RE: [ccp4bb]: Low Resolution Heavy Metal Derivatives

[Savvas N. Savvides]

Title: Message

Hi Nick, 7 angstrom data should be enough to determine Hg sites with high occupancy. A difference Patterson map should do the trick very well provided of course your 'aged and soaked' crystals are still quasi-isomorphous to the original native or to the 'aged and unsoaked'  native crystals. In fact, I even think that your molecular replacement solution would be even better to locate the sites in a difference fourier synthesis [Fo(soak)-Fc(MR); alpha(MR].   However, with such high Z-scores in Phaser your MR solution appears to be enough.  Have you refined the structure at all? Simulated annealing could help quite a bit with the overlapping loops, and several rounds of model (re)building with arp-warp starting with your MR model could help you get there as was pointed out a couple of days ago here on the CCP4-bb (Kerr, Kostrewa, Perrakis et al.). Your 2.4 angs data should be enough for that option!   Best wishes Savvas        -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Noinaj Sent: donderdag 6 april 2006 23:52 To: ccp4bb@dl.ac.uk Subject: [ccp4bb]: Low Resolution Heavy Metal Derivatives Hi,   I have previously collected a native dataset at 2.4 angstroms (a few months ago).  I recently used some old crystals for heavy metal soaks using MeHgCl and was only able to collect to around 7 angstroms.  The crystals seem to have 'died' over time, since I also tried some that weren't soaked for comparison and they didn't diffract past 7 angstroms either.  We were able to observe a fluorescent signal from the crystals indicating the presence of Hg, but my question is, is 7 angstroms enough resolution to find the Hg sites with confidence?    I intended to use this in combination with a molecular replacement solution, that I feel is very good (Z-score is 31 in Phaser), but the crystal packing has significant overlap in some regions.  I am currently soaking new crystals with MeHgCl for better resolution, and trying to grow SelenoMet crystals, so I don't want to waste my time if the data that I have now just isn't worth the effort.   Thanks for any and all help in advance.        Cheers, Nick