Re: [ccp4bb] Mosflm : data process of crystal with huge unit cell
Perhaps it should also be noted here that the long crystal axis should be oriented close to parallel to phi, in order to minimize overlaps. I have heard anecdotal evidence (anyone else?) that this long axis, in plate crystals, is usually normal to the surface of the plates. Perpendicularly-bent/folded loops, in combination with goniometer adjustments, can used to acheive this. The plate is then shot through the edge from all directions. Do others agree with this? It makes sense to me theoretically and in my own experience, anyway. Jacob *** Jacob Pearson Keller Northwestern University Medical Scientist Training Program Dallos Laboratory F. Searle 1-240 2240 Campus Drive Evanston IL 60208 lab: 847.491.2438 cel: 773.608.9185 email: [EMAIL PROTECTED] *** - Original Message - From: [EMAIL PROTECTED] To: CCP4BB@JISCMAIL.AC.UK Sent: Thursday, May 22, 2008 9:47 AM Subject: [ccp4bb] Mosflm : data process of crystal with huge unit cell Dear All I am processing data from a crystal for a large macromolecular complex with mosflm. Cell dimensions are around 120 x 150 x 650, with a p222 spacegroup. To avoid overlaps, we have collected data with a oscillation of 0.1 degrees. When I try to process the data with mosflm, mosaicity decreases along the processing to a very low values (0.05 to 0.1 depending on the images). The result is that I miss a lot of spots from the images. I would appreciate any help regarding : 1.- What's is the best strategy to process such a dataset ? 2.- Which are the critical parameters in mosflm to avoid these problems, and how to modify them ? 3.- and finally, can I use this data or your advice is to try to get crystals in a different spacegroup ? Many thanks Best regards Francisco - Francisco J. Enguita, Ph.D. Macromolecular Crystallography Laboratory ITQB EAN, Av. da República 2781-901 Oeiras Portugal Phone : +351-21-4469663 Fax : +351-21-4433644 E-mail : [EMAIL PROTECTED] -
Re: [ccp4bb] Mosflm : data process of crystal with huge unit cell
Dear Francisco, With a c axis of 650 Angs your phi-overlap problem is severe. When this axis is close to being parallel to the beam, the angular distance in radians between h,k,l and h,k,l+1 for a reflection hkl close to the top or bottom of an image is the angle spanned by c* viewed at a distance d* (=d*[h,k,l]), i.e. roughly c*/d*. This gives an angular distance of 1 degree for a c axis of 180 Angs at a resolution of pi (=3.14159...) Angs. In your case, at a resolution of 3.25 Angs this angular distance is only 1/200 radian, i.e. 0.286 degree; at higher resolution, it is proportionally smaller. Therefore, even a small mosaicity (0.2?) will give you severe phi overlap at the top and bottom of your images (assuming that the spindle axis is horizontal). The best you can hope for is that the estimation of that mosaicity by matching predicted spots to observed spots will give a correct rejection for affected reflections - until refinement against measurements that are linear combinations of intensities of several reflections, rather than intensities of individual reflections, becomes possible. With best wishes, Gerard. -- On Thu, May 22, 2008 at 03:47:24PM +0100, [EMAIL PROTECTED] wrote: Dear All I am processing data from a crystal for a large macromolecular complex with mosflm. Cell dimensions are around 120 x 150 x 650, with a p222 spacegroup. To avoid overlaps, we have collected data with a oscillation of 0.1 degrees. When I try to process the data with mosflm, mosaicity decreases along the processing to a very low values (0.05 to 0.1 depending on the images). The result is that I miss a lot of spots from the images. I would appreciate any help regarding : 1.- What's is the best strategy to process such a dataset ? 2.- Which are the critical parameters in mosflm to avoid these problems, and how to modify them ? 3.- and finally, can I use this data or your advice is to try to get crystals in a different spacegroup ? Many thanks Best regards Francisco - Francisco J. Enguita, Ph.D. Macromolecular Crystallography Laboratory ITQB EAN, Av. da República 2781-901 Oeiras Portugal Phone : +351-21-4469663 Fax : +351-21-4433644 E-mail : [EMAIL PROTECTED] - -- === * * * Gerard Bricogne [EMAIL PROTECTED] * * * * Global Phasing Ltd. * * Sheraton House, Castle Park Tel: +44-(0)1223-353033 * * Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 * * * ===
Re: [ccp4bb] Mosflm : data process of crystal with huge unit cell
Hi Francisco, you could try using these keywords in Mosflm: POSTREF WIDTH 3 #since you have 0.1 degree oscillation and Mosflm can only handle 30 images at once, the default I believe is 5 images if I'm not mistaken SEPARATION CLOSE PROFILE RMSBG 25 POSTREF USEBEAM Then I would start indexing from various images separated let say by 15 degrees use 6 or more if you have, don't go to full resolution first refine the cell estimate the mosaicity, fix the mosaicity and process a low res pass (what is your resolution ?). I assume you didn't run strategy beforehand and checked where you would have overlaps ? If you still have the crystal you could collect another dataset in a different orientation perhaps (assuming you can move kappa). I would then index on two images run strategy and get some images where the predicted overlaps are, then try to optimize the distance to the detector in favour of no overlaps. If you can, use the biggest CCD detector available or image plate if that has a larger detector area. Since you're in P222 you could also offset 2theta to artificially increase your detector area. Set the beam position almost at the edge of the detector and give your crystal a good 360 or more degree spin - sure you will end up with somewhere close to 100GB of data but who cares if you can solve the problem. 650 A is a challenge but not impossible - how many molecules in the asu do you expect ? Can you solve your problem with MR or do you need HA phases - if that is the case, then searching for another crystallization condition might be faster. Another thing to mention is if you have a homehexamer don't forget you can use NCS averaging so if you only collect a 2.5 A dataset your resulting electrondensity maps might look like 2 A after averaging. I suggest to rather be modest in resolution and have no overlaps then be able to solve the structure and go back eventually to optimize for higher resolution etc. With the cell dimensions you obtained you could startup XDS if you want to give it a try. As Graeme mentioned xia2 might be easier for a first time XDS user. Regarding obtaining new / different crystals you could try the Hampton additive screens with your crystal condition, with some luck you might end up with a more favourable crystal form and smaller cell dimensions. Assuming you are working on a macromolecular complex, do you have any EM information perhaps - then you would know if e.g. your cell dimensions correspond to one complex perhaps or more. Then the smallest cell axis you will most likely get will be in the size of your complex. Just some thoughts, Juergen [EMAIL PROTECTED] wrote: Dear All I am processing data from a crystal for a large macromolecular complex with mosflm. Cell dimensions are around 120 x 150 x 650, with a p222 spacegroup. To avoid overlaps, we have collected data with a oscillation of 0.1 degrees. When I try to process the data with mosflm, mosaicity decreases along the processing to a very low values (0.05 to 0.1 depending on the images). The result is that I miss a lot of spots from the images. I would appreciate any help regarding : 1.- What's is the best strategy to process such a dataset ? 2.- Which are the critical parameters in mosflm to avoid these problems, and how to modify them ? 3.- and finally, can I use this data or your advice is to try to get crystals in a different spacegroup ? Many thanks Best regards Francisco - Francisco J. Enguita, Ph.D. Macromolecular Crystallography Laboratory ITQB EAN, Av. da República 2781-901 Oeiras Portugal Phone : +351-21-4469663 Fax : +351-21-4433644 E-mail : [EMAIL PROTECTED] - -- Jürgen Bosch University of Washington Dept. of Biochemistry, K-426 1705 NE Pacific Street Seattle, WA 98195 Box 357742 Phone: +1-206-616-4510 FAX: +1-206-685-7002 Web: http://faculty.washington.edu/jbosch
