Research On Functional Module Detection From PPI Networks Based On Cultural Algorithm

Protein-protein interaction(PPI)network is a complex network of interacted proteins in the living organism.The protein functional module is a collection of proteins that perform a life activity through interactions in PPI network.Accurately detecting functional modules in the PPI network can help people understand the functioning of the life and explain the pathogenesis of the disease.Therefore,to identify the protein function modules effectively,the computing methods which based on clustering emerge in an endless stream.Among them,to detect functional module,the method of group intelligent mechanism has a higher quality.In this paper,Aiming at the problem of PPI network function module detection,some research work based on the double evolution mechanism of cultural algorithm(CA)are carried out from the following two aspects:(1)As a new and effective swarm intelligent algorithm,CA is possible to obtain better results in PPI network function module detection.Therefore,detecting functional module method based on cultural algorithm in PPI networks is proposed.Firstly,an ordered adjacency list encoding scheme was used to model an individual which represents a problem of detecting functional module in the population space.Then,the evolutionary mechanism of cultural algorithm was designed and employed to obtain the optimal solution,where the upper mechanism simulated the evolution of the group experience in the belief space,and the lower mechanism described the evolution of individuals in the population space.The evolution of these two relatively independent spaces is communicated by accepting functions and influencing functions.Finally,by decoding the optimal individuals,the protein functional modules are obtained.Experimental results on three datasets show that the CA-FMD method has obvious advantages in some evaluation metrics compared with other algorithms.(2)To overcome the shortcoming that culture algorithm easily falls into local optima in detecting functional modules of a PPI network,a new culture algorithm with clonal selection strategies for detecting functional modules in PPI networks(called as CSCA-FMD)is proposed.This algorithm takes the cultural algorithm as the basic framework and the population space is redesigned.Three clonal selection strategies in clonal selection algorithm: clone,mutation and selection are incorporated into population space.Firstly,the algorithm determines the individuals cloned according to the cloning probability and determines the number of clones based on the value of the module's density values,and then enlarges the population size by cloning.Then,on the expanded population,the cloned individuals who are mutated are selected according to the decreasing probability of mutation with the increasing algebraic algebra,and according to the similarity information between the nodes,each bit of the selected cloned individual adaptive variation,so that the population has a rich diversity in the early evolution and speed up the convergence rate by reducing the variation in the late evolution.Finally,an individual which is selected from each individual and its derived individuals(clone,mutation)according to the selection probability enters the next generation population by selected operation,so that the population is diversified without degeneration.The experimental results show that CSCA-FMD algorithm not only can obtain better solution performances than that of CA-FMD algorithm,but also has some advantages compared with some other classical algorithms.