蛋白質分子與分子間充滿各種力場的鍵結，在進行分子定位之前須先確定小分子的構形，除了可透過NURBS 之幾何計算找到小分子的最小能量構形外，再使用李亞普諾夫 (Lyapunov)篩選最近全域穩定的對接點使其docking 完成後，分子間呈現穩定度最佳與能量最小，進而加速藥物設計的時程。使用電腦硬體最新的計算技術(HT )來進行藥物設計的複雜計算，並且改進分子動力學計算的方式與藥物資訊的呈現是本研究的宗旨，透過分子模擬三維虛擬圖形顯示，使我們將藥物資訊以視覺化的方式來分析和研究。電腦輔助藥物設計最大的挑戰是分子對接過程。本計畫以Lyapunov 函數中的穩定理論來降低接合位置數以便進一步增進分子對接的效能。所提出的方法是創新的而且有著能量最小化及最佳幾何搜尋為基礎的研究來加速證明蛋白質與藥物前趨物的互動穩定度。本計畫以Lyapunov 的穩定理論來探討零輸入分子動態系統方程式之初始狀態及分子離開平衡點與經過時間t 趨近無限大時所產生的分子狀態軌跡的變化。在之前的研究中我們已經發展以Lyapunov 漸進穩定定理來判別蛋白質分子摺疊的穩定度並加速藥物分子對接的計算。使用Lyapunov 函數時，當特徵值小於一個收斂值，這個分子系統是穩定的，若t 趨近無限大且收斂速率正比Lyapunov 函數則此分子系統稱為Lyapunov 指數漸進穩定，而且也趨近於全體能量最小值。據此用電腦輔助幾何設計NURBS 成形及搜尋技術來找出全體最小能量處與最短路徑來加速分子對接的完成。藉由scoring 函數演算法來決定分子系統的最小能量結構，此scoring 函數演算法包括摺疊力場、焓、能量函數等，Lyapunov 函數及最佳幾何搜尋法也包括在其中。在本計畫亦研究各種蛋白質的摺疊和計算力場之應用，此皆為創新的技術。根據先前我們所出版的研究論文使用WebDeGrator 繪圖系統來建立分子對接過程中的電腦模型描述蛋白質功能的三維結構，亦驗證一些重要的蛋白質摺疊動力學實驗及藥物分子對接計算範例，並且成功地引用Lyapunov 函數與分子動力學研究來支持這些論點。最後，在本研究將對最佳解，分子對接及蛋白質摺疊相關議題也將詳細分析與研究。跨越各種不同研究領域，例如結合生物學，資訊科學，系統工程學及化學等來解決生物資訊的各種新興的議題將是未來最強有力的解決方法。This proposal will employ Lyapunov theory and minimum energy theorem to investigate the stability of the drug docking and to improve the techniques of the computer-aided drug design (CADD). Due to various force-field among protein molecular and ligand molecular, we have to determine to the ligand molecular conformation before proceeding with the drug-docking process. The research will utilize the NURBS geometric calculation to find out the conformation of minimum energy and shortest path. By using Lyapunov theory to sieve out these binding sites with global minimum energy, which gets the docking finished, makes the molecular optimal stability, and accelerates the drug design process. In the project, we will use new hardware technologies for computing the ligand drug-docking such as Hyper-threading, Hyper-transport. To improve the calculation of molecular dynamic and visualize drug information are the goal of this research. By using 3D molecular virtual reality simulation to visualize drug docking and information will be the main purpose of this research. This work will present novel computer graphics and computational approaches for solving the challenges of the computer-aided drug design (CADD). From NCBI’ 97 reports, the most significant challenge is the docking procedure in CADD. The research will propose the Directivity Genetic Algorithm (DGA) to accelerate the binding sites search, and employs Lyapunov’s stability theorem to decrease the number of binding sites for enhancing the docking performance. These novel techniques are significantly based on the application of minimum energy and optimal geometry search strategies to prove the stability of protein – ligand interactions. This study will also discuss the folding of various proteins and calculate the application of force field by using the WebDeGrator system which is utilized to establish the molecular computer modeling for the docking process. Examples of protein folding kinetics and the drug-docking computations will be demonstrated. Finally, the optimal solutions, molecular docking, and protein folding kinetics are also discussed herein. The research will integrate various research fields such as combining biology, information science, system engineering, and combinational chemistry et al, which will solve the problem of bioinformatics and the computer-aided drug design (CADD). 研究期間：10008 ~ 10107
