The Analysis Of Characteristics Of Parasitic Proteins And The Building Of Clustering Model

Nematode parasites do not multiply in the definitive host but molt, grow and mature for a period following infection. In order to survive in their hosts, long-lived nematode parasites have to subvert the host immune defense, using strategies honed during their long co-evolutionary interaction with the host immune system. Excretory-secretory (ES) products are supposed to play a vital role in parasitm. If there is a screening or predicting model for parasitic proteins, it will be helpful for parasitic proteins research. We analyzed the characteristics of the ES products to build up a clustering model of parasitic proteins, which will be the foundationg of the screening model building.Conceptually, several mechanisms affecting evolution to parasitism can be envisioned, including: adaptation of pre-existing genes to encode new functions; changes in genes regulating metabolic or developmental pathways; gene duplication; gene loss; acquisition of genes from other species via horizontal gene transfer. More and more evidences show that the evolution history of secreted parasitic proteins is likely to be that of recent adaptation from ancestral genes that fulfilled other functions in free-living ancestors, we wondered whether this theory works extended in signal peptide-bearing sequences. Are these secretory proteins undergoing accelerated evolution? Taking a genomic approach to characterize potential secreted products, we analyzed putative protein sequences from Brugia malayi whole genom shotgun sequencing project. SignalP analysis was applied to predicted protein sequences to identify potential signal peptides and anchors. We analyzed 552 sequences, of which 88(15.9%) bear predicted signal-sequence coding regions. Comparisons of sequences with homologs from other species, we found that although some of the sequences with signal sequences have not got homologs with others, there are almost the same amounts of the sequences with signals that are highly conserved. It seems that this theory that proteins with N-terminal signal peptides are undergoing accelerated evolution could not work in an extended field. We recommend that further research be needed in this theory.Since rapid evolution of secreted gene products was not evident in greater diversity among signal peptide-bearing sequences than among genes coding for non-secreted proteins. We put our attention to the massage of whole sequence containing. We chose specific amino acid sites which conserved among parasitic protein sequences but different from non-parasitic sequences according to results of multiple alignments. Then we defined the amino acids with numerical value from 1 to 20 respectively. We did clustering analysis, finally built up a clustering model for parasitic protein sequences: using the numerical values for each amino acids of the specific sites of parasitic protein sequences as variables, measured by Pearson correlation and calculated by between groups method to build up the clustering model. We analysed 28 parastic protein families, with our definition of the variables, 24 of them could be clustered successfully, the accuracy rate was 85.7%; 18 of them could be clustered by our model successfully, the accuracy rate was as high as 64.3 %.