人脑发育基因芯片数据的定量关联规则挖掘

Mining quantitative association rules from brain development microarray data

目的探讨对脑发育过程有重要影响的基因及其之间的相互关系。方法采用R!KERT提出的基于半空间的定量关联规则挖掘方法对基因芯片数据进行分析。结果共挖掘出18条最优的定量关联规则,涉及到14条基因,其中有11条基因参与了发育过程,9条基因参与了神经系统的发育,4条基因参与了中枢神经系统及脑的发育;根据关联规则构建了基因之间的作用关系网络图。结论定量关联规则能够从基因芯片数据中挖掘有价值的信息,并且可以为进一步的研究提供信息。

[Objective] To discover important genes related to brain development and to infer their relationships. [Methods] The quantitative association rule based on half-space proposed by Rukert was used to analyze the brain development micmarray data. [Results] Totally 18 best quantitative association rules involving 14 genes were found, among which 11 genes were related to development, 9 genes to nervous system development, 4 genes to central nervous system or brain development. In addition, an interaction graph between genes was inferred from the 18 quantitative association roles. [Conclusion] Quantitative association nile may help to find out valuable information from micmarray data. Also it helps to formulate hypotheses about the function of genes for further research.