Abstract

The function of ganglioside oligosaccharides critically depends on the number, linkage, and spatial arrangement of sialic acid residues, yet direct discrimination of highly sialylated ganglioside oligosaccharides, particularly at the isomeric level, remains challenging due to their high charge density and subtle structural differences. Here, we present a nanopore-based strategy for single-molecule identification of ganglioside oligosaccharides. By engineering the sensing region of an α-hemolysin nanopore with synergistic cationic and aromatic mutations, we markedly enhance the capture of multisialylated glycans and prolong their residence within the pore. This interaction-mediated slowdown converts otherwise transient blockade events into extended trajectories containing rich molecular dynamics information. Beyond conventional blockade amplitude and dwell time descriptors, we introduce time- and spectral-domain features to characterize intraevent current fluctuations, thereby overcoming the key limitations of traditional nanopore analyses. Using this approach, we reliably discriminate 4 linkage isomers of trisialylated ganglioside glycans and accurately identify ganglioside oligosaccharides containing 3 to 5 sialic acid residues in heterogeneous mixtures. When combined with a machine-learning framework, the extracted time- and spectral-domain features enable single-level isomer identification and mixture deconvolution and remain effective even in complex biological matrices such as neural cell line and brain tissue lysates. Together, these results demonstrate that controlling analyte–pore interactions to unlock higher-order dynamical signatures enables nanopore analysis of highly sialylated ganglioside oligosaccharides, providing a general strategy for single-molecule glycan sequence elucidation in biologically relevant environments.