Please use this identifier to cite or link to this item:
標題: A Simultaneous Recurrent Neural Network Model of Genetic Regulatory Network-Identifying Target Gene Regulators from Time Series Gene Expression Data
作者: 陳齊康
關鍵字: 數學類;基礎研究
Gene expression turns information encoded in DNA sequences into geneexpression products such as RNA or proteins. Genes in a living cell interacting witheach through expression products gives rise to the genetic regulatory network (GRN).Using the resemblance of formal principals of GRN and artificial neural networks, wepresent a one hidden layer simultaneous recurrent neural network (SRNN) model ofGRN. Using time series gene expression data, we infer based on the model theregulators of target genes of interest and reconstruct GRN. To objectively evaluate theregulator identification procedure, the time series gene expression data are generatedby an independent artificial GRN with well-defined regulatory and kinetic properties.

基因表現將DNA 序列中的資訊轉換成RNA 或蛋白質。在細胞中基因經由其表現的產物交互作用形成了基因網路。利用基因調控網路在原則上和類神經網路相似的原理,我們提出一個具有一隱藏層之同時回歸類神經網路的基因調控網路模型。經由使用時間序列基因表現資料,我們以所提出的模型為基礎推論所感興趣之目標基因的調控基因,並由此重建基因調控網路。為能客觀地評估所建立之目標基因調控子辨識程序,時間序列基因表現資料將由已知調控性質的人造基因調控網路所生成。
其他識別: NSC97-2118-M005-004
Appears in Collections:應用數學系所

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.