Topology-encoded multiplex diagnostics enabled by strand composition-controlled amplification

The timely differentiation of pathogens causing Fever of Unknown Origin (FUO) relies heavily on multiplex diagnostics. However, scaling up these existing platforms is severely restricted by the spectral overlap of fluorescent probes and the finite number of optical channels. Here, we present a topology-encoded diagnostic strategy that shifts the multiplexing dimension from spectral space to structural space, enabling target discrimination through discrete electrophoretic mobility signatures rather than fluorophore combinations. This platform integrates primer-blocked asymmetric amplification (PBA-amp) with programmable DNA nanostructure sensors (DNA-NS), in which target-specific single-stranded amplicons induce distinct topological transformations of the sensor scaffold. These transformations generate resolvable mobility states corresponding to individual targets, allowing multiplex identification using gel-resolved readouts. We demonstrated the duplex detection of Coxiella burnetii and Orientia tsutsugamushi, two prominent pathogens that cause clinically indistinguishable FUO, achieving an analytical sensitivity of down to 9.76 aM with high specificity against mismatched sequences. In clinical validation using 121 plasma-derived DNA samples, the platform exhibits an overall positive percent agreement of 96.4% -with all discordant cases strictly restricted to low titer samples near the stochastic limit (Ct > 36)- and a negative percent agreement of 100% compared to reference real-time PCR. By decoupling multiplex encoding from spectral constraints, this work establishes a structurally encoded framework for nucleic acid diagnostics, offering a scalable alternative to fluorescence-limited multiplexing while enabling simplified readout without the need for multichannel detection systems.