“Polymer Sequencer” Achieved Through Mass Spectrometry and AI

• Analysis and Evaluation Techniques
• Artificial Intelligence
• Biomedical Applications
• Data-Driven Approaches
• Machine Learning
• Polymeric Materials
• Research Center for Macromolecules and Biomaterials

Polymers are chain-like molecules consisting of hundreds to tens of thousands of monomers in their basic structure. Yusuke Hibi and his colleagues developed the world’s first “polymer sequencer,” a tool that uses mass spectrometry and AI to quantitatively determine the frequency of monomer sequences, which impart diverse functions to polymers.

Randomly Arranged Monomers

Many of the polymers abundant in our daily lives, such as plastics, are copolymers—compounds composed of two or more different kinds of monomers. Even when the same monomers are combined, differences in their sequence can drastically alter the properties and functions of the copolymer. A key factor affecting copolymer properties is the occurrence frequency of specific sequences of a few monomers (subsequences). For instance, in a copolymer consisting of monomers A and B, sequences like AAA and ABAB are examples of subsequences.

Controlling the sequence of monomers during the polymerization process is challenging; typically, random sequences are generated probabilistically. Moreover, there has been no universal analytical technique to quantitatively determine the occurrence frequency of subsequences in randomly sequenced copolymers, making it difficult to analyze the correlation between sequence and material properties or to design materials based on sequence. In response, Hibi and his colleagues proposed a strategy to analyze copolymer mass spectrometry data using AI.