Hi David, Like Stephen, I have found that integrating the peaks tends to remove much of the B-factor dependence of the analysis seen when looking at the anomalous peak heights. I too tend to use mapman and typically integrate a sphere of 2.5-3 A around my anomalous scatters (S, Se, metals). I then compare the integrated anomalous values for each atom and use known atom types (S or Se) to calibrate. You still have to worry about the occupancy of you anomalous scatters, but the B-factor dependence is much lower. Also, for discretely disordered Met/Mse you may need to sum the integrated values from both alternate positions. Regards, Mitch
-----Original Message----- From: CCP4 bulletin board [mailto:[EMAIL PROTECTED] On Behalf Of Stephen Graham Sent: Monday, April 16, 2007 2:23 AM To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] Differentiating bound Mn & Ca. Hi David, You can use Sheldrick's Calcium Bond Valence Sum to descriminate between metals (see Muller, P., Kopke, S., and Sheldrick, G. M. (2003) Acta Crystallogr., Sect. D: Biol. Crystallogr. 59, 32-37) even at low resolution. I have had good success with this method combined with estimation of anomalous contribution scaled against the S atoms as suggested by Kay. Have a look at <gratuitous self plug> Graham, S. C., Bond, C. S., Freeman, H. C., and Guss, J. M. (2005) Biochemistry 44, 13820-13836. </gratuitous self plug> for plenty of examples. To do the anomalous difference calculations I just integrated a 2x2x2 A box around the metal atom and around the S atoms (in MAPMAN) then calculated the ratio of anomalous difference... Cheers, Stephen On 4/16/07, David Briggs <[EMAIL PROTECTED]> wrote: > Dear all. > > I have recently solved a structure in-house, 2.8A, CuKa. > I have a metal ion bound very obvious hepta-valent co-ordination, which > would suggest either Ca or Mn. > Neither was present in the crystallisation setup, but there was some Mg > around, which has contaminants of both Ca & Mn. > At 2.8A, I don't really think I can reliably discriminate between 2.15A & > 2.36A distances to coordinating atoms > (http://tanna.bch.ed.ac.uk/newtargs_06.html ). > The B factors for refined Ca are 18, and Mn 30. The B-factors of > coordinating atoms vary from... 18 > 30 - so no help there. > > I have a nice clear 6sigma anomalous difference peak, but then, according > to http://skuld.bmsc.washington.edu/scatter/ both Ca (f" > ~1.3) and Mn (f" ~2.8) scatter anomalously at that wavelength. > > The obvious solution is go to a synchrotron and scan around the Mn edge and > see what happens, however, whilst waiting for beam time, is there any way I > could... oh I don't know, use the peak in my anomalous difference Fourier to > figure out what anomalous signal would be required to generate a peak of > that size - a sort of back-transform??? > > Is this do-able, and if so, how would one go about it? > > Cheers, > > Dave > > -- > --------------------------------------- > David Briggs, PhD. > Father & Crystallographer > www.dbriggs.talktalk.net > iChat AIM ID: DBassophile > --------------------------------------- > Anyone who is capable of getting themselves made President should on no > account be allowed to do the job. - Douglas Adams -- Dr Stephen Graham Nuffield Medical Fellow Division of Structural Biology Wellcome Trust Centre for Human Genetics Roosevelt Drive Oxford OX3 7BN United Kingdom Phone: +44 1865 287 549