P212121

The Protein Structure Initiative (PSI) is responsible for depositing over 8,000 structures in the protein data bank (PDB). If you are working on a project related to a structure by the PSI then check out the Structure Knowledgebase. This resource has compiled a vast amount of information into a slick search function.

A number of results are produced with my favorite being the experimental protocols through the PepcDBl. Here are the links from the above example:
* PepcDb Summary
* selection
* cloning
* expression
* purification
* crystallization

This organization is very useful and would love to see this type of resource for all the deposited structures.

Related Posts:

• sc-PDB Database For Comparing Binding Sites
• PCDB: Protein Conformational Database
• A Web Interface for PDB Databases
• RBPDB: RNA Binding Database
• Lipid Structural Database: LipidBank

8 Awesome Insights so far | Have Your Say!

1. 

   Tommy
   August 27th, 2010 at 4:25 AM #

   Sean, I checked out the PSI/SBKB. I noticed some of the steps about expression and crystallization are not present in the mmCIF file. For example here is the crystallization description in the mmCIF:

   _exptl_crystal_grow.method ‘vapor diffusion’
   _exptl_crystal_grow.pH 7.0
   _exptl_crystal_grow.temp 293
   _exptl_crystal_grow.pdbx_details ‘50% PEG 200, 0.1M Tris, 50mM lithium sulfate, pH 7.0, vapor diffusion, temperature 293K’

   And here it is in the SBKB:

   The protein was concentrated to 11.6 mg/ml in
   100 mM NaCl, 10mM HEPES, pH 7.5, 2mM EDTA.
   Crystallization was performed in 50% PEG 200,
   0.1M Tris, 50mM lithium sulfate, pH 7.0, by vapor
   diffusion, in sitting drops, at 293K.

   If you come across something useful in the SBKB missing from the PDB/mmCIF, I would appreciate, if you let me know.

   Tommy (a daily reader of your blog)

2. 

   Sean
   August 27th, 2010 at 4:23 PM #

   Hi Tommy,

   I would be interested in seeing if there are any trends in protein concentration that are provided by the PSI. This type of information as you mentioned is not PDB or mmCIF files. I haven’t seen a detailed analysis of protein concentration in regards to crystallization (anyone else?). I have come across two general approaches in this area which is to maximize solubility or to set up drops in the range of 10-20 mg/ml for initial screening. Although the concentration needed for crystallization would be protein dependent it would be interesting to see if there are any patterns such as membrane proteins vs. enzymes, etc…

3. 

   Tommy
   August 27th, 2010 at 9:09 PM #

   I’ll put it in my notebook for later. It’s not completely straight forward to parse the data from the PSI and I don’t quite have the time for it at the moment. I’ll keep you posted.

4. 

   Artem
   August 30th, 2010 at 3:22 PM #

   I did not look at the PSI data very hard (yet) but so far I’ve not found any place that would list a summary of crystallization experiments (per protein) that lists not only *the* condition that gave *the* crystal, but instead all conditions that were tried (per each protein) – with a note on whether they produced crystals or not. A simple table with ‘X” per crystal found would be cool…

5. 

   Tommy
   August 30th, 2010 at 3:33 PM #

   Yes, that’s the problem with science in general. People never list their failures.

   I’m still planning to have a look at the data in the future!

6. 

   Sean
   August 30th, 2010 at 4:07 PM #

   Hi Artem,

   I was looking for a similar table when putting together this post without any luck, which was a bit frustrating. I ended up finding this PDB file through their help examples. I will leave a comment here if I come across it.

7. 

   Arten
   September 1st, 2010 at 8:25 PM #

   Just a note – it’s not about ‘failure’ in this case — each crystallization experiment should be analyzed as a whole so listing only successful conditions means that those of us who (like me) meddle with crystallization propensity analyses and suchlike are deprived of key data. I was (naively) expecting to find out that PSI groups maintain an archive of all crystallization experiments (and they very well might be doing just that!) but so far I’ve not seen thisdata in the public domain.

8. 

   Maria
   September 2nd, 2010 at 3:50 AM #

   I’m glad you like the information that the PSI provides. Until recently I was the editor of the Structural Biology Knowledgebase and I always thought that the most important information on the site was the experimental information. I’m suprised more people don’t routinely search here for starting points for expts. Much of the info – cloning, expression and purification – would be of interest to biochemists, not just structural biologists.

   For “fails” – there are notes about how far along the pathway a protein came, so if it expressed and purified OK the protocol will still be there.