Identification of overlapping protein complexes by fuzzy K-medoids clustering algorithm in yeast protein-protein interaction networks

T-S fuzzy model identification based on an improved interval type-2 fuzzy c-regression model

Fuzzy clustering has been widely applied in T-S fuzzy model identification for nonlinear systems, however, tradition type-1 fuzzy clustering algorithms can’t deal with uncertainties in real world, an improved interval type-2 fuzzy c-regression model (IT2-FCRM) clustering is proposed for T-S fuzzy model identification in this paper. The improved IT2-FCRM adapts a new objective function, which makes the boundary of clustering more clearly and reduces the influence of outliers or noisy data on clustering results. The premise parameters of T-S fuzzy model are upper and lower hyperplanes obtained by improved IT2-FCRM, and the upper and lower hyperplanes are used to build hyper-plane-shaped type-2 Gaussian membership function. Compared with the hyper-sphere-shaped membership function of tradition IT2-FCRM, the hyper-plane-shaped membership function is more coincided with point to plane sample distance described by FCRM clustering. The simulation results of several benchmark problems and a real bed temperature in circulating fluidized bed plant show that the identification algorithm has higher accuracy.

Uncertainty, imprecision, and many-valued logics in protein bioinformatics

Understanding proteins, their structures, functions, mutual interactions, activity in cellular reactions, interactions with drugs, and expression in body cells is a key to efficient medical diagnosis, drug production, and treatment of patients. Machine learning and data exploration methods supported by many-valued logics allow to grasp the imprecision and uncertainties that naturally occur in proteins and other biomolecules. Many-valued logics, like Łukasiewicz logic or fuzzy logic, are non-classical logics that do not restrict the number of truth values to only two values of true or false, but they allow for a larger set of truth degrees. In this paper, we briefly review the use of many-valued logics, especially the fuzzy logic, in bioinformatics. Then, we focus on protein bioinformatics, and present selected applications of many-valued logics in the analysis of complex protein structures, including; (1) potential-based protein similarity searching, (2) matching proteins on the basis of secondary structures, (3) 3D protein structure alignment, (4) prediction of intrinsically disordered proteins, and (5) fuzzy querying in large collections of Big macromolecular Data. Results of presented studies show that the utilization of many-valued logics can enrich the investigations of protein molecules, in which uncertainty and imprecision are prevalent problems. The paper discusses all observed benefits brought by the application of many-valued logics in investigations related to selected protein analyzes carried out by the author.