Engineered Nucleotide Chemicapacitive Microsensor Array Augmented with Physics‐Guided Machine Learning for High‐Throughput Screening of Cannabidiol. Issue 22 (6th May 2022)

• Record Type: Journal Article
• Authors: Yap, Stephanie Hui Kit; Pan, Jieming; Linh, Dao Viet; Zhang, Xiangyu; Wang, Xinghua; Teo, Wei Zhe; Zamburg, Evgeny; Tham, Chen‐Khong; Yew, Wen Shan; Poh, Chueh Loo; Thean, Aaron Voon‐Yew
• Is Part Of: Small. Volume 18:Issue 22(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-05-06
• Subjects: aptamers -- biosensing -- cannabidiol -- interdigitated electrodes -- physics‐guided machine learning; Nanotechnology -- Periodicals; Nanoparticles -- Periodicals; Microtechnology -- Periodicals; 620.5
• DOI: 10.1002/smll.202107659
• Languages: English
• ISSNs: 1613-6810

Abstract

Abstract: The recent legalization of cannabidiol (CBD) to treat neurological conditions such as epilepsy has sparked rising interest across global pharmaceuticals and synthetic biology industries to engineer microbes for sustainable synthetic production of medicinal CBD. Since the process involves screening large amounts of samples, the main challenge is often associated with the conventional screening platform that is time consuming, and laborious with high operating costs. Here, a portable, high‐throughput Aptamer‐based BioSenSing System (ABS 3 ) is introduced for label‐free, low‐cost, fully automated, and highly accurate CBD concentrations' classification in a complex biological environment. The ABS 3 comprises an array of interdigitated microelectrode sensors, each functionalized with different engineered aptamers. To further empower the functionality of the ABS 3, unique electrochemical features from each sensor are synergized using physics‐guided multidimensional analysis. The capabilities of this ABS 3 are demonstrated by achieving excellent CBD concentrations' classification with a high prediction accuracy of 99.98% and a fast testing time of 22 µs per testing sample using the optimized random forest (RF) model. It is foreseen that this approach will be the key to the realistic transformation from fundamental research to system miniaturization for diagnostics of disease biomarkers and drug development in the field of chemical/bioanalytics. Abstract : An Aptamer‐based BioSenSing System (ABS 3 ) comprising an array of aptamer‐functionalized interdigitated microelectrodes for accurate identification and concentration classification of cannabidiol is developed. Unique electrochemical features from each sensor are synergized using physics‐guided multidimensional analysis, and the ABS 3 achieves excellent cannabidiol concentration classification with a high prediction accuracy of 99.98% using the random forest model.