Structural genomics projects are beginning to produce protein structures with unknown functions, thereby creating a need for high-throughput methods to predict functions. Although sequence-based function prediction methods have been used extensively, structure-based prediction is believed to provide higher specificity and sensitivity because functions are closely related to the three-dimensional structures of functional sites, which are more strongly conserved during evolution than sequence. We have developed FCANAL, a method to predict functions using a score matrix obtained from the distances between Cα atoms and frequencies of appearance [1]. The previous report used key residues predicted from sequence comparisons (motifs). In this report, we have expanded the method to include enzymes and binding proteins with key residues predicted on the basis of three-dimensional structures. Using FCANAL, we constructed score matrices for 31 enzymes. When we applied them to all of the structure entries deposited in the Protein Data Bank, FCANAL could detect functional sites with high accuracy. This suggests that FCANAL will help identify the functions of newly determined structures and pinpoint their functionally important regions.

View full abstract

Secretory proteins function as agents for numerous cell-cell interactions and determine the survival strategies adopted by organisms. Using the SOSUI system for membrane proteins (Hirokawa et al., Bioinformatics, 1998) and SOSUIsignal for signal peptides (Gomi et al., CBIJ, 2004), we undertook predictive analyses of secretory proteins from 248 prokaryota using all of the amino acid sequences coded by their respective genomes. The number of secretory proteins exhibited a strong positive correlation with the number of total open reading frames, with analysis of these correlations revealing that prokaryotic organisms could be placed into several groups. Symbiotic or obligate parasitic organisms in eukaryotic cells with less than 1200 open reading frames exhibited a single linear relationship between the number of secretory proteins and the total number of open reading frames. Conversely, free-living organisms with more than 2500 open reading frames could be grouped into three linear relationships. The intercept with the axis of the number of open reading frames in the linear relationships was approximately 300 genes for the survival of symbiotic or obligate parasitic organisms and approximately 700 for the free-living organisms. The factor responsible for distinguishing between the different categories of organisms appeared to be G+C content and the number of open reading frames. The roles of secretory proteins and membrane proteins were discussed on the basis of the ratios of those proteins. The list of all predicted secretory proteins for 248 prokaryota is available through the internet at the URL: http://bp.nuap.nagoya-u.ac.jp/sosui/sosuisignal/SOSUIsignalDB/.

View full abstract

Secondary structure breakers, particularly hairpin structures, impose strong limitations on the global structure of a protein. Three kinds of secondary structure breakers (proline, glycine and amphiphilic residues) were studied in myoglobin and hemoglobin, which are typical all-α type proteins. Secondary structure breakers were located as predicted in about two thirds of the interhelical loops. Charge symmetry analysis provided evidence that high charge symmetry drives the formation of hairpin structures. Based on this information about breakers and hairpin structures, the possibility of folding a protein in silico is discussed.

View full abstract