ltd., Registered in England at 436 Cambridge Science Park, Cambridge CB4 0QA under number 3751674