Acta Biomedica Scientifica (Nov 2014)
APPLICATION OF MATHEMATICAL METHOD PREDICTIONS FOR IDENTIFICATION OF PATTERNS RELATIONS MUTATIONS IN PROTEINS ENCEPHALITIS VIRUS AND A MANIFESTATION OF ITS PHENOTYPIC TRAITS
Abstract
We studied the natural connections between the amino acid sequences of proteins C, prM, E and NS1 virus strains of tick-borne encephalitis (TBE) and their three phenotypic traits -neuroinvasiveness, thermal stability and thermoresistance. Coupling strength is assessed using measures of competitive sequence similarity of each strain with reference strains. For such purposes subsets of strain sections are chosen amino acid composition specifics of which can predict the value of a phenotypic trait of interest. The possibility to predict missing elements in data both in amino acid composition, and in target properties is demonstrated. The relationships between pairs of phenotypic traits of strains were evaluated.