A Temperature-Differential Affinity Biosensor: Model and ${\rm D}$-Optimal Performance Limits

Selectivity is one of the most challenging issues in biosensor design. Several methods have been proposed in the past to overcome nonspecific interference. In particular, it has been shown that temperature curves can be used to simultaneously measure two similar analytes. Here, the performance of such thermal-based systems is analyzed, using least squares estimation and existing models for affinity-based sensing, and it is shown that a D-optimal difference for the sensor temperatures exists. Analysis at this optimized condition yields bounds on the sensitivity and selectivity of this class of sensors. For the first time, thermal discrimination is employed for an affinity-based sensor: an artificial receptor-based system for the detection of the neurotransmitter acetylcholine and its metabolite choline. The system performance demonstrates the practicality of the theoretical results presented.