Fig. 1: Concept for the de novo designed α-helical-barrel differential sensor.
From: Differential sensing with arrays of de novo designed peptide assemblies
a Top: α-Helical barrels (αHBs) are loaded with an environment-sensitive dye giving a fluorescent signal. Bottom: The dye is displaced by an analyte causing a loss of fluorescence that can be measured. b Left: Different αHBs are combined with the environment sensitive dye in multi-well plates. Middle: The resulting array is challenged with different analytes, which can be pure compounds or complex mixtures. Depending on the relative binding strengths of the dye and the analytes for each αHB, dye is displaced differentially across the array to give a ‘fingerprint’ for each analyte. Right: Statistical and machine-learning methods are used to classify the different fingerprints and relate them to the analytes. The resulting models can be used as predictive classifiers for naïve samples. See Supplementary Note and Supplementary Figs. 10–12 for more detail on the data analysis and ML pipelines.
