OPTIMAL PROTEIN-STRUCTURE ALIGNMENTS BY MULTIPLE LINKAGE CLUSTERING -APPLICATION TO DISTANTLY RELATED PROTEINS

A fully automatic procedure for aligning two protein structures is pre sented, It uses as sole structural similarity measure the root mean sq uare (r.m.s.) deviation of superimposed backbone atoms (N, C-alpha,C- C and O) and is designed to yield optimal solutions with respect to th is measure, In a first step, the procedure identifies protein segments with similar conformations in both proteins. In a second step, a nove l multiple linkage clustering algorithm is used to identify segment co mbinations which yield optimal global structure alignments, Several st ructure alignments can usually be obtained for a given pair of protein s, which are exploited here to define automatically the common structu ral core of a protein family, Furthermore, an automatic analysis of th e clustering trees is described which enables detection of rigid-body movements between structure elements, To illustrate the performance of our procedure, we apply it to families of distantly related proteins, One groups the three alpha+beta proteins ubiquitin, ferredoxin and th e B1-domain of protein G. Their common structure motif consists of fou r beta-strands and the only a-helix, with one strand and the helix bei ng displaced as a rigid body relative to the remaining three beta-stra nds. The other family consists of beta-proteins from the Greek key gro up, in particular actinoxanthin, the immunoglobulin variable domain an d plastocyanin, Their consensus motif, composed of five beta-strands a nd a turn, is identified, mostly intact, in all Greek key proteins exc ept the trypsins, and interestingly also in three other beta-protein f amilies, the lipocalins, the neuraminidases and the lectins, This resu lt provides new insights into the evolutionary relationships in the ve ry diverse group of all beta-proteins.