Using machine learning and liquid crystal droplets to identify and quantify endotoxins from different bacterial species

Possibilities and limitations of convolutional neural network machine learning architectures in the characterisation of achiral orthogonal smectic liquid crystals

Machine learning is becoming a valuable tool in the characterisation and property prediction of liquid crystals. It is thus worthwhile to be aware of the possibilities but also the limitations of current machine learning algorithms. In this study we investigated a phase sequence of isotropic - fluid smecticA - hexatic smectic B - soft crystal CrE - crystalline. This is a sequence of transitions between orthogonal phases, which are expected to be difficult to distinguish, because of only minute changes in order. As expected, strong first order transitions such as the liquid to liquid crystal transition and the crystallisation can be distinguished with high accuracy. It is shown that also the hexatic SmB to soft crystal CrE transition is clearly characterised, which represents the transition from short- to long-range order. Limitations of convolutional neural networks can be observed for the fluid to hexatic SmA to SmB transition, where both phases exhibit short-range ordering.

Nanoparticle-Assisted Liquid Crystal Droplet Sensors Enable Analysis of Low-Concentration Species in Aqueous Medium

We introduce nanoparticle-assisted liquid crystal (LC) droplet-based sensors that allow determination of low-level concentrations of aqueous soluble species. The silica nanoparticles functionalized with mixed monolayers composed of two distinct groups, hydrophobic alkane tail- and charged group-terminated silanes, facilitated ternary physical interactions between the model analytes (methylene blue (MB) or methyl orange (MO)) and the nematic mesogens 5CB (4-cyano-4'-pentylbiphenyl), and the interfacial species of the nanoparticle. The response of the LC droplets was measured upon nanoparticle adsorption as a function of analyte concentration, which was characterized by the optical determination of the configuration distributions of the LC droplets. We highlight the importance of the charging and the composition of the nanoparticle interfaces for analytical purposes that allow accurate determination of the concentration of the analytes on the order of 0.01 ppb. Such a low concentration corresponds to a low interfacial coverage of nanoparticles, indicating the promisingly high sensitivity of the sensor platform to target analytes. Distinct from the past examples of the LC-based sensors, the nanoparticle-assisted LC sensors allow detection of the species that do not directly cause an ordering transition at the LC-water interfaces, which allow a broader range of analytical targets. The sensor platform that we report herein can be easily tunable for a range of target molecules and will find use in the determination of a wide range of micropollutants in aqueous environments.

Single-molecule lipopolysaccharides identification and the interplay with biomolecules via nanopore readout

Lipopolysaccharides (LPS) are the major constituent on the cell envelope of all gram-negative bacteria. They are ubiquitous in air, and are toxic inflammatory stimulators for urinary disorders and sepsis. The reported optical, thermal, and electrochemical sensors via the intermolecular interplay of LPS with proteins and aptamers are generally complicated methods. We demonstrate the single-molecule nanopore approach for LPS identification in distinct bacteria as well as the serotypes discrimination. With a 4 nm nanopore, we achieve a detection limit of 10 ng/mL. Both the antibiotic polymyxin B (PMB) and DNA aptamer display specific binding to LPS. The identification of LPS in both human serum and tap water show good performance with nanopore platforms. Our work shows a highly-sensitive and easy-to-handle scheme for clinical and environmental biomarkers determination and provides a promising screening tool for early warning of contamination in water and medical supplies.

Decoding Optical Responses of Contact-Printed Arrays of Thermotropic Liquid Crystals Using Machine Learning: Detection and Reporting of Aqueous Amphiphiles with Enhanced Sensitivity and Selectivity

Surfactants and other amphiphilic molecules are used extensively in household products, industrial processes, and biological applications and are also common environmental contaminants; as such, methods that can detect, sense, or quantify them are of great practical relevance. Aqueous emulsions of thermotropic liquid crystals (LCs) can exhibit distinctive optical responses in the presence of surfactants and have thus emerged as sensitive, rapid, and inexpensive sensors or reporters of environmental amphiphiles. However, many existing LC-in-water emulsions require the use of complicated or expensive instrumentation for quantitative characterization owing to variations in optical responses among individual LC droplets. In many cases, the responses of LC droplets are also analyzed by human inspection, which can miss subtle color or topological changes encoded in LC birefringence patterns. Here, we report an LC-based surfactant sensing platform that takes a step toward addressing several of these issues and can reliably predict concentrations and types of surfactants in aqueous solutions. Our approach uses surface-immobilized, microcontact-printed arrays of micrometer-scale droplets of thermotropic LCs and hierarchical convolutional neural networks (CNNs) to automatically extract and decode rich information about topological defects and color patterns available in optical micrographs of LC droplets to classify and quantify adsorbed surfactants. In addition, we report computational capabilities to determine relevant optical features extracted by the CNN from LC micrographs, which can provide insights into surfactant adsorption phenomena at LC-water interfaces. Overall, the combination of microcontact-printed LC arrays and machine learning provides a convenient and robust platform that could prove useful for developing high-throughput sensors for on-site testing of environmentally or biologically relevant amphiphiles.

Machine learning classification of polar sub-phases in liquid crystal MHPOBC

Experimental polarising microscopy texture images of the fluid smectic phases and sub-phases of the classic liquid crystal MHPOBC were classified as paraelectric (SmA*), ferroelectric (SmC*), ferrielectric (SmC1/3*), and antiferroelectric (SmCA*) using convolutional neural networks, CNNs. Two neural network architectures were tested, a sequential convolutional neural network with varying numbers of layers and a simplified inception model with varying number of inception blocks. Both models are successful in binary classifications between different phases as well as classification between all four phases. Optimised architectures for the multi-phase classification achieved accuracies of (84 ± 2)% and (93 ± 1)% for sequential convolutional and inception networks, respectively. The results of this study contribute to the understanding of how CNNs may be used in classifying liquid crystal phases. Especially the inception model is of sufficient accuracy to allow automated characterization of liquid crystal phase sequences and thus opens a path towards an additional method to determine the phases of novel liquid crystals for applications in electro-optics, photonics or sensors. The outlined procedure of supervised machine learning can be applied to practically all liquid crystal phases and materials, provided the infrastructure of training data and computational power is provided.

Using pretrained machine learning models to predict luminous and solar transmittance controllability of liquid crystal/polymer composites from microstructural images

Polarized optical microscopy (POM) images of polymer network liquid crystals (PNLCs) were first analyzed using a pretrained machine learning model for feature extraction and hierarchical clustering. The analyses worked well in predicting and improving the thermoresponsive changes individually in direct luminous and hemispheric solar transmittance, both of which are crucial properties of energy-saving smart windows. The features of a 1280 × 1920-pixel color POM image were extracted by the latest pretrained algorithm, EfficientNet-B7, as a 2560-dimensional vector and then reduced into a two-dimensional space for clustering and visualization using the uniform manifold approximation and projection (UMAP) algorithm while efficiently preserving the global structures of the distance relationship in a high-dimensional space. The feature vectors in the UMAP space were correlated with the thermoresponsive transmittance and classified using hierarchical clustering analysis. The extracted features belonging to some clusters were also correlated with the fabrication parameters. The PNLCs here were produced from various raw materials under different fabrication conditions. These analyses and predictability are extensively applied to different PNLCs for stimuli-responsive optical devices, such as solar- and privacy-control windows.

Neural networks determination of material elastic constants and structures in nematic complex fluids

Supervised machine learning and artificial neural network approaches can allow for the determination of selected material parameters or structures from a measurable signal without knowing the exact mathematical relationship between them. Here, we demonstrate that material nematic elastic constants and the initial structural material configuration can be found using sequential neural networks applied to the transmmited time-dependent light intensity through the nematic liquid crystal (NLC) sample under crossed polarizers. Specifically, we simulate multiple times the relaxation of the NLC from a random (qeunched) initial state to the equilibirum for random values of elastic constants and, simultaneously, the transmittance of the sample for monochromatic polarized light. The obtained time-dependent light transmittances and the corresponding elastic constants form a training data set on which the neural network is trained, which allows for the determination of the elastic constants, as well as the initial state of the director. Finally, we demonstrate that the neural network trained on numerically generated examples can also be used to determine elastic constants from experimentally measured data, finding good agreement between experiments and neural network predictions.

Optical Fingerprinting of Dynamic Interfacial Reaction Pathways Using Liquid Crystals

Reactions at interfaces between fluid phases are widely used to synthesize small molecules, polymers, and nanoparticles. In situ monitoring of the underlying dynamic reaction pathways remains challenging. Liquid crystals (LCs) have been used to detect simple chemical transformations at interfaces in situations where interface-bound reactants and products trigger distinct equilibrium orientations of LCs. However, whether or not LCs can be used to report complex reaction pathways via nonequilibrium states generated by reactions has not been explored. Here we explore this question using S_N2' nucleophilic substitution reactions that involve a synthetic amphiphile and a series of amine-based nucleophiles with one to four reaction sites. Although all reactants and products generate the same equilibrium LC orientation, we find that each nucleophile defines a distinct set of possible reaction pathways with a characteristic spatial and temporal LC optical response unique to the nucleophile. Additional experiments reveal that the nonequilibrium orientational states of the LCs arise from a combination of dynamic interfacial processes that include adsorption/desorption of reactants, the presence of reaction intermediates on the LC interface, and the generation of interfacial tension gradients (Marangoni stresses). Overall, our results reveal that the spatiotemporal optical outputs of LCs ("optical fingerprints") can be a rich source of information regarding interfacial reactions.

Liquid Crystal Droplet-Based Biosensors: Promising for Point-of-Care Testing

The development of biosensing platforms has been impressively accelerated by advancements in liquid crystal (LC) technology. High response rate, easy operation, and good stability of the LC droplet-based biosensors are all benefits of the long-range order of LC molecules. Bioprobes emerged when LC droplets were combined with biotechnology, and these bioprobes are used extensively for disease diagnosis, food safety, and environmental monitoring. The LC droplet biosensors have high sensitivity and excellent selectivity, making them an attractive tool for the label-free, economical, and real-time detection of different targets. Portable devices work well as the accessory kits for LC droplet-based biosensors to make them easier to use by anyone for on-site monitoring of targets. Herein, we offer a review of the latest developments in the design of LC droplet-based biosensors for qualitative target monitoring and quantitative target analysis.

State-of-the-Art Development in Liquid Crystal Biochemical Sensors

As an emerging stimuli-responsive material, liquid crystal (LC) has attracted great attentions beyond display applications, especially in the area of biochemical sensors. Its high sensitivity and fast response to various biological or chemical analytes make it possible to fabricate a simple, real-time, label-free, and cost-effective LC-based detection platform. Advancements have been achieved in the development of LC-based sensors, both in fundamental research and practical applications. This paper briefly reviews the state-of-the-art research on LC sensors in the biochemical field, from basic properties of LC material to the detection mechanisms of LC sensors that are categorized into LC-solid, LC-aqueous, and LC droplet platforms. In addition, various analytes detected by LCs are presented as a proof of the application value, including metal ions, nucleic acids, proteins, glucose, and some toxic chemical substances. Furthermore, a machine-learning-assisted LC sensing platform is realized to provide a foundation for device intelligence and automatization. It is believed that a portable, convenient, and user-friendly LC-based biochemical sensing device will be achieved in the future.

Chiral nematic liquid crystal droplets as a basis for sensor systems

For a series of phospholipid coated calamitic nematic liquid crystal droplets (5CB, 6CB, 7CB, E7 and MLC7023) of diameter ∼18 μm, the addition of chiral dopant leaves the sign of surface anchoring unchanged. Herein we report that for these chiral nematic droplets an analyte induced transition from a Frank-Pryce structure (with planar anchoring) to a nested-cup structure (with perpendicular anchoring) is accompanied by changes in the intensity of reflected light. We propose this system as both a general scheme for understanding director fields in chiral nematic liquid crystal droplets with perpendicular anchoring and as an ideal candidate to be utilised as the basis for developing cheap, single use LC-based sensor devices.

Liquid crystal microcapillary-based sensors for affordable analytical applications

Stimuli-responsive properties of liquid crystals (LCs), when combined with their optical properties, offer sensitive and rapid sensing applications. Here, we propose and demonstrate a microcapillary-based method to be applied for the online detection of amphiphilic species, which can be further used for tracking biological and chemical species in aqueous media. Specifically, we used compartments (300-1400 μm) of nematic 4-cyano-4'-pentylbiphenyl (5CB) that were positioned into cylindrical glass microcapillaries that promote homeotropic anchoring. The flat surfaces of the cylindrical LC compartments were in contact with an aqueous media. We characterized the equilibrium and nonequilibrium response of LCs upon a change in their anchoring at the aqueous interfaces. Upon anchoring transition, we observed the formation of a positively charged defect at the proximity of the interface that moved to the center of the LC compartment and reached equilibrium, a four-petal configuration. This transition was observed to take an average of 41 ± 19 min., which we related to the motion of the defect due to the imbalance of the elastic forces. During the transition, we observed metastable states which could be removed via thermal treatment. We showed the capillary sensors to be useful considering their ease of additional quantification. We also show that the sensors are reversible that facilitate temporal and cumulative quantification. The findings reported in this study can further be used to develop sensors for specific purposes that require continuous tracking of the chemical and biological species that is critical for the health and safety of the individuals and society.

Liquid Crystals: Versatile Self-Organized Smart Soft Materials

Smart soft materials are envisioned to be the building blocks of the next generation of advanced devices and digitally augmented technologies. In this context, liquid crystals (LCs) owing to their responsive and adaptive attributes could serve as promising smart soft materials. LCs played a critical role in revolutionizing the information display industry in the 20th century. However, in the turn of the 21st century, numerous beyond-display applications of LCs have been demonstrated, which elegantly exploit their controllable stimuli-responsive and adaptive characteristics. For these applications, new LC materials have been rationally designed and developed. In this Review, we present the recent developments in light driven chiral LCs, i.e., cholesteric and blue phases, LC based smart windows that control the entrance of heat and light from outdoor to the interior of buildings and built environments depending on the weather conditions, LC elastomers for bioinspired, biological, and actuator applications, LC based biosensors for detection of proteins, nucleic acids, and viruses, LC based porous membranes for the separation of ions, molecules, and microbes, living LCs, and LCs under macro- and nanoscopic confinement. The Review concludes with a summary and perspectives on the challenges and opportunities for LCs as smart soft materials. This Review is anticipated to stimulate eclectic ideas toward the implementation of the nature's delicate phase of matter in future generations of smart and augmented devices and beyond.

Combination of liquid crystal and deep learning reveals distinct signatures of Parkinson's disease-related wild-type α-synuclein and six pathogenic mutants

α-Synuclein is a central player in Parkinson's disease (PD) pathology. Various point mutations in α-synuclein have been identified to alter the protein-phospholipid binding behavior and cause PD. Therefore, exploration of α-synuclein-phospholipid interaction is important for understanding the PD pathogenesis and helping the early diagnosis of PD. Herein, a phospholipid-decorated liquid crystal (LC)-aqueous interface is constructed to investigate the binding between α-synucleins (wild-type and six familial mutant A30P, E46K, H50Q, G51D, A53E and A53T) and phospholipid. The application of deep learning analyzes and reveals distinct LC signatures generated by the binding of α-synuclein and phospholipid. This system allows for the identification of single point mutant α-synucleins with an average accuracy of 98.3±1.3% in a fast and efficient manner. We propose that this analytical methodology provides a new platform to understand α-synuclein-lipid interactions, and can be potentially developed for easy identification of α-synuclein mutations in common clinic.

Applications of Microfluidics in Liquid Crystal-Based Biosensors

Liquid crystals (LCs) with stimuli-responsive configuration transition and optical anisotropic properties have attracted enormous interest in the development of simple and label-free biosensors. The combination of microfluidics and the LCs offers great advantages over traditional LC-based biosensors including small sample consumption, fast analysis and low cost. Moreover, microfluidic techniques provide a promising tool to fabricate uniform and reproducible LC-based sensing platforms. In this review, we emphasize the recent development of microfluidics in the fabrication and integration of LC-based biosensors, including LC planar sensing platforms and LC droplets. Fabrication and integration of LC-based planar platforms with microfluidics for biosensing applications are first introduced. The generation and entrapment of monodisperse LC droplets with different microfluidic structures, as well as their applications in the detection of chemical and biological species, are then summarized. Finally, the challenges and future perspectives of the development of LC-based microfluidic biosensors are proposed. This review will promote the understanding of microfluidic techniques in LC-based biosensors and facilitate the development of LC-based microfluidic biosensing devices with high performance.