Research

Abstract: Rapid diagnostics remain essential for outbreak response, yet most platforms are inherently limited by their reliance on prior knowledge of specific pathogens. This constraint is especially critical for flaviviruses, a rapidly evolving group with frequent zoonotic spillover and high potential for the emergence of unknown or poorly characterized strains. This work addresses that gap by developing multiplexed lateral flow assays (LFAs) designed not just to detect known viruses, but to identify and flag the unknown through pattern-based sensing. By leveraging antibody cross-reactivity, nanoparticle color encoding, and multivariate analysis, these assays generate distinct signal fingerprints that move beyond binary detection toward identifying unfamiliar viral signatures.

Using flaviviruses as a primary model system, this thesis establishes a sensor-array paradigm in which cross-reactive antibodies and multicolor nanoparticle labels produce unique colorimetric patterns across the Orthoflavivirus genus. Combined with unsupervised learning approaches such as principal component analysis and hierarchical clustering, this platform enables both classification of known viruses and recognition of anomalous, previously unseen patterns indicative of unknown threats. Together, this work presents a generalizable diagnostic framework capable of adapting to viral evolution, offering a path toward rapid, field-deployable detection of emerging and re-emerging flaviviruses without requiring target-specific reagents.