Machine learning for detecting DNA attachment on SPR biosensor

Abstract

Optoelectric biosensors measure the conformational changes of biomolecules and their molecular interactions, allowing researchers to use them in different biomedical diagnostics and analysis activities. Among different biosensors, surface plasmon resonance (SPR)-based biosensors utilize label-free and gold-based plasmonic principles with high precision and accuracy, allowing these gold-based biosensors as one of the preferred methods. The dataset generated from these biosensors are being used in different machine learning (ML) models for disease diagnosis and prognosis, but there is a scarcity of models to develop or assess the accuracy of SPR-based biosensors and ensure a reliable dataset for downstream model development. Current study proposed innovative ML-based DNA detection and classification models from the reflective light angles on different gold surfaces of biosensors and associated properties. We have conducted several statistical analyses and different visualization techniques to evaluate the SPR-based dataset and applied t-SNE feature extraction and min-max normalization to differentiate classifiers of low-variances. We experimented with several ML classifiers, namely support vector machine (SVM), decision tree (DT), multi-layer perceptron (MLP), k-nearest neighbors (KNN), logistic regression (LR) and random forest (RF) and evaluated our findings in terms of different evaluation metrics. Our analysis showed the best accuracy of 0.94 by RF, DT and KNN for DNA classification and 0.96 by RF and KNN for DNA detection tasks. Considering area under the receiver operating characteristic curve (AUC) (0.97), precision (0.96) and F1-score (0.97), we found RF performed best for both tasks. Our research shows the potentiality of ML models in the field of biosensor development, which can be expanded to develop novel disease diagnosis and prognosis tools in the future.

Advances in flame synthesis of nano-scale architectures for chemical, biomolecular, plasmonic, and light sensing

“…The research suggested that this biosensor could enhance diabetes diagnosis, particularly when used alongside standard glucose monitoring methods. 136 This study highlights the potential of this novel biosensor for practical applications, offering a rapid and inexpensive method for enhanced diabetes diagnosis.…”

“…The sensor's DNA aptamer conjugation selectively binds to GA, inducing a plasmonic wavelength shift, allowing for a detection limit of 0.1 μM, which falls well within the human GA range of 20-240 μM. 136 The biosensor demonstrated its capability to provide mean glucose levels over a three-week period, correlating with the glycation status of diabetes patients. Selectivity experiments conducted with various molecules confirmed the sensor's specificity, and tests for sensor reusability yielded positive results.…”

“…In another study, 136 a biosensor designed to detect glycated albumin (GA), an important biomarker for diabetes monitoring, was developed using a novel approach 114. This biosensor utilizes localized surface plasmon resonance (LSPR) for signal transduction, achieved through a fabrication process involving the deposition of gold nanoparticles on a quartz substrate using FSP.…”

A novel hybrid model to predict concomitant diseases for Hashimoto’s thyroiditis

“…For the proposes of DNA classification issues, frequently used techniques in the litreture are Support Vector Machines (SVM) [ 55 – 57 ], Random Forest [ 58 – 60 ], Logistic Regression [ 61 – 63 ], and K-Nearest Neighbors (KNN) [ 53 , 64 ]. The SVM algorithm creates an ideal hyperplane to divide various classes by maximizing the margin.…”

Metasurface‐Driven and Nanomaterial‐Coupled Plasmonic Biosensor for the Rapid and Quantitative Clinical Identification of Neutralizing Antibodies against SARS‐CoV‐2 Variants

“…Surface plasmon resonance (SPR)-based optical biosensors are used because of their real-time, label-free approach, allowing high-throughput results for multi-dimensional research analysis, ranging from fundamental biological and bio-medical research to environmental and agricultural monitoring programs. [20][21][22] Commercial SPR platforms, such as BIAcore 8K and MI-S200D, are based on traditional SPR technology. The working principle is that when a beam of P-polarized light is incident on the end face of a prism within a certain angle range, the interface between the prism and the metal film will generate surface plasma waves.…”