Time/Date: 13:00-15:00, 08 Dec
Location: Room 304, Research Building #5 (総合研究5号館).
http://www.media.kyoto-u.ac.jp/ja/access
First slot:
Talk: “Approximation algorithms for the stable matching problem” (in Japanese)
Speaker: Associate Prof. Shuichi Miyazaki, (Okabe Lab)
Abstract:
The stable matching problem is the problem of finding a “stable” matching for given preference lists of participants. This problem is used in many situations, such as assigning residents to hospitals, university students to laboratories, and so on. In this talk, I give some basic introduction of the stable matching problem, as well as the results on the approximation algorithms for it.
Second slot:
Talk: “Prediction of protein residue contacts and structure-function analysis for time-series data of protein structure.” (in Japanese).
Speaker: PhD candidate, Mayumi Kamada, (Akutsu Lab.).
Abstract:
Proteins have an important role in life activity. They can serve their functions only by forming their structures and interacting with other molecules. Since interaction between protein residues is a clue for understanding these mechanisms, many methods have been developed to predict protein residue contact from amino acid sequence. In this talk, I will introduce our prediction method for predicting protein residue contacts based on co-evolution information using random field model. Proteins are dynamic molecules within the cell, and the dynamics is thought to play a crucial role for structure change and interaction between molecules. For correlation analysis between structure motion and protein function, we extract motion features from time-series data of protein structure using wavelet transform. I will show some analyses using the features.