The characteristic folding event of small proteins is the  crossing
of a single transition-state barrier. Transition states may  reveal
why proteins fold fast and efficiently, and how they avoid
misfolding and aggregation. Therefore, intense experimental efforts
are  currently directed at investigating the folding dynamics of
proteins by  mutational analysis. A central and still controversial
question is how to  reconstruct transition states from the mutational
data. I will present a  novel modeling approach for the
reconstruction of transition states that  resolves controversial
issues. Protein function often involves a characteristic binding
event. During T cell adhesion, the cooperative binding of membrane
proteins triggers signaling networks that eventually lead to T cell
activation. The  membrane proteins are arranged in complex
time-dependent patterns. Based  on modeling and simulations, I will
suggest a mechanism for the T cell  pattern evolution and discuss the
role of the patterns for cell  activation.