Comparison and superposition of low-resolution density maps and atomic models using Gaussian mixture model

2016 Fiscal Year Final Research Report

• Project/Area Number: 26440078
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Biophysics
• Research Institution: Osaka University
• Principal Investigator: KAWABATA TAKESHI 大阪大学, たんぱく質研究所, 寄附研究部門准教授 (60343274)
• Co-Investigator(Renkei-kenkyūsha): SUZUKI HIROFUMI 大阪大学, 蛋白質研究所, 特任助教 (60418572)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 構造・機能予測 / 電子顕微鏡 / 単粒子解析

Outline of Final Research Achievements

3D density maps by electron microscopy have important information about biomolecular structures. To make a full use of these maps, we developed a WEB server “Omokage search” to search the global shape similarity of biological macromolecules in databases for 3D density maps and atomic models. The server searches using the distance-based profiles from the 3D points of the maps and models. It also performs 3D superimpositions using the Gaussian mixture model (GMM), for approximating shapes of the maps and the models. The new algorithm “Gaussian-input GMM” was developed for converting into GMMs more robustly and rapidly. We started to develop the methods for multiple subunit fitting and alpha-helix detection.

Free Research Field

構造バイオインフォマティクス