Protein Crystallization Hits

11/22/2010 8:31:11 AM #

Sounds fancy...

Dolphin

11/22/2010 8:31:24 AM #

But how can we apply this database to "a new protein crystallization project"? The proteins structures in the database are all solved.

Dolphin

11/22/2010 8:32:14 AM #

Hi Dolphin,  
I see where you're getting at. 'NEW' is somewhat ambigous: 'new' as in 'new project' rather than new protein. Or as 'new protein'? In the latter case I'm with you, often a single amino acid change changes crystallization conditions drastically. How can you expect to learn anything from existing crystallization conditions for similar proteins? There's at least some hope that pre-existing crystallization info is useful: there's a hemoglobin study (by G. Gililand, not sure if that was ever published) that showed that hemoglobins from different species crystallize under similar crystallization conditions. Then there are Kinase-specific screens that seem to work. In reality this means that if I want to crystallize a new ABC transporter protein, I'd go into the database and check for xtallization conditions of other proteins in the same family and create a screen around those known conditions.

Dolphin

11/22/2010 8:32:54 AM #

Thank you for the link, I appreciate it!  
  
Keep up the good work!

Dolphin

11/22/2010 8:33:13 AM #

That's a good idea. It will be quite promising if we have a protein has a family member that's already solved. But what if the protein is brand new and there is no protein of this function that's been studied before? can we use the BLAST result to find the high score or high identify protein to trace the possible condition? Let's say I'm going to crystallize protein A. It has the sequence identity of 35% with protein B, of which the structure has been solved. Is it possible to use the crystallization condition of B to try to crystallize protein A? It would be great if there is a database or computational methods for the homologies.

Dolphin

11/22/2010 8:33:35 AM #

> Is it possible to use the crystallization condition of B to try to crystallize protein A?  
  
I seriously doubt it. A specific protein's crystallization behavior difficult to rationalize. Maybe you can exclude portions of the multidimensional crystallization phase space that can be excluded.  
I agree, reliable protein sequence derived crystallization conditions would change our game.

Dolphin