Well that lab can join the WCG as a project and The Grid will recreate the proteins for them.

Kinda of related to the HCC andany other protein-based WCG project.

Cheers.

I know, but maybe The Scribbs Institute can make a project to recreate the proteins for them based on the info the personnel of the destroyed lab can provide, just a thought, the feasibility of the project is yet to be seen though, but if so, maybe a new project for the WCG may be born.

Cheers.

The main problem with protein crystalisation is finding conditions in which it will give you nice crystals of a size and quality you can structure determination on. So I would assume some of the samples they know the conditions although it could still be very expensive to produce if they are isotope tagged or they don't express well to start with in producing the protein. A lot of them would be test conditions for crystalising protein I would imagine which would all have to set up again. Overall its a lot of effort.

As far as I know there is no computational method to try and predict crystalisation conditions. One of the problems is by the nature of what you're trying to do you do not have a structure of the protein to attempt to perform some kind of solvation computation with. So it would have to use predicted structures and even then its wouldn't work like docking simulations that are running here since the solvent and even tiny amounts of trace elements are/can be very important in forming good crystals.

If all the bits of the algorithm could be produced and shown to help at least narrow down possible conditions it would be a massive project for sure.

They would have been growing the crystals to perform X-Ray Crystallography!
They get the shape and structures of the molecules, by working out exactly were each atom is situated. They analyse reflections of light passed through the lattice of the crystal. The crystals are arrays of lined up molecules, only a few mm across. They are able to work out if the light was twisted in various ways using computer analysis and deduce the atoms positions.
A common alternative technique is Nuclear Magnetic Resonance and I wont go into that!
However, computer modelling is also an alternative method. Performed together they complement each other and confirm observations, or indeed reject some.
Basically in computer modelling you work from the amino-acid sequence, their charges, and using calculations about weak and strong Hydrogen bonding, hydrophobic and hydrophilic side chains (and so on), you try to calculate the shape of the molecule, given that molecules tend to structure themselves into conformations that are most stable. It is Physical Biochemistry or molecular modelling.