利用基因群架構整合式分析於藥物反應以及miRNA調控之分析

歸因於生物科技的純熟，現今生物研究多使用高通量偵測儀器來測量基因體表現量，如何分析遽增的高通量基因資料是一個重要的課題，而傳統的單基因分析方式已不敷需求使用，基因群的分析方法取而代之，成為探討生物機制的重要分析方法，不但能有更高的穩定性，同時也可以更有效率的將生物知識與基因體資料緊密連結，來滿足資料分析的需求。這篇論文中，我們利用基因群分類資料庫，來建置兩個系統分析以下議題作為應用:一、分析藥物反應。二、分析微核醣核酸的調控作用。第一個應用中，了解藥物反應是設計藥物的核心問題，但設計開發新的藥物不但費時更要花費巨額金錢，為此，舊藥新用提供了一個契機，加速藥物開發的過程並減少花費，但即便前人的研究成果利用單基因方式分析高通量基因資料，找到可能有新的處方的藥物，而作用機制仍然無法解釋，所以我們提出了一個分析系統，來建置藥物與癌症相關基因群的網絡，以乳癌為例，透過基因群分析藥物反應與病人的存活分析，藥物的資料來自CMAP (Connectivity Map) 資料庫，而病人的資料來自GEO (Gene Expression Omnibus)，將病人的表現量資料使用回歸存活分析 (Cox regression)，找出與病人死亡風險相關的基因群，再透過連結與死亡因素相關的基因群及其作用相關藥物，最後找出162種可能的候選藥物與172個存活相關的基因群相連，並舉一個幹細胞相關的基因群研究為例，展示所分析的6種藥物可能調控其表現，而該基因群的表現經研究發現與乳癌生成有關聯，可望找到治療乳癌的候選藥物的目的。微核醣核酸已被證實其調控基因表現與疾病相關的重要性，以互補配對的機制影響基因的表現，由演算法估算可能影響的基因高達人類三分之一的基因體，已證實當中可能對腫瘤及癌症生成造成影響，然而其實際的調控作用仍然需要更深入的研究探討，缺少有效率的方式來分析微核醣核酸表現資料，所以我們設計了一個以基因群為架構的分析系統，透過矩陣運算，將微核醣核酸的調控投影到生物的作用途徑上，經由一個微核醣核酸與基因的連結矩陣，反映互補配對的機率，再將其受調控的基因投影至含有基因調控的生物途徑矩陣上，最後反映出微核醣核酸對生物途徑的調控作用，以肝母細胞瘤的病人為例，利用該病人的微核醣核酸表現資料，分析出與Wnt相關的生物途徑，該途徑也在前人的研究中發現與該疾病的高度相關，除此之外也找到其他的相關途徑可能對其疾病造成的影響，對比於以往的研究，單獨探討單一微核醣核酸的作用機制，我們所提供的系統能夠有效地分析數個微核醣核酸交互作用後在生物途徑上的作用程度。我們所提供的系統，能有效分析基因體資料在生物功能上的可能作用機制，除此之外也保有相當高的彈性，藥物反應可以應用在不同的疾病，而微核醣核酸的分系系統，可以透過不同的矩陣代入，來提升整體的效能。

關鍵字

基因群分析；乳癌；微核醣核酸調控；基因調控；舊藥新用

並列摘要

With advances in microarrays, using high-throughput technologies to analyze the genome has become an essential part in biological studies. However, the traditional single gene based analysis is not able to meet the requirement of exploring biology mechanisms with efficiency and robustness. Therefore, gene set based analysis as an alternative strategy has prevailed in linking prior biological knowledge and genomic data to conquer this problem. In this study, we proposed two systems for comprehensively analysis 1) drug responses for efficient drug repositioning in cancers and 2) miRNA regulation based on analysis of functionally defined gene sets. Since the response rates of current anti-cancer drugs are not satisfactory and the fact that development of new drugs is time and resource consuming, seeking novel and effective drug repositioning strategy provides an opportunity for revealing of putative anti-cancer drugs. However, previous studies employing single gene based methods were limited in effectiveness and biological interpretation. Thus in the first sub-topic, we adopted the biologically meaningful “cancer-gene set-drug” scheme for discovering novel drug reposition. Taking breast cancer as demonstration, our system first analyzed enrichment of functional gene sets in the expression profiles obtained from the Connectivity Map project. Second, through Cox regression analysis of human breast cancer gene expression data deposited in the Gene Expression Omnibus (GEO) database, gene sets with prognosis significance were further identified. By linking results of the two parts, we identified 162 potential candidates of anti-cancer drugs and showed 172 survival associated functional gene sets had significant enrichment. Choosing a gene set derived from stem cell related study as demonstration, we find 6 new drugs can possibly increase the expression of the gene set and it is associated to breast cancer progression. On the other hand, miRNAs, for the ability to regulate expression levels of up to one third of genes encoded in the human genome, have been proved as crucial players in tumorigenesis and cancer progression. However, the biological effects of changes in miRNA expression profile are not comprehensively explored from a systematic aspect. Addressing this issue, in the second sub-topic, we proposed a miRNA enrichment analysis model for interpreting miRNA expression profiles with respect to functional gene sets. The model was implemented by a series of matrix transformation. Though a predicted miRNA-target gene matrix, the miRNA expression was first projected into enrichment of genes. Followed by another product transformation based on gene-gene interactions, the enrichment of genes was further transformed into enrichment scores of functional gene sets and pathways. As the result of analyzing public miRNA expression data set of a hepatoblastoma cohort, previously reported Wnt signaling pathway as well as other novel functional gene sets were found to be the enriched in hepatoblastoma patients. Different from conventional studies based on processing individual miRNA, our analysis system provided the tool to explore miRNA regulation in the functional level and interpret miRNA data in a more biologically meaningful way. The two systems provided in this study were shown as able to explore biological functions in high-throughput genomic data. Besides, the flexibility of the systems enables their applications in drug repositioning for other complex diseases and enhancing performance of miRNA enrichment analysis. Hence, researchers may find putative drugs for diseases or discover new mechanisms of miRNA regulated functions through our systems.

並列關鍵字

gene set enrichment ； breast cancer ； drug response ； drug reposition ； miRNA regulation ； gene regulation

參考文獻

1. Lu, T.P., et al., Identification of a novel biomarker, SEMA5A, for non-small cell lung carcinoma in nonsmoking women. Cancer Epidemiol Biomarkers Prev, 2010. 19(10): p. 2590-7.

2. Subramanian, A., et al., Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A, 2005. 102(43): p. 15545-50.

3. Li, J., X. Zhu, and J.Y. Chen, Building disease-specific drug-protein connectivity maps from molecular interaction networks and PubMed abstracts. PLoS Comput Biol, 2009. 5(7): p. e1000450.

4. Gottlieb, A., et al., PREDICT: a method for inferring novel drug indications with application to personalized medicine. Mol Syst Biol, 2011. 7: p. 496.

5. Luo, H., et al., DRAR-CPI: a server for identifying drug repositioning potential and adverse drug reactions via the chemical-protein interactome. Nucleic Acids Res, 2011. 39(Web Server issue): p. W492-8.

國際替代計量

利用基因群架構整合式分析於藥物反應以及miRNA調控之分析

全文下載

主題瀏覽