| Abstract
|
A variety of methods are currently availab … A variety of methods are currently available for creating multiple alignments,and these can be used to define and characterize families of related proteins,such as the globins and the immunoglobulins. We have developed a method forusing a multiple alignment to identify an average structural "core," a subsetof atoms with low structural variation. We show how the means and variancesof core-atom positions summarize the commonalities and differences within afamily, making them particularly useful in compiling libraries of proteinfolds. We show further how it is possible to describe the rotation andtranslation relating two core structures, as in two domains of a multi-domainprotein, in a consistent fashion in terms of a "mean" transformation and adeviation about this mean. Once determined, our average core structures (withtheir implicit measure of structural variation) allow us to define a measureof structural similarity more informative than the usual RMS deviation inatomic position, i.e. a "better RMS." Our average structures also permitstraightforward comparisons between variation in structure and sequence ateach position in a family. We have applied our core finding methodology in detail to theimmunoglobulin family. We find that the structural variability we observe justwithin the VL and VH domains anticipates the variability that others haveobserved throughout the whole immunoglobulin superfamily; that a coredefinition based on sequence conservation, somewhat surprisingly, does notagree with one based on structural similarity; and that the cores of the VLand VH domains vary about 5° in relative orientation across the knownstructures. ve orientation across the knownstructures.
|
