Development of a sequence-based prediction method of interacting protein partners for drug target discovery

2016 Fiscal Year Final Research Report

• Project/Area Number: 26870045
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Life / Health / Medical informatics; System genome science
• Research Institution: National Institutes of Biomedical Innovation, Health and Nutrition (2015-2016); Tohoku University (2014)
• Principal Investigator: Murakami Yoichi 国立研究開発法人医薬基盤・健康・栄養研究所, 医薬基盤研究所 創薬デザイン研究センター, プロジェクト研究員 (20548424)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: タンパク質間相互作用 / タンパク質機能 / 機械学習 / 創薬支援

Outline of Final Research Achievements

This study applied previously developed PSOPIA, a method for predicting an interaction between two proteins, for developing a new computational method to predict proteins that interact with a given protein. The new method was trained on a highly imbalanced and large training dataset of human protein-protein interactions and successfully achieved higher prediction performance than PSOPIA. I have also developed a fast system that can automatically make predictions of interactions between a given protein and all the proteins registered in a public database. Furthermore, I have developed a graphical user interface with the new method developed in this study for identifying drug targets and biomarkers.

Free Research Field

バイオインフォマティクス