2020/09/30(Wed) 11:00 -12:00 視訊會議室(圖書館) 3F, Video Conference Room(Library)
Title
Network analysis of the protein system
Speaker
Dr. Geng-Ming Hu (National Taiwan Normal University)Abstract
Network analysis is a useful and common method to help people capture the specific systemic features from a complex system such as a biological system or social system. Here, I would like to introduce our recent study about the protein sequence similarity network of the GPCR family and hydrolases family. We delineate an unsupervised clustering algorithm, minimum span clustering (MSC), apply it to study the GPCR network of 2770 GPCR sequences and the hydrolases network of 3973 hydrolases sequences. The clustering result is visualized by the closest connection relation and minimum spanning tree (MST) algorithm. The clustering and visualization result shows a high detection accuracy and robust relation among sequence/structure/function information. In the GPCR case, we find the MSC result of the main subfamilies can be well explained by estimating the selective pressure of GPCR sequences, which related to the evolution information. We also propose an interactive graph database website, SeQuery, for intuitively visualizing proteome networks of GPCR by integrating the sequence/structure/function information of protein or gene sequences. In the hydrolases case, we try to use the high detection accuracy property to search the new candidate sequence of rare and useful enzymes such as PETase.
Language
演講語言 (Language): in English