WuShock

How would one go about, practically, probing a pool of glycoforms (isoforms) of a particular protein for the type which are most biologically active, that is to say the ones which have the greatest activity? The "brute force method," so-to-speak, would involve assaying every single one and if the potential population is very large then you can imagine how costly and cumbersome the task would be. Much insight could be gleamed from a bioinformatics approach, but I'm wondering does anyone know of a particular approach in the literature? Or does anyone have any suggestions?

I'm giving a lecture called "The Biochemists' Toolbox." My lecture will consist of the aspects of protein engineering (expression vectors, expression systems, and probing protein structure/function via mutagenesis). I'm having trouble finding the pioneers of this field. I'm well aware of PCR (Kary Mullis), however, I don't know where to look for the rest. Any suggestions? Any classical experiments in this field anyone's aware of?