iGemdock 2.0 與 SiMMap: 篩選後分析方法於以結構為基礎之藥物設計

Abstract

以蛋白質結構為基礎的虛擬藥物篩選 (structure-based virtual screening) 常被用於從數百萬種化合物中識別找尋前導藥物，目前也已經有許多的成功的案例，但是對蛋白質和配體 (ligand) 的結合和生物機制不夠了解以及使用不精確的評分方法，使得尋找針對標靶蛋白(target protein) 識別的藥物仍是一個很大的挑戰。 本研究主要實作實驗室發展的兩個概念: (1)透過改進iGEMDOCK的篩選後分析模組 (post-screening analysis module) 識別來找出藥理交互作用 (pharmacological interactions)，(2) 以SiMMap網頁平台建構區域官能基地圖來描述與蛋白質-化合物結合機制相關的關鍵特徵。 為了提升虛擬藥物篩選的精確度和實用性，我們基於過去 GEMDOCK的核心與技術發展了 iGEMDOCK。iGEMDOCK是一個易於使用的圖形化界面整合環境，並利用篩選後分析模組，基於以蛋白質-化合物交互作用能量列表 (protein-compound interaction profiles) 來識別藥理交互作用，並且能夠對化合物做分群分析，最後結合藥理交互作用和以能量為基礎的計分方法為化合物重新排名。 為了找出觸發或阻止標靶蛋白的生物反應所必需的關鍵特徵，我們以SiMMap網頁服務建構區域官能基地圖 (site-moiety map)。利用多個錨點 (anchors) 來描述標靶蛋白和配體之間的關係，並可以提供如蛋白質結合位置 (binding site) 的物化特性 (physico-chemical properties)、對配體官能基 (moiety) 的偏好以及蛋白質袋槽與配體官能基間的交互作用力 (pocket-moiety interaction type)。我們相信 SiMMap能提供生物觀點並有助於藥物設計及前導藥物最佳化。

Structure-based virtual screening is often used to find lead compounds from the hundreds of thousands of compounds. There are already many success stories using structure-based virtual screening. However, because of the incomplete understanding of ligand binding mechanisms and the subsequently imprecise scoring algorithms, How to identify the potential drug molecules of a target protein is still a big challenge. We apply two concepts from our previous works: (1) Deriving pharmacological interactions by improving the post-screening analysis module of iGEMDOCK (2) Designing the SiMMap server to construct site-moiety maps for understanding the key feature of protein-compound binding mechanisms. In order to enhance the accuracy and usefulness of virtual drug screening, based on our core technology GEMDOCK, we developed the iGEMDOCK. iGEMDOCK provides an easy-to-use graphic and integrated environment, utilizing a post-screening analysis module to mine the pharmacological interactions based on protein-compound interaction profiles and to cluster the screened compounds. Finally, iGEMDOCK ranks and visualizes the screening compounds by combining the pharmacological interactions and energy-based scoring function of GEMDOCK. To identify the key features that are essential to trigger or block the biological responses of the target proteins, we presented the SiMMap server to infer the key features by identifying the site-moiety map with anchors. Furthermore, we describe the relationship between the target protein and ligand, moiety preferences and physico-chemical properties of the binding site and pocket-moiety interaction type (E, H or V). We believe that SiMMap is able to provide biological insights for drug discovery and optimization of lead compounds.