Hi Patrick, I'd do the following: - go to www.rcsb.org and choose "Advanced Search" - choose "All/Experimental Type/Molecule Type", choose "Experimental Method" and set to X-ray - click the "+" next to "Add Search Criteria" - in the pulldown menu, choose "Software" and enter "Phenix" (leave as "contains") - click the "+" next to "Add Search Criteria" - in the pulldown menu, choose "Deposit date" and select e.g. 2014-01-01 until 2014-12-31 - then "Submit Query", note the result (2440 Structure Hits)
Repeat this with "Refmac" instead of "Phenix" and note 2098 Structure Hits Repeat with "Buster" and note 373 Structure Hits Repeat with "MAIN" but there are none At this point we need some sanity checks. Are there entries with both "Phenix" and "Refmac" (yes: 99), or "Phenix" and "Buster" (yes: 6), or "Buster" and "Refmac" (yes: 5)? Are there entries which have all 3 packages (no)? We thus capture 2440+2098+373-99-6-5 = 4801 depositions because we shouldn't count those entries twice which cite two packages. How many X-ray depositions were there in 2014? Their total number is 4920. So we didn't capture 4920-4801=119 entries. I don't know a way right now to check what they cite as software. (At this point I realize that I neglected "CNS", and indeed there are 138 entries, and 27 out of these go together with "Phenix". So the updated count is then 2440+2098+373+138-99-6-5-27 = 4912. Almost all 4920 captured!) So it looks like X-ray depositions in 2014 _do_ have reasonable data concerning their usage of software, and it's likely that Phenix (or rather, phenix.refine) was used for ~(2440-99-6)/4920=47%, Refmac5 for ~2098/4920=42%, Buster for ~8%, CNS for ~2.5% . Here, I subtracted the "Phenix and Refmac" entries from the total Phenix count, because it seems likely that Phenix was used for other purposes than refinement. Likewise for "Phenix and Buster". Don't know what to do about the 5 "Buster and Refmac" entries - just ignore them! So within 10 minutes or so one gets a fairly believable picture of which refinement program people used in 2014. Please note that the RCSB is still processing 2014 depositions it seems, so if you repeat this in a couple of weeks/months/years the absolute numbers will be different. HTH, Kay P.S. I'm relying on the critical eyes of other readers to point out my mistakes! On Sat, 24 Jan 2015 09:41:36 -0800, Patrick <patrick.coss...@inbox.com> wrote: >Hi, > >not directly related but I wanted to know how to extract from the PDB all the >PHENIX/CCP4 refined structures currently deposited. > >Do I use the "text" search in quarry to get the result or do I use the >software? > >I tried it for both but seems to give different results (no's). > >I tried for another refinement package (shelx) and same, different no's > >Using software, and looking for contains 320 > >using text search 130 > >Is there another way to get structure of a certain resolution ? > >Patrick > >____________________________________________________________ >Can't remember your password? Do you need a strong and secure password? >Use Password manager! It stores your passwords & protects your account. >Check it out at http://mysecurelogon.com/password-manager
