Prediction And Feature Analyses Of Multi-Functional Enzymes

Multi-functional enzymes(MFEs) are enzymes that Perform multiple functions. According to the mechanisms of multiple functions,the MFEs can be further sub-grouped into MFEs with multiple catalytic domains(MCD-MFEs) and MFEs with single multi-activity domain(SMAD-MFEs).MFEs are found to be beneficial to living systems and provide competitive survival edges in a variety of ways.They are able to employ alternative approaches to coordinating multiple activities and regulating their own expression,which demonstrates evolutionary advantage as part of a clever strategy for generating complexity from existing proteins without expansion of the genome.Combination of multiple functions enables an enzyme to act as a switch point in biochemical or signaling pathways so that a cell can rapidly respond to changes in surrounding environment.Therefore,characterization and identification of MFEs are critical for the better understanding of the molecular mechanisms underlying the crosstalk between different cellular processes.In this study,two support vector machines(SVMs) models were constructed separately for the prediction of MCD-MFEs and SMD-MFEs respectively.The models were trained and optimized using 3,120 annotated MFEs(positive data) derived from Swiss-Prot knowledgebase and 21,833 selected proteins from seed proteins of the domain families in Pfam database excluding those that contain at least one MFE(negative data).Every protein sequence was represented by specific feature vector assembled from encoded representations of tabulated residue properties including amino acid composition,hydrophobicity,normalized Van der Waals volume, polarity,polarizability,charge,surface tension,secondary structure and solvent accessibility for each residue in the sequence.Finally,2,641 novel MFEs were successfully identified from the ExPASy Enzyme database.To evaluate our predictions and further extract the structural,functional and evolutionary patterns of MFEs,further statistical analyses were demonstrated as well.It was found that MFEs are non-evenly distributed in species,and no solid evidences suggest complex life forms like human prefer more MFEs than simple life form like yeast.Based on currently available 3D protein structures,the alpha and beta fold topology was found to be mostly favored for MFEs.Further KEGG ontology(KO) analysis indicated that 90%of MFEs are well preserved in catalyzing several essential cellular processes like the metabolisms of carbohydrates,nucleotides and amino acids.Almost half of MFEs (MCD-MFEs:48.7%SMAD-MFEs:54%) were found to participate in only one biological pathway,whereas another half of MFEs participate in multiple pathways, up to five independent pathways.These suggest that MFEs most likely evolve from early enzymes in primitive life forms.They are well conserved during evolution; however,new MFEs or novel functions were diversified and specified in various forms of genetic variation like gene fusion or exon shuffling.A database was constructed as well in this study to provide comprehensive information of MFEs, which can be freely accessed by http://bioinf.xmu.edu.cn/databases/MFEs/index.htm...