本文選題：時序基因表達數(shù)據(jù)　切入點：基因調(diào)控網(wǎng)絡(luò)　出處：《西安電子科技大學(xué)》2015年碩士論文　論文類型：學(xué)位論文

【摘要】：基因芯片技術(shù)的應(yīng)用使得快速獲得大量基因表達數(shù)據(jù)成為可能,進而為生物信息學(xué)研究提供了必需的數(shù)據(jù)庫,極大地推動了基因數(shù)據(jù)的研究。細(xì)胞的生命活動與細(xì)胞內(nèi)所有基因的表達水平有關(guān),一個基因的表達并不是孤立的,它的表達水平不僅受到其它基因的影響,同時還可能影響其它基因的表達水平,這些基因間的相互調(diào)控關(guān)系構(gòu)成了基因調(diào)控網(wǎng)絡(luò)。同樣地,基因調(diào)控網(wǎng)絡(luò)是眾多基因復(fù)雜調(diào)控的體現(xiàn)。對基因調(diào)控網(wǎng)絡(luò)進行研究能夠幫助我們了解基因與基因之間或者基因及其產(chǎn)物之間的相互關(guān)系;對疾病基因調(diào)控網(wǎng)絡(luò)的分析,可以指導(dǎo)藥物研發(fā)和臨床合理用藥,從而為復(fù)雜疾病的診斷與治療提供有價值的信息。所以重構(gòu)和分析基因調(diào)控網(wǎng)絡(luò)是十分必要的。傳統(tǒng)的調(diào)控網(wǎng)絡(luò)構(gòu)建方法大多是借助各種網(wǎng)絡(luò)模型,如布爾網(wǎng)絡(luò)模型和貝葉斯網(wǎng)絡(luò)模型等等,但這些模型忽略了基因表達數(shù)據(jù)的時序性特征。本文致力于研究利用時序基因表達數(shù)據(jù)重構(gòu)基因調(diào)控網(wǎng)絡(luò)的算法,并提出基于Granger因果關(guān)系的基因調(diào)控網(wǎng)絡(luò)構(gòu)建新算法,以展現(xiàn)基因間的因果調(diào)控關(guān)系。算法的提出基于以下基礎(chǔ):(1)時序基因表達數(shù)據(jù)作為一種特殊的時序數(shù)據(jù),時序數(shù)據(jù)的分析處理方法同樣適用于基因表達數(shù)據(jù)的處理。(2)基因的表達水平是自身表達與基因間相互調(diào)控的體現(xiàn),解讀基因的表達水平可以分析基因間的調(diào)控關(guān)系。同時,Granger因果關(guān)系是分析時序變量因果性的有效工具,可以應(yīng)用于基因表達水平的分析。本文主要涉及自回歸模型和Granger因果關(guān)系等理論知識,我們將由多基因組成的時序基因數(shù)據(jù)出發(fā),利用自回歸模型抽取基因前后時間表達值互相依賴的動態(tài)特性,并利用譜分析的知識將時序數(shù)據(jù)映射到頻域達到降低維度和消除時間對因果性的影響,通過計算基因間兩兩Granger因果調(diào)控關(guān)系,建立包含所有基因在內(nèi)的基于Granger因果關(guān)系的基因調(diào)控網(wǎng)絡(luò)。通過在大腸桿菌和酵母菌模擬數(shù)據(jù)上實驗,驗證了自回歸模型適用于時序基因表達數(shù)據(jù)。在釀酒酵母菌真實基因表達數(shù)據(jù)上實驗,與NIR算法比較基于Granger因果關(guān)系的構(gòu)建算法具有較高的識別率。
[Abstract]:The application of gene chip technology makes it possible to obtain a large number of gene expression data quickly, which provides the necessary database for bioinformatics research. The activity of cell life is related to the level of expression of all genes in the cell. The expression of one gene is not isolated, and its expression level is not only affected by other genes. It may also affect the level of expression of other genes, which form a genetic regulatory network. Gene regulation network is the embodiment of many gene complex regulation. The study of gene regulation network can help us understand the relationship between gene and gene or gene and its products, and analyze disease gene regulation network. Can guide drug research and development and clinical rational use, Thus providing valuable information for the diagnosis and treatment of complex diseases. So it is very necessary to reconstruct and analyze gene regulatory networks. Such as Boolean network model and Bayesian network model, but these models ignore the temporal characteristics of gene expression data. A new algorithm based on Granger causality is proposed to show the relationship between genes. The algorithm is based on the following basic: 1) temporal gene expression data as a special temporal data. The analysis and processing method of time series data is also suitable for the processing of gene expression data. The expression level of gene is the embodiment of self-expression and interregulation between genes. The level of gene expression can be used to analyze the regulatory relationship between genes, and Granger causality is an effective tool to analyze the causality of temporal variables. It can be applied to the analysis of gene expression level. This paper mainly involves the autoregressive model and Granger causality and other theoretical knowledge, we will be composed of multiple gene sequence gene data, The autoregressive model is used to extract the dynamic characteristics of interdependency between time expression values before and after gene expression, and the time series data are mapped to frequency domain to reduce the dimension and eliminate the effect of time on causality by using the knowledge of spectral analysis. By calculating the Granger causality control relationship between genes, a Granger cause-and-effect regulatory network based on Granger causality was established, which includes all genes. The experiments were carried out on the simulated data of Escherichia coli and yeast. It is verified that the autoregressive model is suitable for the sequential gene expression data. Compared with the NIR algorithm, the construction algorithm based on Granger causality has a higher recognition rate in the experiments on the real gene expression data of Saccharomyces cerevisiae.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：Q811.4;O212.1

【相似文獻】

相關(guān)期刊論文 前10條

1 張家軍;蔡傳政;王翼飛;;基因調(diào)控網(wǎng)絡(luò)中的延滯動力學(xué)[J];應(yīng)用科學(xué)學(xué)報;2007年01期

2 郭子龍;紀(jì)兆華;涂華偉;梁艷春;;基因調(diào)控網(wǎng)絡(luò)的研究內(nèi)容及其數(shù)據(jù)分析方法[J];電腦知識與技術(shù);2008年15期

3 陳少白;羅嘉;;一類基因調(diào)控網(wǎng)絡(luò)的定性分析[J];南京信息工程大學(xué)學(xué)報(自然科學(xué)版);2010年05期

4 李慶偉;全俊龍;劉欣;;基因調(diào)控網(wǎng)絡(luò)研究進展[J];遼寧師范大學(xué)學(xué)報(自然科學(xué)版);2013年01期

5 葉緯明;呂彬彬;趙琛;狄增如;;少節(jié)點基因調(diào)控網(wǎng)絡(luò)的控制[J];物理學(xué)報;2013年01期

6 王沛;呂金虎;;基因調(diào)控網(wǎng)絡(luò)的控制:機遇與挑戰(zhàn)[J];自動化學(xué)報;2013年12期

7 易東,李輝智;基因調(diào)控網(wǎng)絡(luò)研究與數(shù)學(xué)模型的建立[J];中國現(xiàn)代醫(yī)學(xué)雜志;2003年24期

8 雷耀山,史定華,王翼飛;基因調(diào)控網(wǎng)絡(luò)的生物信息學(xué)研究[J];自然雜志;2004年01期

9 姜偉;李霞;郭政;李傳星;王麗虹;饒紹奇;;時間延遲基因調(diào)控網(wǎng)絡(luò)重構(gòu)的決策樹方法研究[J];中國科學(xué)(C輯:生命科學(xué));2005年06期

10 張晗,宋滿根,陳國強,駱建華;一種改進的多元回歸估計基因調(diào)控網(wǎng)絡(luò)的方法[J];上海交通大學(xué)學(xué)報;2005年02期

相關(guān)會議論文 前3條

1 熊江輝;李瑩輝;;基因芯片數(shù)據(jù)分析的新方法與基因調(diào)控網(wǎng)絡(luò)推理[A];全面建設(shè)小康社會：中國科技工作者的歷史責(zé)任——中國科協(xié)2003年學(xué)術(shù)年會論文集（上）[C];2003年

2 王亞麗;周彤;;大規(guī)模基因調(diào)控網(wǎng)絡(luò)因果關(guān)系的辨識[A];第二十九屆中國控制會議論文集[C];2010年

3 馮晶;許勇;李娟娟;;非高斯噪聲激勵下基因調(diào)控網(wǎng)絡(luò)的研究[A];第十四屆全國非線性振動暨第十一屆全國非線性動力學(xué)和運動穩(wěn)定性學(xué)術(shù)會議摘要集與會議議程[C];2013年

相關(guān)重要報紙文章 前1條

1 吳佳s，

本文編號：1570189