Machine-Learning Single-Stranded DNA Nanoparticles for Bacterial Analysis

Albumin host for supramolecular fluorescence recognition

Synthetic molecular sensors are crucial for real-time monitoring in biological systems and biotechnological applications, where detecting targets amidst potential interferents is essential. This task is particularly challenging in competitive environments that lacking chemically reactive functional groups, common in agricultural, biological, and environmental contexts. Consequently, scientific efforts have focused on developing sensitive and rapid analytical techniques, with fluorescent sensors emerging as prominent tools. Among these, the albumin-based supramolecular fluorescent indicator displacement assay (AS-FIDA) represents a significant advancement. Our research group has extensively contributed to this field, demonstrating the practical utility of various AS-FIDAs. We pioneered the use of albumin (ALB) as a host molecule in these synthetic chemical sensors, marking a notable advancement. AS-FIDA employs ALB as a versatile host molecule with multiple flexible and asymmetrical binding pockets capable of forming complexes with guest dyes, resulting in ALB@dye ensembles tailored for specific analyte recognition. Recent advancements in AS-FIDA have significantly expanded its applications. This review explores recent advances in ALB-based supramolecular sensors and sensor arrays for detecting biologically and environmentally significant molecules, such as pesticides, hormones, biomarkers, reactive species, mycotoxins, drugs, and carcinogens. The versatility of AS-FIDA positions it as a valuable tool in diverse settings, from laboratory research to practical applications in portable devices, smartphone-assisted on-site monitoring, imaging of living cells, and real sample analysis.

Machine learning supported single-stranded DNA sensor array for multiple foodborne pathogenic and spoilage bacteria identification in milk

Ensuring food safety through rapid and accurate detection of pathogenic bacteria in food products is a critical challenge in the food supply chain. In this study, a non-specific optical sensor array was proposed for the identification of multiple pathogenic bacteria in contaminated milk samples. Fluorescence-labeled single-stranded DNA was efficiently quenched by two-dimensional nanoparticles and subsequently recovered by foreign biomolecules. The recovered fluorescence generated a unique fingerprint for each bacterial species, enabling the sensor array to identify eight bacteria (pathogenic and spoilage) within a few hours. Four traditional machine learning models and two artificial neural networks were applied for classification. The neural network showed a 93.8 % accuracy with a 30-min incubation. Extending the incubation to 120 min increased the accuracy of the multiplayer perceptron to 98.4 %. This sensor array is a novel, low-cost, and high-accuracy approach for the identification of multiple bacteria, providing an alternative to plate counting and ELISA methods.

Surface-Engineered Nanomaterials for Optical Array Based Sensing

Array based sensing governed by optical methods provides fast and economic way for detection of wide variety of analytes where the ideality of detection processes depends on the sensor element's versatile mode of interaction with multiple analytes in an unbiased manner. This can be achieved by either the receptor unit having multiple recognition moiety, or their surface property should possess tuning ability upon fabrication called surface engineering. Nanomaterials have a high surface to volume ratio, making them viable candidates for molecule recognition through surface adsorption phenomena, which makes it ideal to meet the above requirements. Most crucially, by engineering a nanomaterial's surface, one may produce cross-reactive responses for a variety of analytes while focusing solely on a single nanomaterial. Depending on the nature of receptor elements, in the last decade the array-based sensing has been considering as multimodal detection platform which operates through various pathway including single channel, multichannel, binding and indicator displacement assay, sequential ON-OFF sensing, enzyme amplified and nanozyme based sensing etc. In this review we will deliver the working principle for Array-based sensing by using various nanomaterials like nanoparticles, nanosheets, nanodots and self-assembled nanomaterials and their surface functionality for suitable molecular recognition.

Development of Optical Differential Sensing Based on Nanomaterials for Biological Analysis

The discrimination and recognition of biological targets, such as proteins, cells, and bacteria, are of utmost importance in various fields of biological research and production. These include areas like biological medicine, clinical diagnosis, and microbiology analysis. In order to efficiently and cost-effectively identify a specific target from a wide range of possibilities, researchers have developed a technique called differential sensing. Unlike traditional "lock-and-key" sensors that rely on specific interactions between receptors and analytes, differential sensing makes use of cross-reactive receptors. These sensors offer less specificity but can cross-react with a wide range of analytes to produce a large amount of data. Many pattern recognition strategies have been developed and have shown promising results in identifying complex analytes. To create advanced sensor arrays for higher analysis efficiency and larger recognizing range, various nanomaterials have been utilized as sensing probes. These nanomaterials possess distinct molecular affinities, optical/electrical properties, and biological compatibility, and are conveniently functionalized. In this review, our focus is on recently reported optical sensor arrays that utilize nanomaterials to discriminate bioanalytes, including proteins, cells, and bacteria.

Recent Progress in Sensor Arrays: From Construction Principles of Sensing Elements to Applications

The traditional sensors are designed based on the "lock-and-key" strategy with high selectivity and specificity for detecting specific analytes, which however are not suitable for detecting multiple analytes simultaneously. With the help of pattern recognition technologies, the sensor arrays excel in distinguishing subtle changes caused by multitarget analytes with similar structures in a complex system. To construct a sensor array, the multiple sensing elements are undoubtedly indispensable units that will selectively interact with targets to generate the unique "fingerprints" based on the distinct responses, enabling the identification among various analytes through pattern recognition methods. This comprehensive review mainly focuses on the construction strategies and principles of sensing elements, as well as the applications of sensor array for identification and detection of target analytes in a wide range of fields. Furthermore, the present challenges and further perspectives of sensor arrays are discussed in detail.

Implementing neural network approach to create carbon-based optical nanosensor of heavy metal ions in liquid media

The present study is devoted to the creation of multifunctional optical carbon dots-based nanosensor to simultaneously measure concentrations of metal ions (Cu²⁺, Ni²⁺, Cr³⁺) and NO₃^- anion in liquid media. Such nanosensor operates on the basis of its fluorescence (FL) change under the influence of ions in the medium. However, the absence of analytical model, describing CD FL mechanism, the superposition of various luminescence quenching mechanisms during the interaction of carbon dots (CD) with cations, hampers the usage of classical approaches to solve this inverse multiparametric spectroscopic problem. To solve it neural networks were used that analyzed complex fluorescence signal from CD aqueous suspensions comprising Cu²⁺, Ni²⁺, Cr³⁺, NO₃^- ions in the concentration range from 0 to 4.95 mM. The following neural network architectures ensured optical spectroscopy inverse problem solution: multilayer perceptrons, 1D and 2D convolutional neural networks. The developed sensor enables simultaneous determination of the concentrations of heavy metal ions Cu²⁺, Ni²⁺, Cr³⁺ with a root mean squared error of 0.28 mM, 0.79 mM, 0.24 mM respectively. Based on the data given in the literature we can assert that the accuracy of the studied nanosensor satisfies the needs of monitoring the composition of waste and technological water. The developed nanosensor has a unique multimodality: with the simplicity of the synthesis protocol the sensor enables simultaneous determination of three heavy metal ions concentrations, while analogues are being developed mainly to measure the concentration of one (in rare cases two) heavy metal ions.

One-Component Multichannel Sensor Array for Rapid Identification of Bacteria

Bacterial infections routinely cause serious problems to public health. To mitigate the impact of bacterial infections, sensing systems are urgently required for the detection and subsequent epidemiological control of pathogenic organisms. Most conventional approaches are time-consuming and highly instrument- and professional operator-dependent. Here, we developed a novel one-component multichannel array constructed with complex systems made from three modified polyethyleneimine as well as negatively charged graphene oxide, which provided an information-rich multimode response to successfully identify 10 bacteria within minutes via electrostatic interactions and hydrophobic interactions. Furthermore, the concentration of bacteria (from OD₆₀₀ = 0.025 to 1) and the ratio of mixed bacteria were successfully achieved with our smart sensing system. Our designed sensor array also exhibited huge potential in biological samples, such as in urine (OD₆₀₀ = 0.125, 94% accuracy). The way to construct a sensor array with minimal sensor element with abundant signal outputs tremendously saves cost and time, providing a powerful tool for the diagnosis and assessment of bacterial infections in the clinic.

Machine Learning-Assisted Pattern Recognition of Amyloid Beta Aggregates with Fluorescent Conjugated Polymers and Graphite Oxide Electrostatic Complexes

Five fluorescent positively charged poly(para-aryleneethynylene) (P1-P5) were designed to construct electrostatic complexes C1-C5 with negatively charged graphene oxide (GO). The fluorescence of conjugated polymers was quenched by the quencher GO. Three electrostatic complexes were enough to distinguish between 12 proteins with 100% accuracy. Furthermore, using these sensor arrays, we could identify the levels of Aβ40 and Aβ42 aggregates (monomers, oligomers, and fibrils) via employing machine learning algorithms, making it an attractive strategy for early diagnosis of Alzheimer's disease.

Algorithmically Guided Optical Nanosensor Selector (AGONS): Guiding Data Acquisition, Processing, and Discrimination for Biological Sampling

Here, we report a biomarker-free detection of various biological targets through a programmed machine learning algorithm and an automated computational selection process termed algorithmically guided optical nanosensor selector (AGONS). The optical data processed/used by algorithms are obtained through a nanosensor array selected from a library of nanosensors through AGONS. The nanosensors are assembled using two-dimensional nanoparticles (2D-nps) and fluorescently labeled single-stranded DNAs (F-ssDNAs) with random sequences. Both 2D-np and F-ssDNA components are cost-efficient and easy to synthesize, allowing for scaled-up data collection essential for machine learning modeling. The nanosensor library was subjected to various target groups, including proteins, breast cancer cells, and lethal-7 (let-7) miRNA mimics. We have demonstrated that AGONS could select the most essential nanosensors while achieving 100% predictive accuracy in all cases. With this approach, we demonstrate that machine learning can guide the design of nanosensor arrays with greater predictive accuracy while minimizing manpower, material cost, computational resources, instrumentation usage, and time. The biomarker-free detection attribute makes this approach readily available for biological targets without any detectable biomarker. We believe that AGONS can guide optical nanosensor array setups, opening broader opportunities through a biomarker-free detection approach for most challenging biological targets.