Gene-Promoter-Sequence Recognition – an Interpretable and Accurate Fuzzy-Genetic Approach

摘要

Different methods applied to gene promoter recognition share, in general, the same drawback, i.e., they are non-interpretable black-box-type techniques. The main objective of this paper is the application of our fuzzy rule-based classification approach characterized by genetically optimized accuracy-interpretability trade off (using multi-objective evolutionary optimization algorithms (M-OEOAs)) to gene promoter recognition. Two publicly accessible bacterial DNA benchmark data sets, i.e., Molecular Biology (Promoter Gene Sequences) and iPromoter-FSEn benchmark data sets are considered. For comparison purposes, two M-OEOAs are used in our experiments, i.e., the well-known Strength Pareto Evolutionary Algorithm 2 (SPEA2) and our generalization of SPEA2 (referred to as SPEA3) characterized by a higher spread and better-balanced distribution of solutions. Our results for both considered molecular biology data sets are compared with the results of 16 alternative methods (including several state-of-the-art ones) demonstrating the advantages - in terms of system's accuracy-interpretability trade-off optimization - of our approach.