Hi,

It seems the consensus answer is still that crystallization is a stochastic process that cannot be easily predicted. I've included links to useful papers describing the structural genomics consortium (SGC) results and a few other ideas. The mindset now is that we can use robots to set up nL size drops and screen many conditions.

Stability will be key, much more important than any other factor. Is the tetramer very robust and stable? Are there floppy ends? Did you remove tags? These are some things to try to predict.

As for conditions, I have had very good luck with the Index Screen sold by Hampton Research, and it is based on conditions that yielded hits for SGC. I do not mean to endorse a particular supplier though, as other companies likely have a very similar screen.

Hope this helps and good luck!

http://www.ncbi.nlm.nih.gov/pubmed/12718928

"Maximum-likelihood crystallization" from 2003

http://www.ncbi.nlm.nih.gov/pubmed/15325655

"Crystallization data mining in structural genomics: using positive and negative results to optimize protein crystallization screens" from 2004

http://www.ncbi.nlm.nih.gov/pubmed/19079241

"Understanding the physical properties that control protein crystallization by analysis of large-scale experimental data" from 2009

http://www.ncbi.nlm.nih.gov/pubmed/20177794

"Predicting protein crystallization propensity from protein sequence" from 2011

http://bioinformatics.anl.gov/cgi-bin/tools/pdpredictor links to a prediction tool

There is no rule at all but ammonium sulphate is usually a good precipitant, of course it depends on the protein. You can have a look in this review, although it is mainly related with protein complexes, they have examples and references about the most sucessful precipitants for proteins. Maybe you can find also more references in this paper:

Radaev, S., Sun. P.D. (2002) Crystallization of protein-protein complexes. J. Appl. Crystallogr. 35, 674-676.

All crystallographers are cocerned about this question which is "THE QUESTION"..There is no answer. Every protein is a new chalenge. Just read articles and try to make your own idea about the role of salts or PEGs or whatever in the crystallization process. Personnaly I like and I learned a lot from Enrico STURA...search artisles of this guy. He gives very nice discussions and ideas about this problem. Besides the only way to have a sort of answer is to try to srystallize several protein by your own.