ABSTRACT
The gene encoding stage is a fundamental and challenging step in computational biology for gene annotation. The aim of gene encoding scheme is to convert a DNA sequence into a discrete numerical sequence. The encoding process assigns a mathematical descriptor to nucleic acids. The assignment is done based on statistical properties of the structure of nucleic acids. The process of assignment may be dynamic or static. If the descriptor value varies with respect to sequence then it is termed as dynamic gene encoding scheme otherwise treated as a static or fixed gene encoding scheme. In this paper, the dynamic gene encoding schemes devised for the identification of protein coding regions are studied. The comparative analysis is done based on sensitivity, selectivity and accuracy.
