A single source reachable query method for gene regulatory networks

Abstract

With the deepening of research on gene regulatory network, the construction technology of gene regulatory network has been relatively mature. However, according to the characteristics of gene regulatory network, it is still a problem to realize efficient reachable probability query and reachable path query among gene pairs on the constructed gene regulatory network. Therefore, this paper proposes an efficient and accurate single-source reachable query algorithm GRN-RQ based on gene regulatory network, in which the meaning of single-source is: the reachable query between gene pairs with both starting and ending nodes being single nodes can be realized. Algorithm GRN-RQ mainly includes three parts, respectively, an active graph pruning algorithm AP, which is applicable to different node situations, is used to reduce the size of probability graphs, a concurrent reachable query algorithm PSRA for partial instance graph generation and staged graph search, which is used to improve operational efficiency, the dual nodes optimal reachable path query algorithm DOP based on the dual path evaluation rule DPER. Through experimental verification, the algorithm GRN-RQ proposed in this paper can not only quickly obtain the reachable probability between gene pairs and the optimal reachable path between them on the gene regulatory network, but also improve the accuracy of the reachable query result compared with the traditional method.