> Not really - in my hands the ammonium sulfate was one hit out of 7.
>
> While Ivan's question is about Fab complexes with protein antigen, I
> think it brings up a more general question of protein class-dependent
> crystallization bias. While some general trends exist for classes of
> biopolymers (e.g. MPD is number one precipitant for DNA; protein:DNA
> complexes tend to crystallize in PEG-based conditions), a general idea
> of assigning a preferred precipitant to a protein class is, imho,
> pointless. Fabs are a good example - one would think that with half of
> the protein more or less the same in all instances some general trends
> should exist. And perhaps they do, as this
>
>
http://scripts.iucr.org/cgi-bin/paper?S0907444999016224
>
> seems to suggest. But alas, Fab crystallization conditions, once you
> look into it, appear to be just as diverse as the same for proteins in
> general. Crystallization conditions may change radically upon point
> mutation, so why would one expect that a class of proteins sharing some
> 50% identity will show unusual love for PEG, ammonium sulfate, sodium
> malonate or any other "miracle precipitant"?
>
> Consider this. Thanks to great engineering at the Douglas Instruments,
> we can routinely set up ~1000 drops for a given protein. If one of them
> shows a crystalline shower, we celebrate. To me, the fact that we try
> wrong crystallization conditions 99.9% of the time, proves that any
> attempt to predict crystallization conditions beyond vague things like
> "keep pH close to protein pI", "sodium malonate is cool", "PEG and
> ammonium sulfate are two most successful precipitants in history of
> protein crystallography", etc., is futile. Time wasted on looking into
> what is the most common precipitant for a particular class of proteins
> is better spent on setting up more trays.
>
> Cheers,
>
> Ed.
>
> --
> Oh, suddenly throwing a giraffe into a volcano to make water is crazy?
> Julian, King of Lemurs