一种用于基因调控网络建模的CGP-WPSO混合算法

摘要

依靠基因调控网络来预测农作物的表现型,对于保障全球的粮食安全有着极其重要的意义.提出了一种基于笛卡尔遗传规划(Cartesian genetic programming)和线性递减惯性权重粒子群优化(linear decreasing inertia weight particle swarm optimization)的混合算法,用于基因调控网络的建模.进一步,为了验证算法的有效性,将算法应用于拟南芥开花调控系统的模型重建问题.最后通过计算机仿真实验表明,该算法能够根据农作物的基因型和环境情况,重建出能够较精确地预测农作物表现型的基因调控网络模型.%The phenotype of the crops can be predicted through the gene regulatory network (GRN) , which is important for the global food security. This paper proposed a Cartesian genetic programming and linear decreasing inertia weight particle swarm optimization algorithm for GRN modeling. To verify the effectiveness of the proposed algorithm, we applied it to the recovery of the Arabidopsis flowering time control system. The computer simulation indicates that our proposed algorithm is able to infer the GRN model which can predict the phenotype of the crops fairly accurately based on its genotype and environmental conditions.