关键词: 人工智能, 引力场算法, 基因调控网络, 优化算法, 奇异值分解

Abstract: In order to resolve the inefficiency of reconstruction of gene regulatory networks (GRNs) in system biology, we proposed a novel inference algorithm from gene expression data based on differential equation model. In this algorithm, two methods were included for inferring GRNs. One is singular value decomposition method (SVD). The other is gravitation field algorithm (GFA). SVD is used to decompose gene expression data, determine the algorithm solution space, and get all candidate solutions of GRNs. GFA is the kernel part in our algorithm. GFA was divided into four parts: initialization, division of solution space, movement operator, absorption. Random group method was used in division of solution space. Every element movement method was used in movement operator. The proposed algorithm is validated on both the simulated scale-free network and real benchmark gene regulatory network in networks database. Both genetic algorithm and simulated annealing were also used to evaluate gravitation field algorithm. The cross-validation results confirmed the effectiveness of our algorithm, which outperforms significantly other previous algorithms.

Key words: artificial intelligence, gravitation field algorithm, gene regulatory networks, optimal algorithm, singular value decomposition