Conclusion
There is, in general, not a unique value for any particular type of interaction within a protein. Because of tertiary and other extraneous interactions, all of the interactions can be variable and depend on context. There are further complication in that active sites are not selected for optimal stability. This situation makes it all the more important to give much attention to accumulating reliable themodynamic data concerning all aspects of protein behaviour. The combination of the 3D structure determination, both site-directed and random mutagenesis is a powerful tool for studying the stability of proteins. Many features of a protein molecule that contribute to its stability have been identified and , in many case, quantitatively assessed from such combination studies. Further progress requires the acquisition of more high resolution structures of mutants combined with accurate measurements to allow the calibration of computational methods.