- 學位論文
整合圖形演算法與拓樸分類策略預測非微小型細胞肺癌之治療藥物及反序微型核糖核酸
Drugs Predictions and Therapeutic Anti-microRNA Identification for Non-Small Cell Lung Cancer by Integrating Graph Algorithm and Topological Classification
摘要
找出已知藥物應用在其他疾病上(藥物重新定位)已經是一種趨勢且已經有不少成功的案例。根據統計癌症是全球主要死亡的原因之一，其中又以肺癌的死亡率為最高，在過去的研究中，我們建立了一套基於拓樸參數圖形理論的方式預測治療非微小型細胞肺癌(NSCLC)的流程，而且獲得IC50實驗、臨床實驗以及相關文獻的支持。 在過去的研究中，我們已經提出了兩種計算方式(機器學習與拓樸參數分類策略)用於預測非微小型細胞肺癌的潛在藥物。在本文中，我們建立一套新的流程改進了治療非微小型細胞肺癌治療藥物預測的精確度，這個流程整合兩種圖形理論的驗證(拓樸參數分類策略與基於貪婪策略的圖形演算法)，並且從cMap資料庫中推導潛在的標靶治療藥物，再者本研究進一步推導治療NSCLC潛在藥物的標靶基因。 MicroRNA的治療方法代表的是一種新型的基因治療方式，而且已經有許多臨床證明治療方法具有相當的效果。本研究所提出的方法較以往的研究有更高的IC50實驗預測的準確度，本研究提出新的流程推導NSCLC潛在藥物與MicroRNA，預期此流程對治療NSCLC有相當的助益。
並列摘要
The idea of drug repurposing is a recently developed approach that attempts to identify new uses for existing drugs and has already yielded several successes. Lung cancer is one of the leading causes of death in the world. In the previous study we applied the topological graph theory to identify potential therapeutic drugs for non-small cell lung cancer (NSCLC) treatments that are supported by the literature in-vitro IC50 measurements and clinical trials. In previous study, we have proposed two computational approaches, machine learning algorithms and topological classification, for inferring potential drugs for NSCLC. In this thesis, we further established a novel pipeline to improve the drug-repositioning prediction accuracy for NSCLC. This pipeline integrates two computational approaches, i.e. graph algorithm based on greedy method and topological classification, to infer potential-repositioning drugs from the cMap database and obtained the drugs’ targeted genes for therapeutic treatment. MicroRNA-based therapeutics represents another class of drugs that are used in the course of gene therapy. The effectiveness of these drugs is supported by the literature, experimentally determined in-vitro IC50 and clinical trials. This work provides better drug prediction accuracy than competitive research in terms of IC50 measurements. The present study provides a feasible and novel strategy to discover potential drugs and microRNAs.