Machine learning in cell biology - teaching computers to recognize phenotypes

A multiparametric analysis including single-cell and subcellular feature assessment reveals differential behavior of spheroid cultures on distinct ultra-low attachment plate types

Spheroids have become principal three-dimensional models to study cancer, developmental processes, and drug efficacy. Single-cell analysis techniques have emerged as ideal tools to gauge the complexity of cellular responses in these models. However, the single-cell quantitative assessment based on 3D-microscopic data of the subcellular distribution of fluorescence markers, such as the nuclear/cytoplasm ratio of transcription factors, has largely remained elusive. For spheroid generation, ultra-low attachment plates are noteworthy due to their simplicity, compatibility with automation, and experimental and commercial accessibility. However, it is unknown whether and to what degree the plate type impacts spheroid formation and biology. This study developed a novel AI-based pipeline for the analysis of 3D-confocal data of optically cleared large spheroids at the wholemount, single-cell, and sub-cellular levels. To identify relevant samples for the pipeline, automated brightfield microscopy was employed to systematically compare the size and eccentricity of spheroids formed in six different plate types using four distinct human cell lines. This showed that all plate types exhibited similar spheroid-forming capabilities and the gross patterns of growth or shrinkage during 4 days after seeding were comparable. Yet, size and eccentricity varied systematically among specific cell lines and plate types. Based on this prescreen, spheroids of HaCaT keratinocytes and HT-29 cancer cells were further assessed. In HaCaT spheroids, the in-depth analysis revealed a correlation between spheroid size, cell proliferation, and the nuclear/cytoplasm ratio of the transcriptional coactivator, YAP1, as well as an inverse correlation with respect to cell differentiation. These findings, yielded with a spheroid model and at a single-cell level, corroborate earlier concepts of the role of YAP1 in cell proliferation and differentiation of keratinocytes in human skin. Further, the results show that the plate type may influence the outcome of experimental campaigns and that it is advisable to scan different plate types for the optimal configuration during a specific investigation.

The influence of hyperchaoticity, synchronization, and Shannon entropy on the performance of a physical reservoir computer

In this paper, we analyze the dynamic effect of a reservoir computer (RC) on its performance. Modified Kuramoto's coupled oscillators are used to model the RC, and synchronization, Lyapunov spectrum (and dimension), Shannon entropy, and the upper bound of the Kolmogorov-Sinai entropy are employed to characterize the dynamics of the RC. The performance of the RC is analyzed by reproducing the distribution of random, Gaussian, and quantum jumps series (shelved states) since a replica of the time evolution of a completely random series is not possible to generate. We demonstrate that hyperchaotic motion, moderate Shannon entropy, and a higher degree of synchronization of Kuramoto's oscillators lead to the best performance of the RC. Therefore, an appropriate balance of irregularity and order in the oscillator's dynamics leads to better performances.

PhagoStat a scalable and interpretable end to end framework for efficient quantification of cell phagocytosis in neurodegenerative disease studies

Quantifying the phagocytosis of dynamic, unstained cells is essential for evaluating neurodegenerative diseases. However, measuring rapid cell interactions and distinguishing cells from background make this task very challenging when processing time-lapse phase-contrast video microscopy. In this study, we introduce an end-to-end, scalable, and versatile real-time framework for quantifying and analyzing phagocytic activity. Our proposed pipeline is able to process large data-sets and includes a data quality verification module to counteract potential perturbations such as microscope movements and frame blurring. We also propose an explainable cell segmentation module to improve the interpretability of deep learning methods compared to black-box algorithms. This includes two interpretable deep learning capabilities: visual explanation and model simplification. We demonstrate that interpretability in deep learning is not the opposite of high performance, by additionally providing essential deep learning algorithm optimization insights and solutions. Besides, incorporating interpretable modules results in an efficient architecture design and optimized execution time. We apply this pipeline to quantify and analyze microglial cell phagocytosis in frontotemporal dementia (FTD) and obtain statistically reliable results showing that FTD mutant cells are larger and more aggressive than control cells. The method has been tested and validated on several public benchmarks by generating state-of-the art performances. To stimulate translational approaches and future studies, we release an open-source end-to-end pipeline and a unique microglial cells phagocytosis dataset for immune system characterization in neurodegenerative diseases research. This pipeline and the associated dataset will consistently crystallize future advances in this field, promoting the development of efficient and effective interpretable algorithms dedicated to the critical domain of neurodegenerative diseases' characterization. https://github.com/ounissimehdi/PhagoStat .

Super resolution-based methodology for self-supervised segmentation of microscopy images

Data-driven Artificial Intelligence (AI)/Machine learning (ML) image analysis approaches have gained a lot of momentum in analyzing microscopy images in bioengineering, biotechnology, and medicine. The success of these approaches crucially relies on the availability of high-quality microscopy images, which is often a challenge due to the diverse experimental conditions and modes under which these images are obtained. In this study, we propose the use of recent ML-based image super-resolution (SR) techniques for improving the image quality of microscopy images, incorporating them into multiple ML-based image analysis tasks, and describing a comprehensive study, investigating the impact of SR techniques on the segmentation of microscopy images. The impacts of four Generative Adversarial Network (GAN)- and transformer-based SR techniques on microscopy image quality are measured using three well-established quality metrics. These SR techniques are incorporated into multiple deep network pipelines using supervised, contrastive, and non-contrastive self-supervised methods to semantically segment microscopy images from multiple datasets. Our results show that the image quality of microscopy images has a direct influence on the ML model performance and that both supervised and self-supervised network pipelines using SR images perform better by 2%-6% in comparison to baselines, not using SR. Based on our experiments, we also establish that the image quality improvement threshold range [20-64] for the complemented Perception-based Image Quality Evaluator(PIQE) metric can be used as a pre-condition by domain experts to incorporate SR techniques to significantly improve segmentation performance. A plug-and-play software platform developed to integrate SR techniques with various deep networks using supervised and self-supervised learning methods is also presented.

The ReadFree tool for the identification of poor readers: a validation study based on a machine learning approach in monolingual and minority-language children

In this study, we validated the "ReadFree tool", a computerised battery of 12 visual and auditory tasks developed to identify poor readers also in minority-language children (MLC). We tested the task-specific discriminant power on 142 Italian-monolingual participants (8-13 years old) divided into monolingual poor readers (N = 37) and good readers (N = 105) according to standardised Italian reading tests. The performances at the discriminant tasks of the "ReadFree tool" were entered into a classification and regression tree (CART) model to identify monolingual poor and good readers. The set of classification rules extracted from the CART model were applied to the MLC's performance and the ensuing classification was compared to the one based on standardised Italian reading tests. According to the CART model, auditory go-no/go (regular), RAN and Entrainment100bpm were the most discriminant tasks. When compared with the clinical classification, the CART model accuracy was 86% for the monolinguals and 76% for the MLC. Executive functions and timing skills turned out to have a relevant role in reading. Results of the CART model on MLC support the idea that ad hoc standardised tasks that go beyond reading are needed.

Unsupervised machine learning discovers classes in aluminium alloys

Aluminium (Al) alloys are critical to many applications. Although Al alloys have been commercially widespread for over a century, their development has predominantly taken a trial-and-error approach. Furthermore, many discrete studies regarding Al alloys, often application specific, have precluded a broader consolidation of Al alloy classification. Iterative label spreading (ILS), an unsupervised machine learning approach, was used to identify the different classes of Al alloys, drawing from a specifically curated dataset of 1154 Al alloys (including alloy composition and processing conditions). Using ILS, eight classes of Al alloys were identified based on a comprehensive feature set under two descriptors. Further, a decision tree classifier was used to validate the separation of classes.

Rethinking the Dilated Encoder in TE-YOLOF: An Approach Based on Attention Mechanism to Improve Performance for Blood Cell Detection

Blood cell detection is an essential branch of microscopic imaging for disease diagnosis. TE-YOLOF is an effective model for blood cell detection, and was recently found to have an outstanding trade-off between accuracy and model complexity. However, there is a lack of understanding of whether the dilated encoder in TE-YOLOF works well for blood cell detection. To address this issue, we perform a thorough experimental analysis and find the interesting fact that the dilated encoder is not necessary for TE-YOLOF to perform the blood cell detection task. For the purpose of increasing performance on blood cell detection, in this research, we use the attention mechanism to dominate the dilated encoder place in TE-YOLOF and find that the attention mechanism is effective to address this problem. Based upon these findings, we propose a novel approach, named Enhanced Channel Attention Module (ECAM), based on attention mechanism to achieve precision improvement with less growth on model complexity. Furthermore, we examine the proposed ECAM method compared with other tip-top attention mechanisms and find that the proposed attention method is more effective on blood cell detection task. We incorporate the spatial attention mechanism in CBAM with our ECAM to form a new module, which is named Enhanced-CBAM. We propose a new network named Enhanced Channel Attention Network (ENCANet) based upon Enhanced-CBAM to perform blood cell detection on BCCD dataset. This network can increase the accuracy to 90.3 AP while the parameter is only 6.5 M. Our ENCANet is also effective for conducting cross-domain blood cell detection experiments.

Lateral elbow tendinopathy and artificial intelligence: Binary and multilabel findings detection using machine learning algorithms

Background

Ultrasound (US) is a valuable technique to detect degenerative findings and intrasubstance tears in lateral elbow tendinopathy (LET). Machine learning methods allow supporting this radiological diagnosis.

Aim

To assess multilabel classification models using machine learning models to detect degenerative findings and intrasubstance tears in US images with LET diagnosis.

Materials and methods

A retrospective study was performed. US images and medical records from patients with LET diagnosis from January 1st, 2017, to December 30th, 2018, were selected. Datasets were built for training and testing models. For image analysis, features extraction, texture characteristics, intensity distribution, pixel-pixel co-occurrence patterns, and scales granularity were implemented. Six different supervised learning models were implemented for binary and multilabel classification. All models were trained to classify four tendon findings (hypoechogenicity, neovascularity, enthesopathy, and intrasubstance tear). Accuracy indicators and their confidence intervals (CI) were obtained for all models following a K-fold-repeated-cross-validation method. To measure multilabel prediction, multilabel accuracy, sensitivity, specificity, and receiver operating characteristic (ROC) with 95% CI were used.

Results

A total of 30,007 US images (4,324 exams, 2,917 patients) were included in the analysis. The RF model presented the highest mean values in the area under the curve (AUC), sensitivity, and also specificity by each degenerative finding in the binary classification. The AUC and sensitivity showed the best performance in intrasubstance tear with 0.991 [95% CI, 099, 0.99], and 0.775 [95% CI, 0.77, 0.77], respectively. Instead, specificity showed upper values in hypoechogenicity with 0.821 [95% CI, 0.82, −0.82]. In the multilabel classifier, RF also presented the highest performance. The accuracy was 0.772 [95% CI, 0.771, 0.773], a great macro of 0.948 [95% CI, 0.94, 0.94], and a micro of 0.962 [95% CI, 0.96, 0.96] AUC scores were detected. Diagnostic accuracy, sensitivity, and specificity with 95% CI were calculated.

Conclusion

Machine learning algorithms based on US images with LET presented high diagnosis accuracy. Mainly the random forest model shows the best performance in binary and multilabel classifiers, particularly for intrasubstance tears.

Combined nanometric and phylogenetic analysis of unique endocytic compartments in Giardia lamblia sheds light on the evolution of endocytosis in Metamonada

Background

Giardia lamblia, a parasitic protist of the Metamonada supergroup, has evolved one of the most diverged endocytic compartment systems investigated so far. Peripheral endocytic compartments, currently known as peripheral vesicles or vacuoles (PVs), perform bulk uptake of fluid phase material which is then digested and sorted either to the cell cytosol or back to the extracellular space.

Results

Here, we present a quantitative morphological characterization of these organelles using volumetric electron microscopy and super-resolution microscopy (SRM). We defined a morphological classification for the heterogenous population of PVs and performed a comparative analysis of PVs and endosome-like organelles in representatives of phylogenetically related taxa, Spironucleus spp. and Tritrichomonas foetus. To investigate the as-yet insufficiently understood connection between PVs and clathrin assemblies in G. lamblia, we further performed an in-depth search for two key elements of the endocytic machinery, clathrin heavy chain (CHC) and clathrin light chain (CLC), across different lineages in Metamonada. Our data point to the loss of a bona fide CLC in the last Fornicata common ancestor (LFCA) with the emergence of a protein analogous to CLC (GlACLC) in the Giardia genus. Finally, the location of clathrin in the various compartments was quantified.

Conclusions

Taken together, this provides the first comprehensive nanometric view of Giardia’s endocytic system architecture and sheds light on the evolution of GlACLC analogues in the Fornicata supergroup and, specific to Giardia, as a possible adaptation to the formation and maintenance of stable clathrin assemblies at PVs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12915-022-01402-3.

Unleashing high content screening in hit detection – benchmarking AI workflows including novelty detection

Complex mixtures containing natural products are still an interesting source of novel drug candidates. High content screening (HCS) is a popular tool to screen for such. In particular, multiplexed HCS assays promise comprehensive bioactivity profiles, but generate also high amounts of data. Yet, only some machine learning (ML) applications for data analysis are available and these usually require a profound knowledge of the underlying cell biology. Unfortunately, there are no applications that simply predict if samples are biologically active or not (any kind of bioactivity). Within this work, we benchmark ML algorithms for binary classification, starting with classical ML models, which are the standard classifiers of the scikit-learn library or ensemble models of these classifiers (a total of 92 models tested). Followed by a partial least square regression (PLSR)-based classification (44 tested models in total) and simple artificial neural networks (ANNs) with dense layers (72 tested models in total). In addition, a novelty detection (ND) was examined, which is supposed to handle unknown patterns. For the final analysis the models, with and without upstream ND, were tested with two independent data sets. In our analysis, a stacking model, an ensamble model of class ML algorithms, performed best to predict new and unknown data. ND improved the predictions of the models and was useful to handle unknown patterns. Importantly, the classifier presented here can be easily rebuilt and be adapted to the data and demands of other groups. The hit detector (ND + stacking model) is universal and suitable for a broader application to support the search for new drug candidates.

ScanGrow: Deep Learning-Based Live Tracking of Bacterial Growth in Broth

Monitoring the growth of bacterial cultures is one of the most common techniques in microbiology. This is usually achieved by using expensive and bulky spectrophotometric plate readers which periodically measure the optical density of bacterial cultures during the incubation period. In this study, we present a completely novel way of obtaining bacterial growth curves based on the classification of scanned images of cultures rather than using spectrophotometric measurements. We trained a deep learning model with images of bacterial broths contained in microplates, and we integrated it into a custom-made software application that triggers a flatbed scanner to timely capture images, automatically processes the images, and represents all growth curves. The developed tool, ScanGrow, is presented as a low-cost and high-throughput alternative to plate readers, and it only requires a computer connected to a flatbed scanner and equipped with our open-source ScanGrow application. In addition, this application also assists in the pre-processing of data to create and evaluate new models, having the potential to facilitate many routine microbiological techniques.

Generalising from conventional pipelines using deep learning in high-throughput screening workflows

The study of complex diseases relies on large amounts of data to build models toward precision medicine. Such data acquisition is feasible in the context of high-throughput screening, in which the quality of the results relies on the accuracy of the image analysis. Although state-of-the-art solutions for image segmentation employ deep learning approaches, the high cost of manually generating ground truth labels for model training hampers the day-to-day application in experimental laboratories. Alternatively, traditional computer vision-based solutions do not need expensive labels for their implementation. Our work combines both approaches by training a deep learning network using weak training labels automatically generated with conventional computer vision methods. Our network surpasses the conventional segmentation quality by generalising beyond noisy labels, providing a 25% increase of mean intersection over union, and simultaneously reducing the development and inference times. Our solution was embedded into an easy-to-use graphical user interface that allows researchers to assess the predictions and correct potential inaccuracies with minimal human input. To demonstrate the feasibility of training a deep learning solution on a large dataset of noisy labels automatically generated by a conventional pipeline, we compared our solution against the common approach of training a model from a small manually curated dataset by several experts. Our work suggests that humans perform better in context interpretation, such as error assessment, while computers outperform in pixel-by-pixel fine segmentation. Such pipelines are illustrated with a case study on image segmentation for autophagy events. This work aims for better translation of new technologies to real-world settings in microscopy-image analysis.

A versatile active learning workflow for optimization of genetic and metabolic networks

Optimization of biological networks is often limited by wet lab labor and cost, and the lack of convenient computational tools. Here, we describe METIS, a versatile active machine learning workflow with a simple online interface for the data-driven optimization of biological targets with minimal experiments. We demonstrate our workflow for various applications, including cell-free transcription and translation, genetic circuits, and a 27-variable synthetic CO₂-fixation cycle (CETCH cycle), improving these systems between one and two orders of magnitude. For the CETCH cycle, we explore 10²⁵ conditions with only 1,000 experiments to yield the most efficient CO₂-fixation cascade described to date. Beyond optimization, our workflow also quantifies the relative importance of individual factors to the performance of a system identifying unknown interactions and bottlenecks. Overall, our workflow opens the way for convenient optimization and prototyping of genetic and metabolic networks with customizable adjustments according to user experience, experimental setup, and laboratory facilities.

Optimization of biological networks is often limited by wet lab labor and cost, and the lack of convenient computational tools. Here, aimed at democratization and standardization, the authors describe METIS, a modular and versatile active machine learning workflow with a simple online interface for the optimization of biological target functions with minimal experimental datasets.

Precise measurement of nanoscopic septin ring structures with deep learning-assisted quantitative superresolution microscopy

The combination of image analysis and superresolution microscopy methods allows for unprecedented insight into the organization of macromolecular assemblies in cells. Advances in deep learning (DL)-based object recognition enable the automated processing of large amounts of data, resulting in high accuracy through averaging. However, while the analysis of highly symmetric structures of constant size allows for a resolution approaching the dimensions of structural biology, DL-based image recognition may introduce bias. This prohibits the development of readouts for processes that involve significant changes in size or shape of amorphous macromolecular complexes. Here we address this problem by using changes of septin ring structures in single molecule localization-based superresolution microscopy data as a paradigm. We identify potential sources of bias resulting from different training approaches by rigorous testing of trained models using real or simulated data covering a wide range of possible results. In a quantitative comparison of our models, we find that a trade-off exists between measurement accuracy and the range of recognized phenotypes. Using our thus verified models, we find that septin ring size can be explained by the number of subunits they are assembled from alone. Furthermore, we provide a new experimental system for the investigation of septin polymerization.

Induced Pluripotent Stem Cell-Based Drug Screening by Use of Artificial Intelligence

Induced pluripotent stem cells (iPSCs) are terminally differentiated somatic cells that differentiate into various cell types. iPSCs are expected to be used for disease modeling and for developing novel treatments because differentiated cells from iPSCs can recapitulate the cellular pathology of patients with genetic mutations. However, a barrier to using iPSCs for comprehensive drug screening is the difficulty of evaluating their pathophysiology. Recently, the accuracy of image analysis has dramatically improved with the development of artificial intelligence (AI) technology. In the field of cell biology, it has become possible to estimate cell types and states by examining cellular morphology obtained from simple microscopic images. AI can evaluate disease-specific phenotypes of iPS-derived cells from label-free microscopic images; thus, AI can be utilized for disease-specific drug screening using iPSCs. In addition to image analysis, various AI-based methods can be applied to drug development, including phenotype prediction by analyzing genomic data and virtual screening by analyzing structural formulas and protein-protein interactions of compounds. In the future, combining AI methods may rapidly accelerate drug discovery using iPSCs. In this review, we explain the details of AI technology and the application of AI for iPSC-based drug screening.

Live-cell fluorescence spectral imaging as a data science challenge

Live-cell fluorescence spectral imaging is an evolving modality of microscopy that uses specific properties of fluorophores, such as excitation or emission spectra, to detect multiple molecules and structures in intact cells. The main challenge of analyzing live-cell fluorescence spectral imaging data is the precise quantification of fluorescent molecules despite the weak signals and high noise found when imaging living cells under non-phototoxic conditions. Beyond the optimization of fluorophores and microscopy setups, quantifying multiple fluorophores requires algorithms that separate or unmix the contributions of the numerous fluorescent signals recorded at the single pixel level. This review aims to provide both the experimental scientist and the data analyst with a straightforward description of the evolution of spectral unmixing algorithms for fluorescence live-cell imaging. We show how the initial systems of linear equations used to determine the concentration of fluorophores in a pixel progressively evolved into matrix factorization, clustering, and deep learning approaches. We outline potential future trends on combining fluorescence spectral imaging with label-free detection methods, fluorescence lifetime imaging, and deep learning image analysis.

Utility of extracellular vesicles as a potential biological indicator of physiological resilience during military operational stress

Extracellular vesicles (EVs) transport biological content between cells to mediate physiological processes. The association between EVs and resilience, the ability to cope with stress, is unknown. Using unbiased machine learning approaches, we aimed to identify a biological profile of resilience. Twenty servicemen (27.8 ± 5.9 years) completed the Connor Davidson Resilience (CD-RISC) questionnaire and were exposed to daily physical and cognitive exertion with 48-hr sleep and caloric restriction. Blood samples from baseline and the second day of stress were analyzed for neuroendocrine biomarkers impacted by military stress. EVs were isolated from plasma and stained with antibodies associated with exosomes (CD63), microvesicles (VAMP3), and apoptotic bodies (THSD1). Individuals were separated into high (n = 10, CD-RISC > 90) and low (n = 10, CD-RISC < 79) resilience. EV features were stratified by size, then down-selected using regression trees and compared between groups. Diagnostic accuracy was assessed using receiver operating characteristic curves. Compared to low resilience, high resilience demonstrated a greater increase in variability of THSD1 local bright spot intensities among large-sized EVs in response to stress (p = 0.002, Hedges' g = 1.59). Among medium-sized EVs, high resilience exhibited a greater decrease in side scatter intensity (p = 0.014, Hedges' g = 1.17). Both features demonstrated high to moderate diagnostic accuracy for high resilience (AUC = 0.90 and 0.79). In contrast, neuroendocrine biomarker concentrations were similar between groups. The increase in variability among THSD1 + EVs in high, but not low, resilient individuals following stress may suggest high resilience is accompanied by stress-triggered apoptotic adaptations to the environment that are not detected in neuroendocrine biomarkers.

A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation

DNA double-strand breaks (DSBs) are the most lethal form of damage to cells from irradiation. γ-H2AX (phosphorylated form of H2AX histone variant) has become one of the most reliable and sensitive biomarkers of DNA DSBs. However, the γ-H2AX foci assay still has limitations in the time consumed for manual scoring and possible variability between scorers. This study proposed a novel automated foci scoring method using a deep convolutional neural network based on a You-Only-Look-Once (YOLO) algorithm to quantify γ-H2AX foci in peripheral blood samples. FociRad, a two-stage deep learning approach, consisted of mononuclear cell (MNC) and γ-H2AX foci detections. Whole blood samples were irradiated with X-rays from a 6 MV linear accelerator at 1, 2, 4 or 6 Gy. Images were captured using confocal microscopy. Then, dose-response calibration curves were established and implemented with unseen dataset. The results of the FociRad model were comparable with manual scoring. MNC detection yielded 96.6% accuracy, 96.7% sensitivity and 96.5% specificity. γ-H2AX foci detection showed very good F1 scores (> 0.9). Implementation of calibration curve in the range of 0-4 Gy gave mean absolute difference of estimated doses less than 1 Gy compared to actual doses. In addition, the evaluation times of FociRad were very short (< 0.5 min per 100 images), while the time for manual scoring increased with the number of foci. In conclusion, FociRad was the first automated foci scoring method to use a YOLO algorithm with high detection performance and fast evaluation time, which opens the door for large-scale applications in radiation triage.

Machine Learning Classifies Ferroptosis and Apoptosis Cell Death Modalities with TfR1 Immunostaining

Determining cell death mechanisms occurring in patient and animal tissues is a longstanding goal that requires suitable biomarkers and accurate quantification. However, effective methods remain elusive. To develop more powerful and unbiased analytic frameworks, we developed a machine learning approach for automated cell death classification. Image sets were collected of HT-1080 fibrosarcoma cells undergoing ferroptosis or apoptosis and stained with an anti-transferrin receptor 1 (TfR1) antibody, together with nuclear and F-actin staining. Features were extracted using high-content-analysis software, and a classifier was constructed by fitting a multinomial logistic lasso regression model to the data. The prediction accuracy of the classifier within three classes (control, ferroptosis, apoptosis) was 93%. Thus, TfR1 staining, combined with nuclear and F-actin staining, can reliably detect both apoptotic and ferroptotis cells when cell features are analyzed in an unbiased manner using machine learning, providing a method for unbiased analysis of modes of cell death.

Implicitly perturbed Hamiltonian as a class of versatile and general-purpose molecular representations for machine learning

Unraveling challenging problems by machine learning has recently become a hot topic in many scientific disciplines. For developing rigorous machine-learning models to study problems of interest in molecular sciences, translating molecular structures to quantitative representations as suitable machine-learning inputs play a central role. Many different molecular representations and the state-of-the-art ones, although efficient in studying numerous molecular features, still are suboptimal in many challenging cases, as discussed in the context of the present research. The main aim of the present study is to introduce the Implicitly Perturbed Hamiltonian (ImPerHam) as a class of versatile representations for more efficient machine learning of challenging problems in molecular sciences. ImPerHam representations are defined as energy attributes of the molecular Hamiltonian, implicitly perturbed by a number of hypothetic or real arbitrary solvents based on continuum solvation models. We demonstrate the outstanding performance of machine-learning models based on ImPerHam representations for three diverse and challenging cases of predicting inhibition of the CYP450 enzyme, high precision, and transferrable evaluation of non-covalent interaction energy of molecular systems, and accurately reproducing solvation free energies for large benchmark sets.

Molecular representations are fundamental tools for machine-learning models. The current work introduces a new set of molecular representations demonstrated to enable accurate predictions of molecular conformational energy and solvation free energy.

Statistical distortion of supervised learning predictions in optical microscopy induced by image compression

The growth of data throughput in optical microscopy has triggered the extensive use of supervised learning (SL) models on compressed datasets for automated analysis. Investigating the effects of image compression on SL predictions is therefore pivotal to assess their reliability, especially for clinical use. We quantify the statistical distortions induced by compression through the comparison of predictions on compressed data to the raw predictive uncertainty, numerically estimated from the raw noise statistics measured via sensor calibration. Predictions on cell segmentation parameters are altered by up to 15% and more than 10 standard deviations after 16-to-8 bits pixel depth reduction and 10:1 JPEG compression. JPEG formats with higher compression ratios show significantly larger distortions. Interestingly, a recent metrologically accurate algorithm, offering up to 10:1 compression ratio, provides a prediction spread equivalent to that stemming from raw noise. The method described here allows to set a lower bound to the predictive uncertainty of a SL task and can be generalized to determine the statistical distortions originated from a variety of processing pipelines in AI-assisted fields.

Deep Learning-Enabled Technologies for Bioimage Analysis

Deep learning (DL) is a subfield of machine learning (ML), which has recently demonstrated its potency to significantly improve the quantification and classification workflows in biomedical and clinical applications. Among the end applications profoundly benefitting from DL, cellular morphology quantification is one of the pioneers. Here, we first briefly explain fundamental concepts in DL and then we review some of the emerging DL-enabled applications in cell morphology quantification in the fields of embryology, point-of-care ovulation testing, as a predictive tool for fetal heart pregnancy, cancer diagnostics via classification of cancer histology images, autosomal polycystic kidney disease, and chronic kidney diseases.

Moving Towards Induced Pluripotent Stem Cell-based Therapies with Artificial Intelligence and Machine Learning

The advent of induced pluripotent stem cell (iPSC) technology, which allows to transform one cell type into another, holds the promise to produce therapeutic cells and organs on demand. Realization of this objective is contingent on the ability to demonstrate quality and safety of the cellular product for its intended use. Bottlenecks and backlogs to the clinical use of iPSCs have been fully outlined and a need has emerged for safer and standardized protocols to trigger cell reprogramming and functional differentiation. Amidst great challenges, in particular associated with lengthy culture time and laborious cell characterization, a demand for faster and more accurate methods for the validation of cell identity and function at different stages of the iPSC manufacturing process has risen. Artificial intelligence-based methods are proving helpful for these complex tasks and might revolutionize the way iPSCs are managed to create surrogate cells and organs. Here, we briefly review recent progress in artificial intelligence approaches for evaluation of iPSCs and their derivatives in experimental studies.

Quantitative Evaluation of Cardiac Cell Interactions and Responses to Cyclic Strain

The heart has a dynamic mechanical environment contributed by its unique cellular composition and the resultant complex tissue structure. In pathological heart tissue, both the mechanics and cell composition can change and influence each other. As a result, the interplay between the cell phenotype and mechanical stimulation needs to be considered to understand the biophysical cell interactions and organization in healthy and diseased myocardium. In this work, we hypothesized that the overall tissue organization is controlled by varying densities of cardiomyocytes and fibroblasts in the heart. In order to test this hypothesis, we utilized a combination of mechanical strain, co-cultures of different cell types, and inhibitory drugs that block intercellular junction formation. To accomplish this, an image analysis pipeline was developed to automatically measure cell type-specific organization relative to the stretch direction. The results indicated that cardiac cell type-specific densities influence the overall organization of heart tissue such that it is possible to model healthy and fibrotic heart tissue in vitro. This study provides insight into how to mimic the dynamic mechanical environment of the heart in engineered tissue as well as providing valuable information about the process of cardiac remodeling and repair in diseased hearts.

Neural network fast-classifies biological images through features selecting to power automated microscopy

Artificial intelligence is nowadays used for cell detection and classification in optical microscopy during post-acquisition analysis. The microscopes are now fully automated and next expected to be smart by making acquisition decisions based on the images. It calls for analysing them on the fly. Biology further imposes training on a reduced data set due to cost and time to prepare the samples and have the data sets annotated by experts. We propose a real-time image processing compliant with these specifications by balancing accurate detection and execution performance. We characterised the images using a generic, high-dimensional feature extractor. We then classified the images using machine learning to understand the contribution of each feature in decision and execution time. We found that the non-linear-classifier random forests outperformed Fisher's linear discriminant. More importantly, the most discriminant and time-consuming features could be excluded without significant accuracy loss, offering a substantial gain in execution time. It suggests a feature-group redundancy likely related to the biology of the observed cells. We offer a method to select fast and discriminant features. In our assay, a 79.6 ± 2.4% accurate classification of a cell took 68.7 ± 3.5 ms (mean ± SD, 5-fold cross-validation nested in 10 bootstrap repeats), corresponding to 14 cells per second, dispatched into eight phases of the cell cycle, using 12 feature groups and operating a consumer market ARM-based embedded system. A simple neural network offered similar performances paving the way to faster training and classification, using parallel execution on a general-purpose graphic processing unit. Finally, this strategy is also usable for deep neural networks paving the way to optimizing these algorithms for smart microscopy.

Generative Adversarial Domain Adaptation for Nucleus Quantification in Images of Tissue Immunohistochemically Stained for Ki-67

PURPOSE

We focus on the problem of scarcity of annotated training data for nucleus recognition in Ki-67 immunohistochemistry (IHC)–stained pancreatic neuroendocrine tumor (NET) images. We hypothesize that deep learning–based domain adaptation is helpful for nucleus recognition when image annotations are unavailable in target data sets.

METHODS

We considered 2 different institutional pancreatic NET data sets: one (ie, source) containing 38 cases with 114 annotated images and the other (ie, target) containing 72 cases with 20 annotated images. The gold standards were manually annotated by 1 pathologist. We developed a novel deep learning–based domain adaptation framework to count different types of nuclei (ie, immunopositive tumor, immunonegative tumor, nontumor nuclei). We compared the proposed method with several recent fully supervised deep learning models, such as fully convolutional network-8s (FCN-8s), U-Net, fully convolutional regression network (FCRN) A, FCRNB, and fully residual convolutional network (FRCN). We also evaluated the proposed method by learning with a mixture of converted source images and real target annotations.

RESULTS

Our method achieved an F₁ score of 81.3% and 62.3% for nucleus detection and classification in the target data set, respectively. Our method outperformed FCN-8s (53.6% and 43.6% for nucleus detection and classification, respectively), U-Net (61.1% and 47.6%), FCRNA (63.4% and 55.8%), and FCRNB (68.2% and 60.6%) in terms of F₁ score and was competitive with FRCN (81.7% and 70.7%). In addition, learning with a mixture of converted source images and only a small set of real target labels could further boost the performance.

CONCLUSION

This study demonstrates that deep learning–based domain adaptation is helpful for nucleus recognition in Ki-67 IHC stained images when target data annotations are not available. It would improve the applicability of deep learning models designed for downstream supervised learning tasks on different data sets.

Machine Learning and Artificial Intelligence for the Prediction of Host-Pathogen Interactions: A Viral Case

The research of interactions between the pathogens and their hosts is key for understanding the biology of infection. Commencing on the level of individual molecules, these interactions define the behavior of infectious agents and the outcomes they elicit. Discovery of host-pathogen interactions (HPIs) conventionally involves a stepwise laborious research process. Yet, amid the global pandemic the urge for rapid discovery acceleration through the novel computational methodologies has become ever so poignant. This review explores the challenges of HPI discovery and investigates the efforts currently undertaken to apply the latest machine learning (ML) and artificial intelligence (AI) methodologies to this field. This includes applications to molecular and genetic data, as well as image and language data. Furthermore, a number of breakthroughs, obstacles, along with prospects of AI for host-pathogen interactions (HPI), are discussed.

Modeling and Predicting the Cell Migration Properties from Scratch Wound Healing Assay on Cisplatin-Resistant Ovarian Cancer Cell Lines Using Artificial Neural Network

The study of artificial neural networks (ANN) has undergone a tremendous revolution in recent years, boosted by deep learning tools. The presence of a greater number of learning tools and their applications, in particular, favors this revolution. However, there is a significant need to deal with the issue of implementing a systematic method during the development phase of the ANN to increase its performance. A multilayer feedforward neural network (FNN) was proposed in this paper to predict the cell migration assay on cisplatin-sensitive and cisplatin-resistant (CisR) ovarian cancer (OC) cell lines via scratch wound healing assay. An FNN training algorithm model was generated using the MATLAB fitting function in a MATLAB script to accomplish this task. The input parameters were types of cell lines, times, and wound area, and outputs were relative wound area, percentage of wound closure, and wound healing speed. In addition, we tested and compared the initial accuracy of various supervised learning classifier and support vector regression (SVR) algorithms. The proposed ANN model achieved good agreement with the experimental data and minimized error between the estimated and experimental values. The conclusions drawn demonstrate that the developed ANN model is a useful, accurate, fast, and inexpensive method to predict cancerous cell migration characteristics evaluated via scratch wound healing assay.

Review: Application of Artificial Intelligence in Phenomics

Plant phenomics has been rapidly advancing over the past few years. This advancement is attributed to the increased innovation and availability of new technologies which can enable the high-throughput phenotyping of complex plant traits. The application of artificial intelligence in various domains of science has also grown exponentially in recent years. Notably, the computer vision, machine learning, and deep learning aspects of artificial intelligence have been successfully integrated into non-invasive imaging techniques. This integration is gradually improving the efficiency of data collection and analysis through the application of machine and deep learning for robust image analysis. In addition, artificial intelligence has fostered the development of software and tools applied in field phenotyping for data collection and management. These include open-source devices and tools which are enabling community driven research and data-sharing, thereby availing the large amounts of data required for the accurate study of phenotypes. This paper reviews more than one hundred current state-of-the-art papers concerning AI-applied plant phenotyping published between 2010 and 2020. It provides an overview of current phenotyping technologies and the ongoing integration of artificial intelligence into plant phenotyping. Lastly, the limitations of the current approaches/methods and future directions are discussed.

Versatile Microfluidic Platform for Automated Live-Cell Hyperspectral Imaging Applied to Cold Climate Cyanobacterial Biofilms

Microfluidic bioanalytical platforms are driving discoveries from synthetic biology to the health sciences. In this work, we present a platform for in vivo live-cell imaging and automated species detection in mixed cyanobacterial biofilms from cold climate environments. Using a multimodal microscope with custom optics applied to a chip with six parallel growth channels, we monitored biofilm dynamics via continuous imaging at natural irradiance levels. Machine learning algorithms were applied to the collected hyperspectral images for automatic segmentation of mixed-species biofilms into individual species of cyanobacteria with similar filamentous morphology. The coupling of microfluidic technology with modern multimodal imaging and computer vision systems provides a versatile platform for the study of cause-and-effect scenarios of cyanobacterial biofilms, which are important elements of many ecosystems, including lakes and rivers of the polar regions.

Size Effect of a Piezoelectric Patch on a Rectangular Plate with the Neural Network Model

Artificial neural networks have been widely used in many studies, such as the prediction of the piezoelectric effect of the plate of engineering structures in vibration and noise reduction. In this paper, an artificial neural network (ANN) model was employed to explore the piezoelectric patch size and thickness’s effect on the first order natural frequency and displacement amplitude of a plate. With the finite element method (FEM), a rectangular plate actuated by a piezoelectric patch was analyzed with various patch sizes. The FEM data was later used to build an ANN model. The dynamic response of the plate was predicted by the ANN model and validated with FEM in terms of 1st order natural frequency and displacement amplitude. Results from case studies showed that with the input of patch length, width and thickness, ANN model can accurately predict both natural frequency and displacement amplitude. When the input of ANN model was simplified to patch size and thickness or the volume of the patch, the accuracy became worse and worse. The influence of the patch size and thickness on the first order natural frequency was coupled and the maximal and minimal values were predicted based on the ANN model.

Machine Learning Algorithms, Applied to Intact Islets of Langerhans, Demonstrate Significantly Enhanced Insulin Staining at the Capillary Interface of Human Pancreatic β Cells

Pancreatic β cells secrete the hormone insulin into the bloodstream and are critical in the control of blood glucose concentrations. β cells are clustered in the micro-organs of the islets of Langerhans, which have a rich capillary network. Recent work has highlighted the intimate spatial connections between β cells and these capillaries, which lead to the targeting of insulin secretion to the region where the β cells contact the capillary basement membrane. In addition, β cells orientate with respect to the capillary contact point and many proteins are differentially distributed at the capillary interface compared with the rest of the cell. Here, we set out to develop an automated image analysis approach to identify individual β cells within intact islets and to determine if the distribution of insulin across the cells was polarised. Our results show that a U-Net machine learning algorithm correctly identified β cells and their orientation with respect to the capillaries. Using this information, we then quantified insulin distribution across the β cells to show enrichment at the capillary interface. We conclude that machine learning is a useful analytical tool to interrogate large image datasets and analyse sub-cellular organisation.

Microscopy deep learning predicts virus infections and reveals mechanics of lytic-infected cells

Imaging across scales reveals disease mechanisms in organisms, tissues, and cells. Yet, particular infection phenotypes, such as virus-induced cell lysis, have remained difficult to study. Here, we developed imaging modalities and deep learning procedures to identify herpesvirus and adenovirus (AdV) infected cells without virus-specific stainings. Fluorescence microscopy of vital DNA-dyes and live-cell imaging revealed learnable virus-specific nuclear patterns transferable to related viruses of the same family. Deep learning predicted two major AdV infection outcomes, non-lytic (nonspreading) and lytic (spreading) infections, up to about 20 hr prior to cell lysis. Using these predictive algorithms, lytic and non-lytic nuclei had the same levels of green fluorescent protein (GFP)-tagged virion proteins but lytic nuclei enriched the virion proteins faster, and collapsed more extensively upon laser-rupture than non-lytic nuclei, revealing impaired mechanical properties of lytic nuclei. Our algorithms may be used to infer infection phenotypes of emerging viruses, enhance single cell biology, and facilitate differential diagnosis of non-lytic and lytic infections.

•

Artificial intelligence identifies HSV- and AdV-infected cells without specific probes.

•

Imaging lytic-infected cells reveals nuclear envelope rupture and AdV dissemination.

•

Live cell imaging and neural networks presciently pinpoint lytic-infected cells.

•

Lytic-infected cell nuclei have mechanical properties distinct from non-lytic nuclei.

Regression plane concept for analysing continuous cellular processes with machine learning

Biological processes are inherently continuous, and the chance of phenotypic discovery is significantly restricted by discretising them. Using multi-parametric active regression we introduce the Regression Plane (RP), a user-friendly discovery tool enabling class-free phenotypic supervised machine learning, to describe and explore biological data in a continuous manner. First, we compare traditional classification with regression in a simulated experimental setup. Second, we use our framework to identify genes involved in regulating triglyceride levels in human cells. Subsequently, we analyse a time-lapse dataset on mitosis to demonstrate that the proposed methodology is capable of modelling complex processes at infinite resolution. Finally, we show that hemocyte differentiation in Drosophila melanogaster has continuous characteristics.

Machine learning-based investigation of the cancer protein secretory pathway

Deregulation of the protein secretory pathway (PSP) is linked to many hallmarks of cancer, such as promoting tissue invasion and modulating cell-cell signaling. The collection of secreted proteins processed by the PSP, known as the secretome, is often studied due to its potential as a reservoir of tumor biomarkers. However, there has been less focus on the protein components of the secretory machinery itself. We therefore investigated the expression changes in secretory pathway components across many different cancer types. Specifically, we implemented a dual approach involving differential expression analysis and machine learning to identify PSP genes whose expression was associated with key tumor characteristics: mutation of p53, cancer status, and tumor stage. Eight different machine learning algorithms were included in the analysis to enable comparison between methods and to focus on signals that were robust to algorithm type. The machine learning approach was validated by identifying PSP genes known to be regulated by p53, and even outperformed the differential expression analysis approach. Among the different analysis methods and cancer types, the kinesin family members KIF20A and KIF23 were consistently among the top genes associated with malignant transformation or tumor stage. However, unlike most cancer types which exhibited elevated KIF20A expression that remained relatively constant across tumor stages, renal carcinomas displayed a more gradual increase that continued with increasing disease severity. Collectively, our study demonstrates the complementary nature of a combined differential expression and machine learning approach for analyzing gene expression data, and highlights key PSP components relevant to features of tumor pathophysiology that may constitute potential therapeutic targets.

The secretory pathway is a series of intracellular compartments and enzymes that process and export proteins from the cell to its surrounding environment. Dysfunction of the secretory pathway is associated with many diseases, including cancer, and therefore constitutes a potential target for novel therapeutic strategies. The large number of interacting components that comprise the secretory pathway pose a challenge when attempting to identify where the dysfunction originates or how to restore healthy function. To improve our understanding of how the secretory pathway is changed within tumors, we used gene expression data from normal tissue and tumor samples from thousands of individuals which included many different types of cancers. The data was analyzed using different machine learning algorithms which we trained to predict sample characteristics, such as disease severity. This training quantified the relative degree to which each gene was associated with the tumor characteristic, allowing us to predict which secretory pathway components were important for processes such as tumor progression—both within specific cancer types and across many different cancer types. The machine learning-based approach demonstrated excellent performance compared to traditional gene expression analysis methods and identified several secretory pathway components with strong evidence of involvement in tumor development.

A guide to accurate reporting in digital image processing - can anyone reproduce your quantitative analysis?

Considerable attention has been recently paid to improving replicability and reproducibility in life science research. This has resulted in commendable efforts to standardize a variety of reagents, assays, cell lines and other resources. However, given that microscopy is a dominant tool for biologists, comparatively little discussion has been offered regarding how the proper reporting and documentation of microscopy relevant details should be handled. Image processing is a critical step of almost any microscopy-based experiment; however, improper, or incomplete reporting of its use in the literature is pervasive. The chosen details of an image processing workflow can dramatically determine the outcome of subsequent analyses, and indeed, the overall conclusions of a study. This Review aims to illustrate how proper reporting of image processing methodology improves scientific reproducibility and strengthens the biological conclusions derived from the results.

Machine learning and applications in microbiology

To understand the intricacies of microorganisms at the molecular level requires making sense of copious volumes of data such that it may now be humanly impossible to detect insightful data patterns without an artificial intelligence application called machine learning. Applying machine learning to address biological problems is expected to grow at an unprecedented rate, yet it is perceived by the uninitiated as a mysterious and daunting entity entrusted to the domain of mathematicians and computer scientists. The aim of this review is to identify key points required to start the journey of becoming an effective machine learning practitioner. These key points are further reinforced with an evaluation of how machine learning has been applied so far in a broad scope of real-life microbiology examples. This includes predicting drug targets or vaccine candidates, diagnosing microorganisms causing infectious diseases, classifying drug resistance against antimicrobial medicines, predicting disease outbreaks and exploring microbial interactions. Our hope is to inspire microbiologists and other related researchers to join the emerging machine learning revolution.

Rapid analysis of streaming platelet images by semi-unsupervised learning

We developed a fast and accurate deep learning approach employing a semi-unsupervised learning system (SULS) for capturing the real-time noisy, sparse, and ambiguous images of platelet activation. Outperforming several leading supervised learning methods when applied to segment various platelet morphologies, the SULS detects their complex boundaries at submicron resolutions and it massively decreases to only a few hours for segmenting streaming images of 45 million platelets that would have taken 40 years to annotate manually. For the first time, the fast dynamics of pseudopod formation and platelet morphological changes including membrane tethers and transient tethering to vessels are accurately captured.

TEMImageNet training library and AtomSegNet deep-learning models for high-precision atom segmentation, localization, denoising, and deblurring of atomic-resolution images

Atom segmentation and localization, noise reduction and deblurring of atomic-resolution scanning transmission electron microscopy (STEM) images with high precision and robustness is a challenging task. Although several conventional algorithms, such has thresholding, edge detection and clustering, can achieve reasonable performance in some predefined sceneries, they tend to fail when interferences from the background are strong and unpredictable. Particularly, for atomic-resolution STEM images, so far there is no well-established algorithm that is robust enough to segment or detect all atomic columns when there is large thickness variation in a recorded image. Herein, we report the development of a training library and a deep learning method that can perform robust and precise atom segmentation, localization, denoising, and super-resolution processing of experimental images. Despite using simulated images as training datasets, the deep-learning model can self-adapt to experimental STEM images and shows outstanding performance in atom detection and localization in challenging contrast conditions and the precision consistently outperforms the state-of-the-art two-dimensional Gaussian fit method. Taking a step further, we have deployed our deep-learning models to a desktop app with a graphical user interface and the app is free and open-source. We have also built a TEM ImageNet project website for easy browsing and downloading of the training data.

Identification of Most Relevant Features for Classification of Francisella tularensis using Machine Learning

Background:

Francisella tularensis is a stealth pathogen fatal for animals and humans. Ease of its propagation, coupled with high capacity for ailment and death makes it a potential candidate for biological weapon.

Objective:

Work related to the pathogen’s classification and factors affecting its prolonged existence in soil is limited to statistical measures. Machine learning other than conventional analysis methods may be applied to better predict epidemiological modeling for this soil-borne pathogen.

Methods:

Feature-ranking algorithms namely; relief, correlation and oneR are used for soil attribute ranking. Moreover, classification algorithms; SVM, random forest, naive bayes, logistic regression and MLP are used for classification of the soil attribute dataset for Francisella tularensis positive and negative soils.

Results:

Feature-ranking methods concluded that clay, nitrogen, organic matter, soluble salts, zinc, silt and nickel are the most significant attributes while potassium, phosphorous, iron, calcium, copper, chromium and sand are the least contributing risk factors for the persistence of the pathogen. However, clay is the most significant and potassium is the least contributing attribute. Data analysis suggests that feature-ranking using relief produced classification accuracy of 84.35% for multilayer perceptron; 82.99% for linear regression; 80.27% for SVM and random forest; and 78.23% for naive bayes, which is better than other ranking methods. MLP outperforms other classifiers by generating an accuracy of 84.35%, 82.99% and 81.63% for feature-ranking using relief, correlation and oneR algorithms, respectively.

Conclusion:

These models can significantly improve accuracy and can minimize the risk of incorrect classification. They further help in controlling epidemics and thereby minimizing the socio-economic impact on the society.

Application of Machine Learning in 3D Bioprinting: Focus on Development of Big Data and Digital Twin

The application of machine learning (ML) in bioprinting has attracted considerable attention recently. Many have focused on the benefits and potential of ML, but a clear overview of how ML shapes the future of three-dimensional (3D) bioprinting is still lacking. Here, it is proposed that two missing links, Big Data and Digital Twin, are the key to articulate the vision of future 3D bioprinting. Creating training databases from Big Data curation and building digital twins of human organs with cellular resolution and properties are the most important and urgent challenges. With these missing links, it is envisioned that future 3D bioprinting will become more digital and in silico, and eventually strike a balance between virtual and physical experiments toward the most efficient utilization of bioprinting resources. Furthermore, the virtual component of bioprinting and biofabrication, namely, digital bioprinting, will become a new growth point for digital industry and information technology in future.

Bioimage Analysis and Cell Motility

Bioimage analysis (BIA) has historically helped study how and why cells move; biological experiments evolved in intimate feedback with the most classical image processing techniques because they contribute objectivity and reproducibility to an eminently qualitative science. Cell segmentation, tracking, and morphology descriptors are all discussed here. Using ameboid motility as a case study, these methods help us illustrate how proper quantification can augment biological data, for example, by choosing mathematical representations that amplify initially subtle differences, by statistically uncovering general laws or by integrating physical insight. More recently, the non-invasive nature of quantitative imaging is fertilizing two blooming fields: mechanobiology, where many biophysical measurements remain inaccessible, and microenvironments, where the quest for physiological relevance has exploded data size. From relief to remedy, this trend indicates that BIA is to become a main vector of biological discovery as human visual analysis struggles against ever more complex data.

Supervised machine learning for automated classification of human Wharton's Jelly cells and mechanosensory hair cells

Tissue engineering and gene therapy strategies offer new ways to repair permanent damage to mechanosensory hair cells (MHCs) by differentiating human Wharton's Jelly cells (HWJCs). Conventionally, these strategies require the classification of each cell as differentiated or undifferentiated. Automated classification tools, however, may serve as a novel method to rapidly classify these cells. In this paper, images from previous work, where HWJCs were differentiated into MHC-like cells, were examined. Various cell features were extracted from these images, and those which were pertinent to classification were identified. Different machine learning models were then developed, some using all extracted data and some using only certain features. To evaluate model performance, the area under the curve (AUC) of the receiver operating characteristic curve was primarily used. This paper found that limiting algorithms to certain features consistently improved performance. The top performing model, a voting classifier model consisting of two logistic regressions, a support vector machine, and a random forest classifier, obtained an AUC of 0.9638. Ultimately, this paper illustrates the viability of a novel machine learning pipeline to automate the classification of undifferentiated and differentiated cells. In the future, this research could aid in automated strategies that determine the viability of MHC-like cells after differentiation.

Predicting Hot Spots Using a Deep Neural Network Approach

Targeting protein-protein interactions is a challenge and crucial task of the drug discovery process. A good starting point for rational drug design is the identification of hot spots (HS) at protein-protein interfaces, typically conserved residues that contribute most significantly to the binding. In this chapter, we depict point-by-point an in-house pipeline used for HS prediction using only sequence-based features from the well-known SpotOn dataset of soluble proteins (Moreira et al., Sci Rep 7:8007, 2017), through the implementation of a deep neural network. The presented pipeline is divided into three steps: (1) feature extraction, (2) deep learning classification, and (3) model evaluation. We present all the available resources, including code snippets, the main dataset, and the free and open-source modules/packages necessary for full replication of the protocol. The users should be able to develop an HS prediction model with accuracy, precision, recall, and AUROC of 0.96, 0.93, 0.91, and 0.86, respectively.

Sensitivity-Enhancing Strategies in Optical Biosensing

High-sensitivity detection of minute quantities or concentration variations of analytes of clinical importance is critical for biosensing to ensure accurate disease diagnostics and reliable health monitoring. A variety of sensitivity-improving concepts have been proposed from chemical, physical, and biological perspectives. In this review, elements that are responsible for sensitivity enhancement are classified and discussed in accordance with their operating steps in a typical biosensing workflow that runs through sampling, analyte recognition, and signal transduction. With a focus on optical biosensing, exemplary sensitivity-improving strategies are introduced, which can be developed into "plug-and-play" modules for many current and future sensors, and discuss their mechanisms to enhance biosensing performance. Three major strategies are covered: i) amplification of signal transduction by polymerization and nanocatalysts, ii) diffusion-limit-breaking systems for enhancing sensor-analyte contact and subsequent analyte recognition by fluid-mixing and analyte-concentrating, and iii) combined approaches that utilize renal concentration at the sampling and recognition steps and chemical signal amplification at the signal transduction step.

Developing a Qualification and Verification Strategy for Digital Tissue Image Analysis in Toxicological Pathology

Digital tissue image analysis is a computational method for analyzing whole-slide images and extracting large, complex, and quantitative data sets. However, as with any analysis method, the quality of generated results is dependent on a well-designed quality control system for the entire digital pathology workflow. Such system requires clear procedural controls, appropriate user training, and involvement of specialists to oversee key steps of the workflow. The toxicologic pathologist is responsible for reporting data obtained by digital image analysis and therefore needs to ensure that it is correct. To accomplish that, they must understand the main parameters of the quality control system and should play an integral part in its conception and implementation. This manuscript describes the most common digital tissue image analysis end points and potential sources of analysis errors. In addition, it outlines recommended approaches for ensuring quality and correctness of results for both classical and machine-learning based image analysis solutions, as adapted from a recently proposed Food and Drug Administration regulatory framework for modifications to artificial intelligence/machine learning-based software as a medical device. These approaches are beneficial for any type of toxicopathologic study which uses the described end points and can be adjusted based on the intended use of the image analysis solution.

Classification of the mechanism of toxicity as applied to human cell line ECV304

The objective of this study was to identify the pattern of cytotoxicity testing of the human cell line ECV304 using three techniques of an ensemble learning algorithm (bagging, boosting and stacking). The study of cell morphology of ECV304 cell line was conducted using impedimetric measurement. Three types of toxins were applied to the ECV304 cell line namely 1 mM hydrogen peroxide (H₂O₂), 5% dimethyl sulfoxide and 10 μg Saponin. The measurement was conducted using electrodes and lock-in amplifier to detect impedance changes during cytotoxicity testing within a frequency range 200 and 830 kHz. The results were analysed, processed and extracted using detrended fluctuation analysis to obtain characteristics and features of the cells when exposed to the each of the toxins. Three ensemble algorithms applied showed slightly different results on the performance for classifying the data set from the feature extraction that was performed. However, the results show that the cell reaction to the toxins could be classified.

A Cell Adhesion-Based Reconstitution Method for Studying Cell Polarity

Cell polarity is an evolutionarily conserved process of asymmetric spatial organization within cells and is essential to tissue structure, signal transduction, cell migration, and cell division. The establishment and maintenance of polarity typically involves extensive protein-protein interactions that can be made further intricate by cell cycle-dependent regulation. These aspects can make interpreting phenotypes within traditional in vivo genetic systems challenging due to pleiotropic effects in loss-of-function experiments. Minimal reconstitution methods offer investigators the advantage of stricter control of otherwise complex systems and allow for more direct assessment of the role of individual components to the process of interest. Here I provide a detailed protocol for a cell adhesion-based method of inducing cell polarity within non-polarized Drosophila S2 cells. This technique is simple, cost effective, moderate throughput, and amenable to RNAi-based loss-of-function studies. The ability to “plug-and-play” genes of interest allows investigators to easily assess the contribution of individual protein domains and post-translational modifications to their function. The system is ideally suited to test not only the requirement of individual components but also their sufficiency, and can provide important insight into the epistatic relationship among multiple components in a protein complex. Although designed for use within Drosophila cells, the general premise and protocol should be easily adapted to mammalian cell culture or other systems that may better suit the interests of potential users.