Does structure matter?

[dbernick]

I think that there are a few key reasons why we care about protein structures:

• It is a modeling method that incorporates our understanding of how these polymers fold. We can propose a model and then test it against a structure derived from a crystal or concentrated form of the real polymer. In this sense, predictive modeling gives us a way to test our understanding.
• Structures are conserved more than sequence. A structural prediction can give us hypotheses about the function of the molecule in question.
• It provides the possibility of detecting binding partners -- allowing us to build hypotheses about complexes and pathways.

It may be that this course area use some thoughts on a follow-on course that takes structure prediction into its possible uses. Maybe a seminar in human disease caused by aberrant structure, or maybe a seminar series on metal binding proteins and the coordinating structures that nature has found (how to tell a zinc binding protein from an Iron binding protein). How are binding partners found, and.. is there a faster way?

[KevinKarplus]

I believe that Dietlind's course on protein function does have the sort of material David is asking for. Unfortunately, neither protein course is being offered this year. I think that eventually we'll get to a schedule where we'll alternate year—offering one course one year, the other the next.