以生物資訊最佳化方法發展電腦輔助疫苗設計系統

Abstract

本計畫為「以生物資訊最佳化方法發展電腦輔助疫苗設計系統」之三年期研究計畫， 主要目的是發展一套輔助免疫學家設計疫苗之電腦系統。本計畫之核心為發展數個能適用 於探勘各種生物免疫現象之重要特徵因子的高性能大量參數最佳化之生物資訊演算法，並 結合生物免疫知識庫發展成一整合型重要特徵因子探勘、分析、註記的應用系統。研發生 物資訊演算法包含了三個重要的步驟：（1）蒐集各種已知物理化學、演化與結構等能夠有 效解釋生物現象之特徵因子；（2）結合生物知識與演算法技巧來建立生物設計模型並轉換 成最佳化設計問題；（3）發展特定的高性能演算法來解決最佳化設計問題。整合式重要特 徵探勘系統可從大量資訊中探勘重要特徵來解釋各種生物現象，並應用於探勘各免疫反應 相關之重要特徵因子來建立電腦輔助疫苗設計系統。 本計畫擬以三年完成電腦輔助疫苗設計系統，每年之主題研究相輔相成，生物知識收 集與演算法設計並重，簡述如下： 第一年：發展核心之高效能重要特徵因子探勘演算法，並研究細胞毒性T 細胞相關免疫反 應之重要特徵。 第二年：研究輔助T 細胞致免疫性及B 細胞抗原決定部位之重要特徵與預測演算法。 第三年：發展免疫相關重要特徵資料庫並建構整合性電腦輔助疫苗設計系統。 本計畫之目標包括：（1）深入瞭解影響免疫反應之特徵；（2）建立電腦輔助疫苗設計 系統；（3）整合三年研究成果建立可供查詢之特徵資料庫。

This is a three-year project: Developing Computer-aided Vaccine Design Systems Using Bio-inspired Optimization Methods. The objectives are to develop computer-aided systems to help immunologist for vaccine design. The core project is two-fold: 1) to develop various high-performance optimization algorithms for solving large-scale parameter optimization problems of bioinformatics to mine informative physicochemical properties from known experimental data; and 2) to integrate immune knowledge base and application system for analysis and annotation of informative features. Developing these algorithms involves three important phase: (a) collection of various features including physicochemical properties, evolutionary information and structure information; (b) design of optimization problems by identifying system parameters to be optimized from combining bio-knowledge and computing techniques, and (c) design of powerful optimization algorithms for obtaining near-optimal solutions. The integrated feature mining systems can extract potentially good features from a large number of physicochemical properties in the biological database and apply them to vaccine design systems. The individual projects of the three year focus on both study of vaccine design and algorithm development, which are fully cooperated, described below. Year 1 : Develop the core high-performance feature mining system, and study cytotoxic T lymphocyte related immune response by mining informative features. Year 2 : Study the immune responses of T helper cell and B cell, and their prediction algorithms. Year 3 : Establish integrated computer-aided vaccine design systems, and develop an informative feature database of immune systems. The goals of this project are 1) identify informative features to further understand immune systems; 2) construct computer-aided vaccine design systems, and 3) develop an informative feature database of immune systems.