Identification of data-permitted predictors of Boolean networks via observed data

Identification of predictors of Boolean networks from observed data is an earlier-stage work of determining Boolean networks via experiments. This paper aims to identify predictors that are consistent with the observed data from experiments. Based on the algebraic expression of Boolean works which is established by aid of a new modelling tool, called semi-tensor product of matrices which is proposed in recent years, we propose a method of partially determining logic functions of a node via observed experimental data. Further, a necessary and sufficient condition that judging whether a set of candidate nodes can be a predictor of a node is established. The approach is quite different from others such as computer-algorithm-based and provides a new angle and means to understand and analyze the structures of Boolean networks. Finally, an example is provided to demonstrate how the condition identifies all the possible predictors of a node.