From: Rolf H. <rh...@fl...> - 2013-08-19 12:38:43
|
On 08/19/2013 01:19 PM, Rolf Huehne wrote: > On 08/19/2013 02:10 AM, Robert Hanson wrote: >> For example, your database shows structures. What's the best default >> orientation to show such that the longest side of the box is along X and >> the shortest depth is toward the user? >> >> rotate BEST >> > I thought that this might be an intended usage. But I didn't understand > it from the description. > > Have you compared the results with the much faster (and atom number > independent) approximation using the boundbox plane sizes, like we are > using it in the Jena3D viewer (jena3d.jenage.de)? > > Although it doesn't work perfectly for all PDB entries it works overall > quite well. And if you have to precalculate about 300.000 images for all > PDB structures it makes quite a difference if the selection of the > orientation takes milliseconds or in the range of seconds. > I did a little comparison between "http://chemapps.stolaf.edu/jmol/jsmol/simple.htm" and our predefined images. It was hard to find an example that would need improvement, looking quickly at the thumbnails of the first 10000 protein structures in a customized JenaLib entry list ("http://jenalib.fli-leibniz.de/cgi-bin/entry_list.pl?TYPE=customized&TABLE=&STATISTICS=&DATABASE_FILTER=pdb&DATABASE_FILTER=ndb&METHOD_FILTER=x-ray_neutron_methods&METHOD_FILTER=nmr_methods&METHOD_FILTER=electron_methods&METHOD_FILTER=other_methods&METHOD_FILTER=theoretical_models&MOLECULE_TYPE_FILTER=protein&SORT1=pdb_expdata_type&SORT2=idb_id&SORT3=default&LIST_INFO=serial_number&LIST_INFO=image&LIST_INFO=pdb_expdata_type&LIST_INFO=pdb_entry_type_list&LIST_INFO=pdb_release_date&LIST_INFO=pdb_header&DATE=any&DAY1=&MONTH1=&YEAR1=&DAY2=&MONTH2=&YEAR2=&OUTPUT_FORMAT=pre&WIDTH=200&LIMIT=limit&ENTRY_LIMIT=10000&SEARCH_TERMS=&CODE_LIST=&VIEW=build+entry+list"). The only one I found this way was 1PFD, although it looks not so bad. But in my opinion the "rotate best" result looks even a little bit worse. So I just picked some test candidates looking good in Jena3D/JenaLib and tried "rotate best": - 1AZN and 1B4M look good - 3J0Q looks not quite as good - 1AW0 and 1CRP look worse but still better than default - 108L and 1A23 look not good Regards, Rolf -- Rolf Huehne Postdoc Leibniz Institute for Age Research - Fritz Lipmann Institute (FLI) Beutenbergstrasse 11 07745 Jena, Germany Phone: +49 3641 65 6205 Fax: +49 3641 65 6210 E-Mail: rh...@fl... Website: http://www.fli-leibniz.de Scientific Director: Prof. Dr. K. Lenhard Rudolph Head of Administration: Dr. Daniele Barthel Chairman of Board of Trustees: Dennys Klein VAT No: DE 153 925 464 Register of Associations: No. 230296, Amtsgericht Jena Tax Number: 162/141/08228 |