[[alternative]]利用計算機探勘DNA 晶片資料以推測基因調控網路

[[abstract]]In this paper, we propose a reverse-engineering strategy to predict the interactions between genes within a genetic regulatory network. In implementing this strategy, first of all, we suggest hypotheses to simplify the complexity of the modeling problem. The proposed network modeling is represented as a directed graph G= (V, E), associating interactions at the transcription level with those at the protein level. After that, we enumerate all possible models subject to the hypotheses. And then, we simulate the models computationally to identify candidate models whose simulated results are consistent with the input data. In proof of this method, we take a well-known genetic regulatory network in yeast Saccharomyces cerevisia as a sample to verify the proposed inference method. The result shows that one of the candidate models has identical interactions to the sample’s interactions, which is experimentally determined by other biochemical experiments. In conclusion, the computational inference method is useful and feasible that helps biologists save lots of time and money, compared with the traditionally biochemical experiments