[10a-N302-1~11] 18.1 Materials Informatics

9:45 AM - 10:00 AM JST(12:45 AM - 1:00 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[10a-N302-4] Exploration of Novel Switch-Type Fluorescent Molecules Using Molecular Generative AI

〇Yumi Tazuya¹, Yoshiharu Mori², Hisashi Ono³, Mizuki Sugimoto³, Orie Takayama³, Kenjiro Hanaoka³, Kiyoshi Yagi⁴, Koichiro Kato^1,5 (1.Grad. Sch. Eng., Kyushu Univ., 2.Grad. Sch. ISEE., Kyushu Univ., 3.Grad. Sch. Pharm., Keio Univ., 4.Grad. Sch. Pure Appl. Sci., Univ. of Tsukuba, 5.CMS, Kyushu Univ.)

Keywords:

Molecule Search,Machine Learning,Fluorescent molecules

Rhodamine dyes possess high fluorescence quantum yield (QY) and excellent photostability, and are used in applications such as fluorescence imaging. However, since elucidating quenching mechanisms has historically been crucial for the development of switchable fluorescent molecules, this study aimed to create novel switchable fluorescent molecules by exploring low-QY rhodamine dyes, based on this reverse approach. We constructed a machine learning model to predict QY based on experimental data and incorporated it into the evaluation function of a molecular generation AI to identify novel fluorescent molecules predicted to have low QY while retaining a rhodamine skeleton.