Combination of Single-Molecule Electrical Measurements and Machine Learning for the Identification of Single Biomolecules

Optical Image Sensors for Smart Analytical Chemiluminescence Biosensors

Optical biosensors have emerged as a powerful tool in analytical biochemistry, offering high sensitivity and specificity in the detection of various biomolecules. This article explores the advancements in the integration of optical biosensors with microfluidic technologies, creating lab-on-a-chip (LOC) platforms that enable rapid, efficient, and miniaturized analysis at the point of need. These LOC platforms leverage optical phenomena such as chemiluminescence and electrochemiluminescence to achieve real-time detection and quantification of analytes, making them ideal for applications in medical diagnostics, environmental monitoring, and food safety. Various optical detectors used for detecting chemiluminescence are reviewed, including single-point detectors such as photomultiplier tubes (PMT) and avalanche photodiodes (APD), and pixelated detectors such as charge-coupled devices (CCD) and complementary metal-oxide-semiconductor (CMOS) sensors. A significant advancement discussed in this review is the integration of optical biosensors with pixelated image sensors, particularly CMOS image sensors. These sensors provide numerous advantages over traditional single-point detectors, including high-resolution imaging, spatially resolved measurements, and the ability to simultaneously detect multiple analytes. Their compact size, low power consumption, and cost-effectiveness further enhance their suitability for portable and point-of-care diagnostic devices. In the future, the integration of machine learning algorithms with these technologies promises to enhance data analysis and interpretation, driving the development of more sophisticated, efficient, and accessible diagnostic tools for diverse applications.

Machine learning empowered next generation DNA sequencing: perspective and prospectus

The pursuit of ultra-rapid, cost-effective, and accurate DNA sequencing is a highly sought after aspect of personalized medicine development. With recent advancements, mainstream machine learning (ML) algorithms hold immense promise for high throughput DNA sequencing at the single nucleotide level. While ML has revolutionized multiple domains of nanoscience and nanotechnology, its implementation in DNA sequencing is still in its preliminary stages. ML-aided DNA sequencing is especially appealing, as ML has the potential to decipher complex patterns and extract knowledge from complex datasets. Herein, we present a holistic framework of ML-aided next-generation DNA sequencing with domain knowledge to set directions toward the development of artificially intelligent DNA sequencers. This perspective focuses on the current state-of-the-art ML-aided DNA sequencing, exploring the opportunities as well as the future challenges in this field. In addition, we provide our personal viewpoints on the critical issues that require attention in the context of ML-aided DNA sequencing.

Reduction of Biosensor False Responses and Time Delay Using Dynamic Response and Theory-Guided Machine Learning

Here, we provide a new methodology for reducing false results and time delay of biosensors, which are barriers to industrial, healthcare, military, and consumer applications. We show that integrating machine learning with domain knowledge in biosensing can complement and improve the biosensor accuracy and speed relative to the performance achieved by traditional regression analysis of a standard curve based on the biosensor steady-state response. The methodology was validated by rapid and accurate quantification of microRNA across the nanomolar to femtomolar range using the dynamic response of cantilever biosensors. Theory-guided feature engineering improved the performance and efficiency of several classification models relative to the performance achieved using traditional feature engineering methods (TSFRESH). In addition to the entire dynamic response, the technique enabled rapid and accurate quantification of the target analyte concentration and false-positive and false-negative results using the initial transient response, thereby reducing the required data acquisition time (i.e., time delay). We show that model explainability can be achieved by combining theory-guided feature engineering and feature importance analysis. The performance of multiple classifiers using both TSFRESH- and theory-based features from the biosensor's initial transient response was similar to that achieved using the entire dynamic response with data augmentation. We also show that the methodology can guide design of experiments for high-performance biosensing applications, specifically, the selection of data acquisition parameters (e.g., time) based on potential application-dependent performance thresholds. This work provides an example of the opportunities for improving biosensor performance, such as reducing biosensor false results and time delay, using explainable machine learning models supervised by domain knowledge in biosensing.

Deciphering DNA nucleotide sequences and their rotation dynamics with interpretable machine learning integrated C3N nanopores

A solid-state nanopore combined with the quantum transport method has garnered substantial attention and intrigue for DNA sequencing due to its potential for providing rapid and accurate sequencing results, which could have numerous applications in disease diagnosis and personalized medicine. However, the intricate and multifaceted nature of the experimental protocol poses a formidable challenge in attaining precise single nucleotide analysis. Here, we report a machine learning (ML) framework combined with the quantum transport method to accelerate high-throughput single nucleotide recognition with C3N nanopores. The optimized eXtreme Gradient Boosting Regression (XGBR) algorithm has predicted the fingerprint transmission of each unknown nucleotide and their rotation dynamics with root mean square error scores as low as 0.07. Interpretability of ML black box models with the game theory-based SHapley Additive exPlanation method has provided a quasi-explanation for the model working principle and the complex relationship between electrode-nucleotide coupling and transmission. Moreover, a comprehensive ML classification of nucleotides based on binary, ternary, and quaternary combinations shows maximum accuracy and F1 scores of 100%. The results suggest that ML in tandem with a nanopore device can potentially alleviate the experimental hurdles associated with quantum tunneling and facilitate fast and high-precision DNA sequencing.

High Accuracy Protein Identification: Fusion of Solid-State Nanopore Sensing and Machine Learning

Proteins are arguably one of the most important class of biomarkers for health diagnostic purposes. Label-free solid-state nanopore sensing is a versatile technique for sensing and analyzing biomolecules such as proteins at single-molecule level. While molecular-level information on size, shape, and charge of proteins can be assessed by nanopores, the identification of proteins with comparable sizes remains a challenge. Here, solid-state nanopore sensing is combined with machine learning to address this challenge. The translocations of four similarly sized proteins is assessed using amplifiers with bandwidths (BWs) of 100 kHz and 10 MHz, the highest bandwidth reported for protein sensing, using nanopores fabricated in <10 nm thick silicon nitride membranes. F-values of up to 65.9% and 83.2% (without clustering of the protein signals) are achieved with 100 kHz and 10 MHz BW measurements, respectively, for identification of the four proteins. The accuracy of protein identification is further enhanced by classifying the signals into different clusters based on signal attributes, with F-value and specificity of up to 88.7% and 96.4%, respectively, for combinations of four proteins. The combined use of high bandwidth instruments, advanced clustering and machine learning methods allows label-free identification of proteins with high accuracy.

Machine Learning Prediction of the Transmission Function for Protein Sequencing with Graphene Nanoslit

Protein sequencing has rapidly changed the landscape of healthcare and life science by accelerating the growth of diagnostics and personalized medicines for a variety of fatal diseases. Next-generation nanopore/nanoslit sequencing is promising to achieve single-molecule resolution with chromosome-size-long readability. However, due to inherent complexity, high-throughput sequencing of all 20 amino acids demands different approaches. Aiming to accelerate the detection of amino acids, a general machine learning (ML) method has been developed for quick and accurate prediction of the transmission function for amino acid sequencing. Among the utilized ML models, the XGBoost regression model is found to be the most effective algorithm for fast prediction of the transmission function with a very low test root-mean-square error (RMSE ∼0.05). In addition, using the random forest ML classification technique, we are able to classify the neutral amino acids with a prediction accuracy of 100%. Therefore, our approach is an initiative for the prediction of the transmission function through ML and can provide a platform for the quick identification of amino acids with high accuracy.

Combining dynamic Monte Carlo with machine learning to study nanoparticle translocation

Resistive pulse sensing (RPS) measurements of nanoparticle translocation have the ability to provide information on single-particle level characteristics, such as diameter or mobility, as well as ensemble averages. However, interpreting these measurements is complex and requires an understanding of nanoparticle dynamics in confined spaces as well as the ways in which nanoparticles disrupt ion transport while inside a nanopore. Here, we combine Dynamic Monte Carlo (DMC) simulations with Machine Learning (ML) and Poisson-Nernst-Planck calculations to simultaneously simulate nanoparticle dynamics and ion transport during hundreds of independent particle translocations as a function of nanoparticle size, electrophoretic mobility, and nanopore length. The use of DMC simulations allowed us to explicitly investigate the effects of Brownian motion and nanoparticle/nanopore characteristics on the amplitude and duration of translocation signals. Simulation results were verified with experimental RPS measurements and found to be in quantitative agreement.

Single-Molecule Classification of Aspartic Acid and Leucine by Molecular Recognition through Hydrogen Bonding and Time-Series Analysis

Amino acid detection/identification methods are important for understanding biological systems. In this study, we developed the single-molecule measurement for investigated quantum tunneling enhancement by chemical modification and machine learning based time series analysis for develop accurate amino acid discrimination. We performed single-molecule measurement of L-aspartic Acid (Asp) and L-leucine (Leu) with mercaptoacetic acid (MAA) chemical modified nano-gap. The measured current was investigated by machine learning based time series analysis method for accurate amino acid discrimination. Compared to measurements using bare nano-gap, it is found that MAA modification improves the difference in the conductance-time profiles between Asp and Leu through the hydrogen bonding facilitated tunneling phenomena. It is also found that this method enables determination of relative concentration. even in the mixture of Asp and Leu. It improves selective analysis for amino acids, and therefore would be applicable in medicine, diagnosis, and single-molecule peptide sequencing.

Nanodevices for Biological and Medical Applications: Development of Single-Molecule Electrical Measurement Method

A comprehensive detection of a wide variety of diagnostic markers is required for the realization of personalized medicine. As a sensor to realize such personalized medicine, a single molecule electrical measurement method using nanodevices is currently attracting interest for its comprehensive simultaneous detection of various target markers for use in biological and medical application. Single-molecule electrical measurement using nanodevices, such as nanopore, nanogap, or nanopipette devices, has the following features:; high sensitivity, low-cost, high-throughput detection, easy-portability, low-cost availability by mass production technologies, and the possibility of integration of various functions and multiple sensors. In this review, I focus on the medical applications of single- molecule electrical measurement using nanodevices. This review provides information on the current status and future prospects of nanodevice-based single-molecule electrical measurement technology, which is making a full-scale contribution to realizing personalized medicine in the future. Future prospects include some discussion on of the current issues on the expansion of the application requirements for single-mole-cule measurement.

A Guide to Signal Processing Algorithms for Nanopore Sensors

Nanopore technology holds great promise for a wide range of applications such as biomedical sensing, chemical detection, desalination, and energy conversion. For sensing performed in electrolytes in particular, abundant information about the translocating analytes is hidden in the fluctuating monitoring ionic current contributed from interactions between the analytes and the nanopore. Such ionic currents are inevitably affected by noise; hence, signal processing is an inseparable component of sensing in order to identify the hidden features in the signals and to analyze them. This Guide starts from untangling the signal processing flow and categorizing the various algorithms developed to extracting the useful information. By sorting the algorithms under Machine Learning (ML)-based versus non-ML-based, their underlying architectures and properties are systematically evaluated. For each category, the development tactics and features of the algorithms with implementation examples are discussed by referring to their common signal processing flow graphically summarized in a chart and by highlighting their key issues tabulated for clear comparison. How to get started with building up an ML-based algorithm is subsequently presented. The specific properties of the ML-based algorithms are then discussed in terms of learning strategy, performance evaluation, experimental repeatability and reliability, data preparation, and data utilization strategy. This Guide is concluded by outlining strategies and considerations for prospect algorithms.

Anisotropic acoustic phonon polariton-enhanced infrared spectroscopy for single molecule detection

Nanoscale Fourier transform infrared spectroscopy (nano-FTIR) based on scanning probe microscopy enables the identification of the chemical composition and structure of surface species with a high spatial resolution (∼10 nm), which is crucial for exploring catalytic reaction processes, cellular processes, virus detection, etc. However, the characterization of a single molecule with nano-FTIR is still challenging due to the weak coupling between the molecule and infrared light due to a large size mismatch. Here, we propose a novel structure (monolayer α-MoO₃/air nanogap/Au) to excite anisotropic acoustic phonon polaritons (APhPs) with ultra-high field confinement (mode volume, V_APhPs∼ 10^-11V₀) and electromagnetic energy enhancement (>10⁷), which largely enhance the interaction of single molecules with infrared light. In addition, the anisotropic APhP-assisted nano-FTIR can detect single molecular dipoles in directions both along and perpendicular to the probe axis, while pristine nano-FTIR mainly detects molecular dipoles along the probe axis. The proposed structure provides a way to detect a single molecule, which will impact the fields of biology, chemistry, energy, and environment through fundamental research and applications.

Adeno-associated virus characterization for cargo discrimination through nanopore responsiveness

Solid-state nanopore (SSN)-based analytical methods have found abundant use in genomics and proteomics with fledgling contributions to virology - a clinically critical field with emphasis on both infectious and designer-drug carriers. Here we demonstrate the ability of SSN to successfully discriminate adeno-associated viruses (AAVs) based on their genetic cargo [double-stranded DNA (AAVdsDNA), single-stranded DNA (AAVssDNA) or none (AAVempty)], devoid of digestion steps, through nanopore-induced electro-deformation (characterized by relative current change; ΔI/I0). The deformation order was found to be AAVempty > AAVssDNA > AAVdsDNA. A deep learning algorithm was developed by integrating support vector machine with an existing neural network, which successfully classified AAVs from SSN resistive-pulses (characteristic of genetic cargo) with >95% accuracy - a potential tool for clinical and biomedical applications. Subsequently, the presence of AAVempty in spiked AAVdsDNA was flagged using the ΔI/I0 distribution characteristics of the two types for mixtures composed of ∼75 : 25% and ∼40 : 60% (in concentration) AAVempty : AAVdsDNA.

Advancing Biosensors with Machine Learning

Chemometrics play a critical role in biosensors-based detection, analysis, and diagnosis. Nowadays, as a branch of artificial intelligence (AI), machine learning (ML) have achieved impressive advances. However, novel advanced ML methods, especially deep learning, which is famous for image analysis, facial recognition, and speech recognition, has remained relatively elusive to the biosensor community. Herein, how ML can be beneficial to biosensors is systematically discussed. The advantages and drawbacks of most popular ML algorithms are summarized on the basis of sensing data analysis. Specially, deep learning methods such as convolutional neural network (CNN) and recurrent neural network (RNN) are emphasized. Diverse ML-assisted electrochemical biosensors, wearable electronics, SERS and other spectra-based biosensors, fluorescence biosensors and colorimetric biosensors are comprehensively discussed. Furthermore, biosensor networks and multibiosensor data fusion are introduced. This review will nicely bridge ML with biosensors, and greatly expand chemometrics for detection, analysis, and diagnosis.