3D protein analysis based on the multiple view images

2014 Fiscal Year Final Research Report

• Project/Area Number: 25540062
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: University of Tsukuba
• Principal Investigator: FUKUI Kazuhiro 筑波大学, システム情報系, 教授 (40375423)
• Co-Investigator(Kenkyū-buntansha): SAIGO Hiroto 九州工業大学, 情報工学研究院, 准教授 (90586124)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Keywords: タンパク質 / ３次元構造 / 多視点画像 / 相互部分空間法 / 画像認識

Outline of Final Research Achievements

In this research project, we have proposed a new similarity measure for protein structure comparison that uses a set of multi-view visualization images of 3D protein structures. The advantage of our method is that distinctive structural information can be embedded in the visualization. In this approach, each set of multi-view images is represented by a subspace, while applying LBP feature extraction to the image set. The similarity between two protein structures is then characterized by the distance between two points which are corresponding to different two subspaces, on a Grassman manifold. The proposed method was evaluated by classification experiments on 7 protein classes based on SCOP (alpha, beta, alpha/beta, alpha+beta, multidomain protein, membrane, and small protein), using 700 randomly selected proteins. In the experiments, our method significantly outperformed the CE and FATCAT alignment methods. In addition, we have constructed a view-based protein comparison (VPC) system.

Free Research Field

コンピュータビジョン