Identifying Antimicrobial Peptides in Silico

瀏覽統計 Email 通知 RSS Feed

簡歷

基本資料

Project title

Code/計畫編號

NSC102-2221-E019-060

Translated Name/計畫中文名

抗菌胜太的電腦辨識模組

Project Coordinator/計畫主持人

Kuan Y. Chang

Funding Organization/主管機關

National Science and Technology Council

Co-Investigator(s)/共同執行人

王建國

Department/Unit

Department of Computer Science and Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=3087428

Year

2013

Start date/計畫起

01-08-2013

Expected Completion/計畫迄

31-07-2014

Bugetid/研究經費

603千元

ResearchField/研究領域

資訊科學--軟體

"預測抗菌胜肽是重要的。隨著具抗藥性的微生物不斷蔓延，可具醫療用途的抗菌胜肽提供一種戰勝細菌、黴菌、病毒及寄生物的希望。此外，由於日益豐富的蛋白體資料，使得利用電腦模組去發掘天然的抗菌胜肽更為可行。本計畫要研究抗菌胜肽以三個層面: 序列、物理化學及結構特性。我們要建立一套系統，其以一組分別運用隱藏馬可夫模組、支持向量機、隨機森林、強化決定樹與線程的預測程式來預測。此系統表現會被評估。為避免過度訓練，會執行交叉驗證。最後，一海洋生物的蛋白體會被檢驗。被預測最可能是抗菌胜肽的序列會被合成。其是否具抗菌力會由實驗證明。本計畫的四個研究目標如下:  目標一: 檢驗蛋白質保守區域與抗菌胜肽活化區的關係。  目標二: 利用強化決定樹建立一套新的電腦模組，並評估此模組的表現。  目標三: 檢驗抗菌胜肽的結構，並將其特性融入預測系統。  目標四: 探索一海洋生物的蛋白體，利用此預測系統，辨識未知、天然的抗菌胜肽。""Predicting antimicrobial peptides (AMPs) is important. With multidrug-resistant microbes prevailing, AMPs, which could be used as therapeutic agents, offer a hope to fight against bacteria, fungi, viruses, and parasites. Besides, abundant proteome data available to the public makes the discovery of novel naturally occurred AMPs using computational approaches feasible. In this project, we plan to study AMPs in terms of three perspectives: sequence, physicochemical, and structural properties of the AMPs. We will develop a system with a panel of the predictors using hidden Markov model, support vector machine, random forest, boosted decision tree, and threading. The performance of this system will be evaluated. To make sure no over-training, cross-validation analysis will be done. Eventually, a selected marine proteome will be examined by the system. The top predicted AMP sequences will be synthesized and the antimicrobial ability of these sequences will be verified experimentally. Four specific aims of this proposed project are as follows: Specific Aim 1. Examine the relationship of protein domains with the active sites of AMPs. Specific Aim 2. Establish a novel computational model using a boosted decision tree and evaluate its performance. Specific Aim 3. Examine the structural aspects of AMPs and integrate the properties into the prediction model. Specific Aim 4. Explore the selected marine proteome to identify novel naturally occurred AMPs."

DSpace CRIS

Identifying Antimicrobial Peptides in Silico

基本資料

Description