Abstract
中文摘要蛋白質褶疊的研究一直是從事電腦模擬的生命科學家最大的難題，不旦是因為目前對於蛋白質褶疊的機制缺乏一致的看法，同時巨大的蛋白質分子更對電腦的計算速度提出嚴苛的挑戰。在這本論文中，我們並不打算繼續創造更多蛋白質疊合的機制，我們的重點是在於如合減輕電腦計算的負擔，使得電腦模擬蛋白質分子的運動更有效率。我們創造了一種簡化的蛋白質分子的表達方式，是將每一個氨基酸的側鏈簡化成一個單一代表性的原子，這個代表性的原子保有原來整個側鏈的分子量，及電荷。整個蛋白質經過如此的簡化之後，可以使得整個計算的時間減少一個次方以上，一般個人型的工作站即可勝任。以這個方式通過煩雜的計算之後，再將原本的側鏈還原，以計算這兩種模型之間能量上的差異。運用這個方式來模擬蛋白質分子的反褶疊，可以得到令人滿意的結果。AbstractProtein folding has been one of the biggest issues instructural biology during the years. Many folding mechanismshave been proposed to explain why protein molecules fold so fastwhen little time is given. However, formatting such kind oftheory requires many calculations and is very CPU-intensive,usually supercomputers are involved. To get ride of thislimitation, we develop a fast calculation routine which can berun on a personal workstation, and yield satisfying results.Simplified protein model, which we are using, cuts off the atomnumber participated in simulation, while still giving physicalmeaning for the structure. In this model, side chain atoms arereduced to a single sphere having the same mass and charge andthen go through the entire molecular dynamics simulation.Afterwards, complete side chain configuration was restored andgive the complete energy of the structure status. With thismanipulation, unfolding simulation can be conducted within aweek. This procedure is easy to be implemented and worth furtherdevelopment.