Elucidation of the mechanism responsible for diverse ligand recognition using feature extraction methods from machine learning

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K06139
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 43060:System genome science-related
• Research Institution: Suntory Foundation for Life Sciences
• Principal Investigator: Shiraishi Akira 公益財団法人サントリー生命科学財団, 生物有機科学研究所・統合生体分子機能研究部, 研究員 (50710729)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: ペプチド / MRGX / Substance P / BAM8-22 / 機械学習 / 特徴抽出

Outline of Final Research Achievements

We developed IDL Score, a method that can estimate residues contributing to interactions using PD-incorporated SVM, which predicts peptide-GPCR interactions based on the sequence of G protein-coupled receptors (GPCRs) and peptides. Using this method, we estimated the receptor residues that are important for the specific interactions between SP and MRGX2 and BAM and MRGX1, respectively. Based on the estimated residues, we constructed mutants of MRGX1/2 and found the activity for SP-MRGX1 mutant and BAM-MRGX2 mutant pairs, while intact receptors do not interact with these peptides. Furthermore, we elucidated the evolutionary process of activity of orthologs of other species based on mutations in these interacting factors. These results indicate that the IDL Score is useful not only for residues involved in activity, but also for estimating the acquisition of function during receptor evolution.

Free Research Field

バイオインフォマティクス

Academic Significance and Societal Importance of the Research Achievements

これまで、受容体がリガンドを認識に関与する残基の探索は、三次元結晶構造に基づいた相互作用残基の探索か、網羅的な変異体の構築に基づいていた。しかし、前者は直接相互作用がない残基を見出すことは難しく、後者は膨大な変異体実験を要し、時間・コストがかかっていた。本成果により確立した手法は、体系的かつ効率的に受容体とペプチドの相互作用に関わる分子メカニズムを明らかにできる。ペプチドは生体内で様々な生物学的役割を果たしていることから、本研究は内分泌学の発展に寄与する。さらに、ペプチド認識に関わる残基の解明はペプチド認識が関与する点変異が関連する進化や遺伝病のメカニズム解明にもつながる。