In-silico-assisted derivatization of triarylboranes for the catalytic reductive functionalization of aniline-derived amino acids and peptides with H2.

Cheminformatics-based machine learning (ML) has been employed to determine optimal reaction conditions, including catalyst structures, in the field of synthetic chemistry. However, such ML-focused strategies have remained largely unexplored in the context of catalytic molecular transformations using Lewis-acidic main-group elements, probably due to the absence of a candidate library and effective guidelines (parameters) for the prediction of the activity of main-group elements. Here, the construction of a triarylborane library and its application to an ML-assisted approach for the catalytic reductive alkylation of aniline-derived amino acids and C-terminal-protected peptides with aldehydes and H2 is reported. A combined theoretical and experimental approach identified the optimal borane, i.e., B(2,3,5,6-Cl4-C6H)(2,6-F2-3,5-(CF3)2-C6H)2, which exhibits remarkable functional-group compatibility toward aniline derivatives in the presence of 4-methyltetrahydropyran. The present catalytic system generates H2O as the sole byproduct.

Metabolic syndrome is linked to the incidence of pancreatic cancer.

Background: Although previous studies have showed that metabolic syndrome is one of the contributors of pancreatic cancer, there is no clear consensus that early stages of metabolic syndrome are linked to increased incidence of pancreatic cancer. Therefore, we confirmed the linkage between metabolic syndrome and pancreatic cancer, and shown that even early stage of metabolic syndrome is linked to pancreatic cancer in the retrospective observational study. Methods: We recruited approximately 4.6 million Japanese in 2005 and followed up these subjects for more than 10 years. At the time of the enrollment, after obtaining clinical data with prescribed drugs and examining the presence or absence of metabolic syndrome (MetS), we followed up on these subjects with and without MetS to examine the incidence of pancreatic cancer. The modified criteria of the National Cholesterol Education Program Adult Treatment Panel III (NCEP/ATPIII) were used to define MetS. Findings: During the 40.7-month average follow-up period for 2,707,296 subjects with complete data for identifying MetS and important risk factors without pancreatic cancer before the enrollment, 87,857 suffered from pancreatic cancer. Pancreatic cancers occurred in 16,154 of 331,229 subjects (4.9%) in the MetS group and 71,703 of 2,376,067 patients (3.0%) in the non-MetS group (hazard ratio (HR), 1.37; 95% confidence interval [CI], 1.34-1.39; p < 0.0001 after the adjustment with age, smoking and sex). As the number of the constituent factors of MetS increased from one to five, the incidence of pancreatic cancer correspondingly increased (HR: 1.11, 1.23, 1.42, 1.66 and 2.03 using Cox proportional hazard models, p < 0.0001 each). When we defined MetS using the Japanese criteria, the results are in accord with the results using NCEP/ATPIII. Especially pre-metabolic syndrome (pre-MetS) in the Japanese criteria was tightly linked to the incidence of pancreatic cancers. Interpretation: MetS is confirmed to be linked to pancreatic cancer. Although we cannot conclude causality. We also demonstrated the link between pre-MetS and pancreatic cancer. Funding: The sponsors of the study were Japanese Heart Foundation and Japan Cardiovascular Research Foundation. This is also partially supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan; and Grants-in-Aid from the Japan Agency for Medical Research and Development.

Green-light wavelength-selective organic solar cells for agrivoltaics: dependence of wavelength on photosynthetic rate.

There is a growing demand for the development of novel solar power systems that can simultaneously solve the problems associated with both energy generation and food supply in agriculture. Green-light wavelength-selective organic solar cells (OSCs), whose transmitted blue and red light can be utilized to promote plant growth were recently reported by our group. However, the influence of wavelength variation on the photosynthetic rate in green-light wavelength-selective OSCs remains unclear. In this study, we report on the design and synthesis of new electron-accepting π-conjugated molecules containing cyclopentene-annelated thiophene with a spiro-substituted 2,7-bis(2-ethylhexyl)fluorene (FT) unit (TT-FT-ID) as a green-light wavelength-selective nonfullerene acceptor along with a reference compound TT-T-ID. Photophysical measurements indicate that the introduction of the FT unit leads to an absorption band with a small full width at half maximum in films, leading to the ability to fine-tune the absorption length. Concerning the optimization of the conditions for the fabrication of the active layers, which are composed of a green-light wavelength-selective donor polymer of poly(3-hexylthiophene) (P3HT) and the new acceptors, Bayesian optimization based on Gaussian process regression was applied to minimize the experimental batches. The green-light wavelength-selective factor (SG) and the PCEs in the green-light region (PCE-GR) of the P3HT:TT-FT-ID-based device were determined to be 0.52 and 8.6%, respectively, which are higher values than those of the P3HT:TT-T-ID blend film. The P3HT:TT-FT-ID blend film increased the photosynthetic rate of green pepper compared to that of the P3HT:TT-T-ID blend film. These results indicate that the fine-tuning of the absorbance required for crop growth is an important issue in developing green-light wavelength-selective OSCs for agrivoltaics.

High-precision rapid testing of omicron SARS-CoV-2 variants in clinical samples using AI-nanopore.

A digital platform that can rapidly and accurately diagnose pathogenic viral variants, including SARS-CoV-2, will minimize pandemics, public anxiety, and economic losses. We recently reported an artificial intelligence (AI)-nanopore platform that enables testing for Wuhan SARS-CoV-2 with high sensitivity and specificity within five minutes. However, which parts of the virus are recognized by the platform are unknown. Similarly, whether the platform can detect SARS-CoV-2 variants or the presence of the virus in clinical samples needs further study. Here, we demonstrated the platform can distinguish SARS-CoV-2 variants. Further, it identified mutated Wuhan SARS-CoV-2 expressing spike proteins of the delta and omicron variants, indicating it discriminates spike proteins. Finally, we used the platform to identify omicron variants with a sensitivity and specificity of 100% and 94%, respectively, in saliva specimens from COVID-19 patients. Thus, our results demonstrate the AI-nanopore platform is an effective diagnostic tool for SARS-CoV-2 variants.

Bayesian Optimization-Assisted Screening to Identify Improved Reaction Conditions for Spiro-Dithiolane Synthesis.

Bayesian optimization (BO)-assisted screening was applied to identify improved reaction conditions toward a hundred-gram scale-up synthesis of 2,3,7,8-tetrathiaspiro[4.4]nonane (1), a key synthetic intermediate of 2,2-bis(mercaptomethyl)propane-1,3-dithiol [tetramercaptan pentaerythritol]. Starting from the initial training set (ITS) consisting of six trials sampled by random screening for BO, suitable parameters were predicted (78% conversion yield of spiro-dithiolane 1) within seven experiments. Moreover, BO-assisted screening with the ITS selected by Latin hypercube sampling (LHS) further improved the yield of 1 to 89% within the eight trials. The established conditions were confirmed to be satisfactory for a hundred grams scale-up synthesis of 1.

Predicting heart failure onset in the general population using a novel data-mining artificial intelligence method.

We aimed to identify combinations of clinical factors that predict heart failure (HF) onset using a novel limitless-arity multiple-testing procedure (LAMP). We also determined if increases in numbers of predictive combinations of factors increases the probability of developing HF. We recruited people without HF who received health check-ups in 2010, who were followed annually for 4 years. Using 32,547 people, LAMP was performed to identify combinations of factors of fewer than four factors that could predict the onset of HF. The ability of the method to predict the probability of HF onset based on the number of matching predictive combinations of factors was determined in 275,658 people. We identified 549 combinations of factors for the onset of HF. Then we classified 275,658 people into six groups who had 0, 1-50, 51-100, 101-150, 151-200 or 201-250 predictive combinations of factors for the onset of HF. We found that the probability of HF progressively increased as the number of predictive combinations of factors increased. We identified combinations of variables that predict HF onset. An increased number of matching predictive combinations for the onset of HF increased the probability of HF onset.

Identification of Bacterial Drug-Resistant Cells by the Convolutional Neural Network in Transmission Electron Microscope Images.

The emergence of bacteria that are resistant to antibiotics is common in areas where antibiotics are used widely. The current standard procedure for detecting bacterial drug resistance is based on bacterial growth under antibiotic treatments. Here we describe the morphological changes in enoxacin-resistant Escherichia coli cells and the computational method used to identify these resistant cells in transmission electron microscopy (TEM) images without using antibiotics. Our approach was to create patches from TEM images of enoxacin-sensitive and enoxacin-resistant E. coli strains, use a convolutional neural network for patch classification, and identify the strains on the basis of the classification results. The proposed method was highly accurate in classifying cells, achieving an accuracy rate of 0.94. Using a gradient-weighted class activation mapping to visualize the region of interest, enoxacin-resistant and enoxacin-sensitive cells were characterized by comparing differences in the envelope. Moreover, Pearson's correlation coefficients suggested that four genes, including lpp, the gene encoding the major outer membrane lipoprotein, were strongly associated with the image features of enoxacin-resistant cells.

Bayesian optimization with constraint on passed charge for multiparameter screening of electrochemical reductive carboxylation in a flow microreactor.

Multiparameter screening of reductive carboxylation in an electrochemical flow microreactor was performed using a Bayesian optimization (BO) strategy. The developed algorithm features a constraint on passed charge for the electrochemical reaction, which led to suitable conditions being instantaneously found for the desired reaction. Analysis of the BO-suggested conditions underscored the physicochemical validity.

Bayesian optimization-driven parallel-screening of multiple parameters for the flow synthesis of biaryl compounds.

Traditional optimization methods using one variable at a time approach waste time and chemicals and assume that different parameters are independent from one another. Hence, a simpler, more practical, and rapid process for predicting reaction conditions that can be applied to several manufacturing environmentally sustainable processes is highly desirable. In this study, biaryl compounds were synthesized efficiently using an organic Brønsted acid catalyst in a flow system. Bayesian optimization-assisted multi-parameter screening, which employs one-hot encoding and appropriate acquisition function, rapidly predicted the suitable conditions for the synthesis of 2-amino-2'-hydroxy-biaryls (maximum yield of 96%). The established protocol was also applied in an optimization process for the efficient synthesis of 2,2'-dihydroxy biaryls (up to 97% yield). The optimized reaction conditions were successfully applied to gram-scale synthesis. We believe our algorithm can be beneficial as it can screen a reactor design without complicated quantification and descriptors.

Deep Learning-Enhanced Nanopore Sensing of Single-Nanoparticle Translocation Dynamics.

Noise is ubiquitous in real space that hinders detection of minute yet important signals in electrical sensors. Here, the authors report on a deep learning approach for denoising ionic current in resistive pulse sensing. Electrophoretically-driven translocation motions of single-nanoparticles in a nano-corrugated nanopore are detected. The noise is reduced by a convolutional auto-encoding neural network, designed to iteratively compare and minimize differences between a pair of waveforms via a gradient descent optimization. This denoising in a high-dimensional feature space is demonstrated to allow detection of the corrugation-derived wavy signals that cannot be identified in the raw curves nor after digital processing in frequency domains under the given noise floor, thereby enabled in-situ tracking to electrokinetic analysis of fast-moving single- and double-nanoparticles. The ability of the unlabeled learning to remove noise without compromising temporal resolution may be useful in solid-state nanopore sensing of protein structure and polynucleotide sequence.

Detecting Single Molecule Deoxyribonucleic Acid in a Cell Using a Three-Dimensionally Integrated Nanopore.

Amplification-free genome analysis can revolutionize biology and medicine by uncovering genetic variations among individuals. Here, the authors report on a 3D-integrated nanopore for electrolysis to in situ detection of single-molecule DNA in a cell by ionic current measurements. It consists of a SiO2 multipore sheet and a SiNx nanopore membrane stacked vertically on a Si wafer. Single cell lysis is demonstrated by 106  V m-1 -level electrostatic field focused at the multinanopore. The intracellular molecules are then directly detected as they move through a sensing zone, wherein the authors find telegraphic current signatures reflecting folding degrees of freedom of the millimeter-long polynucleotides threaded through the SiNx nanopore. The present device concept may enable on-chip single-molecule sequencing to multi-omics analyses at a single-cell level.

Analysis on Microstructure-Property Linkages of Filled Rubber Using Machine Learning and Molecular Dynamics Simulations.

A better understanding of the microstructure-property relationship can be achieved by sampling and analyzing a microstructure leading to a desired material property. During the simulation of filled rubber, this approach includes extracting common aggregates from a complex filler morphology consisting of hundreds of filler particles. However, a method for extracting a core structure that determines the rubber mechanical properties has not been established yet. In this study, we analyzed complex filler morphologies that generated extremely high stress using two machine learning techniques. First, filler morphology was quantified by persistent homology and then vectorized using persistence image as the input data. After that, a binary classification model involving logistic regression analysis was developed by training a dataset consisting of the vectorized morphology and stress-based class. The filler aggregates contributing to the desired mechanical properties were extracted based on the trained regression coefficients. Second, a convolutional neural network was employed to establish a classification model by training a dataset containing the imaged filler morphology and class. The aggregates strongly contributing to stress generation were extracted by a kernel. The aggregates extracted by both models were compared, and their shapes and distributions producing high stress levels were discussed. Finally, we confirmed the effects of the extracted aggregates on the mechanical property, namely the validity of the proposed method for extracting stress-contributing fillers, by performing coarse-grained molecular dynamics simulations.

Combining machine learning and nanopore construction creates an artificial intelligence nanopore for coronavirus detection.

High-throughput, high-accuracy detection of emerging viruses allows for the control of disease outbreaks. Currently, reverse transcription-polymerase chain reaction (RT-PCR) is currently the most-widely used technology to diagnose the presence of SARS-CoV-2. However, RT-PCR requires the extraction of viral RNA from clinical specimens to obtain high sensitivity. Here, we report a method for detecting novel coronaviruses with high sensitivity by using nanopores together with artificial intelligence, a relatively simple procedure that does not require RNA extraction. Our final platform, which we call the artificially intelligent nanopore, consists of machine learning software on a server, a portable high-speed and high-precision current measuring instrument, and scalable, cost-effective semiconducting nanopore modules. We show that artificially intelligent nanopores are successful in accurately identifying four types of coronaviruses similar in size, HCoV-229E, SARS-CoV, MERS-CoV, and SARS-CoV-2. Detection of SARS-CoV-2 in saliva specimen is achieved with a sensitivity of 90% and specificity of 96% with a 5-minute measurement.

A photoswitchable fluorescent protein for hours-time-lapse and sub-second-resolved super-resolution imaging.

Reversibly photoswitchable fluorescent proteins (RSFPs) are a class of fluorescent proteins whose fluorescence can be turned on and off by light irradiation. RSFPs have become essential tools for super-resolution (SR) imaging. Because most SR imaging techniques require high-power-density illumination, mitigating phototoxicity in cells due to intense light irradiation has been a challenge. Although we previously developed an RSFP named Kohinoor to achieve SR imaging with low phototoxicity, the photoproperties were insufficient to move a step further to explore the cellular dynamics by SR imaging. Here, we show an improved version of RSFP, Kohinoor2.0, which is suitable for SR imaging of cellular processes. Kohinoor2.0 shows a 2.6-fold higher fluorescence intensity, 2.5-fold faster chromophore maturation and 1.5-fold faster off-switching than Kohinoor. The analysis of the pH dependence of the visible absorption band revealed that Kohinoor2.0 and Kohinoor were in equilibria among multiple fluorescently bright and dark states, with the mutations introduced into Kohinoor2.0 bringing about a higher stabilization of the fluorescently bright states compared to Kohinoor. Using Kohinoor2.0 with our SR imaging technique, super-resolution polarization demodulation/on-state polarization angle narrowing, we conducted 4-h time-lapse SR imaging of an actin filament network in mammalian cells with a total acquisition time of 480 s without a noticeable indication of phototoxicity. Furthermore, we demonstrated the SR imaging of mitochondria dynamics at a time resolution of 0.5 s, in which the fusion and fission processes were clearly visualized. Thus, Kohinoor2.0 is shown to be an invaluable RSFP for the SR imaging of cellular dynamics.

Correction: Exploration of flow reaction conditions using machine-learning for enantioselective organocatalyzed Rauhut-Currier and [3+2] annulation sequence.

Correction for 'Exploration of flow reaction conditions using machine-learning for enantioselective organocatalyzed Rauhut-Currier and [3+2] annulation sequence' by Masaru Kondo et al., Chem. Commun., 2020, 56, 1259-1262, DOI: 10.1039/C9CC08526B.

Synthesis of computer simulation and machine learning for achieving the best material properties of filled rubber.

Molecular dynamics (MD) simulation is used to analyze the mechanical properties of polymerized and nanoscale filled rubber. Unfortunately, the computation time for a simulation can require several months' computing power, because the interactions of thousands of filler particles must be calculated. To alleviate this problem, we introduce a surrogate convolutional neural network model to achieve faster and more accurate predictions. The major difficulty when employing machine-learning-based surrogate models is the shortage of training data, contributing to the huge simulation costs. To derive a highly accurate surrogate model using only a small amount of training data, we increase the number of training instances by dividing the large-scale simulation results into 3D images of middle-scale filler morphologies and corresponding regional stresses. The images include fringe regions to reflect the influence of the filler constituents outside the core regions. The resultant surrogate model provides higher prediction accuracy than that trained only by images of the entire region. Afterwards, we extract the fillers that dominate the mechanical properties using the surrogate model and we confirm their validity using MD.

Digital Pathology Platform for Respiratory Tract Infection Diagnosis via Multiplex Single-Particle Detections.

The variability of bioparticles remains a key barrier to realizing the competent potential of nanoscale detection into a digital diagnosis of an extraneous object that causes an infectious disease. Here, we report label-free virus identification based on machine-learning classification. Single virus particles were detected using nanopores, and resistive-pulse waveforms were analyzed multilaterally using artificial intelligence. In the discrimination, over 99% accuracy for five different virus species was demonstrated. This advance is accessed through the classification of virus-derived ionic current signal patterns reflecting their intrinsic physical properties in a high-dimensional feature space. Moreover, consideration of viral similarity based on the accuracies indicates the contributing factors in the recognitions. The present findings offer the prospect of a novel surveillance system applicable to detection of multiple viruses including new strains.

Machine learning-driven electronic identifications of single pathogenic bacteria.

A rapid method for screening pathogens can revolutionize health care by enabling infection control through medication before symptom. Here we report on label-free single-cell identifications of clinically-important pathogenic bacteria by using a polymer-integrated low thickness-to-diameter aspect ratio pore and machine learning-driven resistive pulse analyses. A high-spatiotemporal resolution of this electrical sensor enabled to observe galvanotactic response intrinsic to the microbes during their translocation. We demonstrated discrimination of the cellular motility via signal pattern classifications in a high-dimensional feature space. As the detection-to-decision can be completed within milliseconds, the present technique may be used for real-time screening of pathogenic bacteria for environmental and medical applications.

Nano-corrugated Nanochannels for In Situ Tracking of Single-Nanoparticle Translocation Dynamics.

Dynamic motions of materials in liquid present a wealth of information concerning their physical properties. While fluorescence microscopy has been widely utilized for single-particle observations, the method cannot be used for characterizing fast motions of nanoscale objects due to the limited spatiotemporal resolution. Here, we report on a nanostructure strategy for nanoscale tracking of single nanoparticles. We fabricated a straight conduit in a SiO2 layer on a Si wafer with lithographically defined 30 nm-sized protrusions formed on the side walls. We performed resistive pulse measurements at a 1 MHz sampling rate wherein we found n-stepped current traces signifying n number of nanoparticles moving concurrently inside the nanochannel. Ensemble average of the ionic current signals revealed a peculiar feature reflecting the slightly stronger ion blockage at the nanoconstrictions between the protrusions, thereby proving the ability of nano-corrugation as physical gates to signify the precise positions of objects inside the nanofluidic channel. This in situ tracking approach elucidated steady-state motions of the nanoparticles moving at a constant speed under the counter-balanced electrophoretic and viscous drag forces, which also allowed estimations of their surface charge densities. The present method can be utilized as a speedometer for nanoscale objects of virtually any size as long as they are able to be put through the sensing zones with potential applications for single-molecule time-of-flight mass spectrometry.