BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Yong-Jin Huang (Kyoto University) DTSTART:20260312T160000Z DTEND:20260312T163000Z DTSTAMP:20260421T124952Z UID:MoRN/138 DESCRIPTION:Title: A Structural Approach to Identifying Indicator Species in Chemical Reaction Networks\nby Yong-Jin Huang (Kyoto University) as part of Seminar on the Mathematics of Reaction Networks\n\n\nAbstract\nCellular phenotypes di splay high diversity\, reflecting the complex functional states of individ ual cells. While transcriptomics has traditionally been used to determine these states\, recent advances in single-cell technologies are shifting in terest toward more detailed classification via metabolomic phenotyping\, w hich directly reflects cellular function. However\, the large number of me tabolites poses challenges for both measurement and computational clusteri ng in the task of phenotypic classification. A fundamental question theref ore arises: which subset of species suffices to represent the system's ove rall state?\n\nThis talk introduces a novel theory that\, based solely on the structural information of chemical reaction networks\, identifies indi cator species—whose concentrations uniquely determine all others and thu s distinguish multistable equilibria. An implementing algorithm is applied to biochemical pathway databases. Numerical experiments demonstrate that classification based solely on these indicator species matches or surpasse s full-set accuracy\, with superior robustness under measurement noise. Th ese results provide a rigorous\, topology-based foundation for selecting i ndicator species\, advancing metabolic phenotyping and biomarker discovery .\n LOCATION:https://researchseminars.org/talk/MoRN/138/ END:VEVENT END:VCALENDAR