 |
 |
Hi, "Question posed in a different way - One experienced crystallographer is being supplied with crystals of proteins - how many can one solve in a year (40 hour weeks, with 5 week holidays)." Under your assumptions with slight modifications (median length ~365 aa, all Se-Met-mostly MAD, some SAD, average resolution ~2.0A or better), rate of structure determination at JCSG for the past week is 168 protein structures with of which ~84% are unique. The rate of structure determination is calculated every week and varies a bit and can be found at: http://www.mcsg.anl.gov/ (Click on the tab "SG Progress" on the left. Information on how the statistics are arrived at is also presented. "The number of structures deposited in the PDB is increasing, the number of groups doing structural work is also increasing. Given these two scenarios - how much has efficiency increased? We know average quality has not improved :-)" According to your paper in Acta D, 2007, "Multivariate statistical analysis revealed that while technological improvements have increased the number of structures determined, the overall quality of structures has remained constant. The quality of structures deposited by structural genomics initiatives are generally better than the quality of structures from individual investigator laboratories" Doesn't the above imply that since the quality of SG structures is better than the quality from non-SG, there has been improvement in quality? Efficiency has definitely improved given the total no. of structures being solved which has brought down the dollar cost per structure. I believe there are a couple of publications which discuss the reduction in cost per structure from ~$250K for non-SG labs to ~$100K or below for SG structures (taking all costs into account). Regards, Debanu. -- Debanu Das, JCSG, SSRL. -----Original Message----- From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Ramaswamy Sent: Tuesday, March 18, 2008 4:03 PM To: [log in to unmask] Subject: [ccp4bb] Off-topic structural genomics related question. Hello all, I apologize for this slightly off topic structural genomics kind of question(s). The number of structures deposited in the PDB is increasing, the number of groups doing structural work is also increasing. Given these two scenarios - how much has efficiency increased? We know average quality has not improved :-) Question posed in a different way - One experienced crystallographer is being supplied with crystals of proteins - how many can one solve in a year (40 hour weeks, with 5 week holidays). Assumptions: 1. The average size of the protein is 250 amino acids. 2. All are MAD structures with Se-Met readily available 3. Beam time is not bottle neck. 4. Average resolution 2.3-2.5 (so, there are some at 3.0 and some at 1.5A resolution). May be some of the SG groups or PX beamlines have already computed numbers like these. I would appreciate if they are shared with me (along with the assumptions that have gone into the calculations). Thanks. Rams. S. Ramaswamy. Department of Biochemistry University of Iowa.