辨識微RNA標靶基因的相對R平方法及探討微RNA與標靶基因之間tissue特性的演算法

Abstract

近期的研究中指出了一個小且非編碼RNA，稱之為微RNA（miRNA），它可以降解調控其標靶基因，並被認為在許多生物歷程中扮演著重要角色。近年來，許多研究致力於發現新的微RNA和其標靶基因的辨識。雖然學者已經預測了許多的微RNA，但實際通過實驗驗證的微RNA卻非常少。為了加速微RNA的標靶基因預測結果以提供實驗驗證，文獻中已經利用序列分析(sequence analysis)及微陣列表達分析(microarray expression analysis)的方法來預測微RNA的標靶基因。此篇論文聚焦於人類微RNA靶標基因的預測，並提出了廣義相對R平方法（RRSM）找到許多高可信度的靶標基因，並且證明這個方法是優於現有的其他方法。

雖然廣義相對的R平方法(RRSM)提供許多高可信度的標靶基因並從過去的文獻研究中獲得證實，但是廣義相對的R平方法所設定臨界值為固定的常數，是無法依據不同基因的特性來改變。為了找到可以更適切地應用在真實資料的方法，我們提出了一種建立在相對R平方法的統計量的分布上並可以根據不同變數來選取臨界值的方法，根據不同變數來選取臨界值的方法所得到的預測結果成功地改善了過去建立在臨界值為固定常數的相對R平方法所得的預測結果，並在TarBase 和 miRTarBase兩種資料庫中獲得驗證。

此外，我們發現對於一個微RNA而言，它及其標靶基因只在特定的tissue中有作用。在文獻中並沒有一套有系統的方法透過微陣列(microarray)的資料來探討微RNA及其標靶基因是否在特定的tissue中有作用。在我們的研究中，利用已被實驗證實的微RNA及其標靶基因提出演算法來探討微RNA及其標靶基因在哪些特定的tissue中有作用。而我們的方法所得到微RNA及其標靶基因在特定的tissue中有作用的結論跟文獻所提的理論一致。

Recent studies have revealed a small non-coding RNA, microRNA (miRNA) down-regulates its mRNA targets, which is regarded as an important role in various biological processes. In recent years, there have been many studies concentrated on the discovery of new miRNAs and identification of their mRNA targets. Although researchers have identified many miRNAs, few miRNA targets have been identified by actual experimental methods. To expedite the identification of miRNA targets for experimental verification, in the literature approaches based on the sequence or microarray expression analysis have been established to discover the potential miRNA targets. We focus on the human miRNA target prediction and propose a generalized relative R squared method (RRSM) to find many high-confidence targets, which is shown to be superior to some existing methods. Although many high-confidence targets from RRSM have been confirmed from previous studies, the thresholds of RRSM are set to be fixed constants, which do not depend on the characteristic of a gene. To find a more feasible method for real data applications, we propose a variable threshold selection method based on the distribution of the relative R squared statistic, which is shown significantly improvement of the prediction results of RRSM only based on a fixed threshold criterion. In addition, we show that the interactions may be only functional on some specific-tissues which depend on the characteristic of a miRNA. There have been no systematic methods established in the literatures to investigate the relationship between miRNA target interactions and tissue specificity through microarray data. In this study, we propose an algorithm to investigate tissue-specificity of miRNAs based on experimental miRNA target interactions. The tissue-specificity result by our method is in accordance with the literatures.