| Resumo: |
The structural genomics initiatives have had as one of their aims to improve our understanding of protein function by providing representative structures for many structurally uncharacterised protein families. As suggested by the recent assessment of the Protein Structure Initiative (Structural Genomics Initiative, funded by the NIH), doubts have arisen as to whether Structural Genomics as initially planned were really beneficial to our understanding of biological issues, and in particular of protein function.
A few protein domain superfamilies have been shown to account for unexpectedly large numbers of proteins encoded in fully sequenced genomes. These large superfamilies are generally very diverse, spanning a wide range of functions, both in terms of molecular activities and biological processes. Some of these superfamilies, such as the Rossmann-fold P-loop nucleotide hydrolases or the TIM-barrel glycosidases, have been the subject of extensive structural studies which in turn have shed light on how evolution of the sequence and structure properties produce functional diversity amongst homologues. Recently, the Structure-Function Linkage Database (SFLD) has been setup with the aim of helping the study of structure-function correlations in such superfamilies. Since the evolutionary success of these large superfamilies suggests biological importance, several Structural Genomics Centers have focused on providing full structural coverage for representatives of all sequence families in these superfamilies.
In this work we evaluate structure/function diversity in a set of these large superfamilies and attempt to assess the quality and quantity of biological information gained from Structural Genomics.

|
