August 18, 2009 15:37
Sometimes I wonder how the world will look like in a few years and: how will we be crystallizing proteins in 5 years from now? Of course it is impossible to predict new technological breakthroughs. However, I think there are technologies that are in their infancy now that will be widely used in 5 years. That allows us to extrapolate a bit. Here are some trends that point to fewer primary crystallization screening experiments and less protein sample requirements within 5 years time:
1. Better primary crystallization screening regimes. As more systematically generated protein crystallization data becomes available, it will become easier to extract best practices for the 'first pass' round of crystallization screening . This could go either generic (weeding out the duds) or customized for a particular protein. I think the jury is still out there. Maybe Emerald will have a sequence-in screen-out online tool? ("getting there with escreenbuilder", says Mark). Hard to believe though that there will be a reliable sequence-based crystallization cocktail predictor (please prove me wrong!).
2. More Seeding methods: Looks like seeding and seeding-like techniques will make it big. From what I hear, the microseed matrix screening works well in many labs. What about universal nucleants such as laser pulses and grains?
3. More ultra-low volume crystallization setups. My first crystallization experiments that I set up in the 90's were on the scale of 10 ul. Now the volume of typical crystallization experiments are down to 100 to ca. 10 nano liters (check out Emerald's MPCS). While the technology to go pico Liters is there, for now 10 nano Liters seems to form a practical barrier. Less volume makes sense only if you can get enough crystal mass for X-ray diffraction, or you're only interested in crytallizability and not physical crystals. But how much is crystal mass is enough? The requirement for crystal size has continued to decrease, with structures of crystals less than 10 um possible now and possibly routine in 5 years (think NSLS-II ). For now I don't really think lower than 1 nL crystallization volume makes much sense, things become really messy.
Looking forward to re-read this blog post in 2014 (with foot in mouth?),