[10p-N302-1~10] AI for Measurement Science: Toward New Measurement Paradigms Enabled by the Deep Integration of Information and Measurement Sciences

3:45 PM - 4:15 PM JST(6:45 AM - 7:15 AM UTC)

[10p-N302-6] Automating Molecular Discovery with an Integrated Platform of AI and 3D-Phenomics

〇Hideyuki Shimizu¹ (1.Inst. Science Tokyo)

Keywords:

antimicrobial peptides,Tissue clearing method

We developed an AI model that predicts antimicrobial activity from amino acid sequences and applied it to discover and design novel antimicrobial peptides. Top candidates achieved a 90% hit rate, with half surpassing the activity of existing antibiotics. We are also advancing large-scale drug screening using patient-derived glioblastoma organoids, enabled by a 3D phenomics platform integrating tissue clearing and 3D point cloud analysis. Our vision is to establish an automated, integrated platform spanning molecular design to biological evaluation.