使用高通量定序測小RNA及mRNA降解率計算分析方法

Abstract

在小分子核糖核酸（miRNA）對信使核糖核酸（mRNA）的調控為目前已知的一種生物機制，其中這兩者的生命週期在機制中的影響亦不容忽視，所以透過了解小分子核糖核酸與信使核糖核酸的降解率，將更有助於了解其調控機制當中更細緻的行為。透過次世代定序的高解析度特性，我們可以利用一些實驗方法修改新生核糖核酸序列的核苷酸，使最終實驗結果出現T/C轉換，並以此作為標記，推算核糖核酸之降解率。故其實驗結果之資料分析工具，必須具有允許T/C轉換之高效能、高度容錯（insertion、deletion）之排比器

在本篇論文中，我們結合了各種序列排比之方法，包含 FM-index, Smith-Waterman演算法等，以及匹配計分策略，並且運用單指令流多資料流技術（SIMD）開發出新的排比器完成實驗分析之需求，同時設計完整的分析流程計算出小分子核糖核酸及信使核糖核酸之降解率。

我們也在模擬資料測試檢驗本論文所設計方法之效率以及準確度，並通過與各種現有之工具比較確認其在可用度及效能上皆展現出明顯優勢。最後，在真資料分析中本篇論文演示了完整的分析過程，並分別計算出特定的小分子核糖核酸及信使核糖核酸之降解率。

The interactive of miRNA and mRNA, such as post-transcriptional regulation is known biological mechanism. To understand the effect of the life cycle of these characters is important, so measuring the degradation rate of miRNA and mRNA can be a useful information to understand the detailed behavior of this mechanism. To measure the degradation rate, we replace the nucleotides of newborn RNAs from U to 4SU and cause the T/C conversion in NGS sequencing result. To achieve these requirements, an aligner which allows T/C conversion, high fault tolerant, fast and accurate is needed in the analysis tool. In this paper, we combine different kind of alignment method, including FM-index, Smith-Waterman, etc. We also propose an alignment scoring function and accurate the aligner use SIMD technology. Then, a pipeline for miRNA and mRNA degradation rate analysis was designed. The mappability and performance of our tools are verified in simulation test and shown a better ability than the others. The complete analysis workflow was demonstrated in real data analysis, and the degradation rate of specific miRNA and mRNA is computed.