Improving Real-Time In Vivo Molecular Monitoring: Multi-Gaussian with Laplacian Voltammogram Fitting Enhances the Precision of Electrochemical Aptamer-Based Sensors

Abstract

Electrochemical, aptamer-based (EAB) sensors are the first technology supporting high-frequency, real-time measurements of the concentrations of specific drugs, metabolites, and biomarkers in the body that is independent of the chemical reactivity of its analytes. To achieve this, EAB sensors employ the binding-induced folding of an electrode-attached, redox-reporter-modified aptamer to produce an electrochemical output easily monitored using square wave voltammetry. Using such sensors, multiple research groups have achieved the seconds-resolved, multihour measurement of multiple drugs and metabolites in situ in the veins, brains, and peripheral solid tissues of live animals. Historically, the large volume of voltammograms (hundreds per hour) produced by in vivo EAB sensors have been fitted using simple polynomials to extract the peak heights from which target concentrations are estimated. This, however, can lead to misestimation of peak heights due to overfitting of noise or poor correction of peak shouldering. In response, here we describe an alternative method of fitting EAB sensor voltammograms that improves the accuracy of “problematic” (i.e., noisy, or heavily “shouldered”) data sets while simultaneously reducing sensor noise.