Comments on: The Europeans are cleaning up the PDB, woohoo!!! http://rosettadesigngroup.com/blog/410/the-europeans-are-cleaning-up-the-pdb-woohoo/ Trading tips and news on modeling proteins, DNA, RNA, small molecules, folding, docking, design..... Thu, 04 Mar 2010 18:31:13 -0700 http://wordpress.org/?v=2.9.1 hourly 1 By: Ivan http://rosettadesigngroup.com/blog/410/the-europeans-are-cleaning-up-the-pdb-woohoo/comment-page-1/#comment-7462 Ivan Thu, 06 Aug 2009 10:53:15 +0000 http://rosettadesigngroup.com/blog/?p=410#comment-7462 This idea could be interesting, because sometimes crystallographers are interested only in the structure of active center and not in the orientation of peripheral side-chains. Another thing, they don't mention the efforts of Garib Murshudov et al., who checks the pdb entries for data treatment errors, for example like that http://scripts.iucr.org/cgi-bin/paper?S0907444905036759 (and also tests his BALBES software on the pdb entries) This idea could be interesting, because sometimes crystallographers are interested only in the structure of active center and not in the orientation of peripheral side-chains.
Another thing, they don’t mention the efforts of Garib Murshudov et al., who checks the pdb entries for data treatment errors, for example like that http://scripts.iucr.org/cgi-bin/paper?S0907444905036759 (and also tests his BALBES software on the pdb entries)

]]> By: Xavier Ambroggio http://rosettadesigngroup.com/blog/410/the-europeans-are-cleaning-up-the-pdb-woohoo/comment-page-1/#comment-6631 Xavier Ambroggio Tue, 07 Jul 2009 16:22:38 +0000 http://rosettadesigngroup.com/blog/?p=410#comment-6631 So I read the paper and "cleaning" is definitely right, as opposed to "fixing". The pdb is like a box of treasures in your attic- some of the treasures are in great shape (e.g. well built, high-resolution models), some need dusting and polishing (e.g. majority of structures in this paper), and some need major fixing (e.g. Chang et al. mentioned in this paper). The lack of structure factors for coordinates (did the paper say ~half?) is a major problem but the good structures may allow us to fix the broken structures by bootstrapping using the good structures and sequences. Obviously, a PDB quality Interpol/NATO type organization is needed with servers that run a battery of tests on PDB structures and perhaps give measures of confidence of structures in the PDB based on criteria like resolution, phasing method, depositor (gulp), etc. It's great to see a paper like this as it's a step in the right direction. So I read the paper and “cleaning” is definitely right, as opposed to “fixing”. The pdb is like a box of treasures in your attic- some of the treasures are in great shape (e.g. well built, high-resolution models), some need dusting and polishing (e.g. majority of structures in this paper), and some need major fixing (e.g. Chang et al. mentioned in this paper). The lack of structure factors for coordinates (did the paper say ~half?) is a major problem but the good structures may allow us to fix the broken structures by bootstrapping using the good structures and sequences. Obviously, a PDB quality Interpol/NATO type organization is needed with servers that run a battery of tests on PDB structures and perhaps give measures of confidence of structures in the PDB based on criteria like resolution, phasing method, depositor (gulp), etc. It’s great to see a paper like this as it’s a step in the right direction.

]]> By: David Hall http://rosettadesigngroup.com/blog/410/the-europeans-are-cleaning-up-the-pdb-woohoo/comment-page-1/#comment-6625 David Hall Tue, 07 Jul 2009 13:42:50 +0000 http://rosettadesigngroup.com/blog/?p=410#comment-6625 It seems like a good idea, but I can't completely evaluate it because they made a decision that messes with a lot of tools. They replace all the HETATM records with ATOM records. Considering all the compliments they pay to PDB remediation and complaints about data deposited by other groups, this is horrifying. I guess I should write to someone and complain. It seems like a good idea, but I can’t completely evaluate it because they made a decision that messes with a lot of tools. They replace all the HETATM records with ATOM records. Considering all the compliments they pay to PDB remediation and complaints about data deposited by other groups, this is horrifying. I guess I should write to someone and complain.

]]> By: Egon Willighagen http://rosettadesigngroup.com/blog/410/the-europeans-are-cleaning-up-the-pdb-woohoo/comment-page-1/#comment-6512 Egon Willighagen Sat, 04 Jul 2009 06:06:13 +0000 http://rosettadesigngroup.com/blog/?p=410#comment-6512 Ha, the paper is finally out! Congrats to Gert! The idea is very simple: recalculate the structures with the current much faster computers, allowing more accurate searching of the optimal solution. Additionally, the current code should have less bugs, and use better algorithms. Plenty of reasons why a recalculation makes sense. Ha, the paper is finally out! Congrats to Gert!

The idea is very simple: recalculate the structures with the current much faster computers, allowing more accurate searching of the optimal solution. Additionally, the current code should have less bugs, and use better algorithms. Plenty of reasons why a recalculation makes sense.

]]>