Image-based profiling for drug discovery: due for a machine-learning upgrade?

Label-free prediction of cell painting from brightfield images

Cell Painting is a high-content image-based assay applied in drug discovery to predict bioactivity, assess toxicity and understand mechanisms of action of chemical and genetic perturbations. We investigate label-free Cell Painting by predicting the five fluorescent Cell Painting channels from brightfield input. We train and validate two deep learning models with a dataset representing 17 batches, and we evaluate on batches treated with compounds from a phenotypic set. The mean Pearson correlation coefficient of the predicted images across all channels is 0.84. Without incorporating features into the model training, we achieved a mean correlation of 0.45 with ground truth features extracted using a segmentation-based feature extraction pipeline. Additionally, we identified 30 features which correlated greater than 0.8 to the ground truth. Toxicity analysis on the label-free Cell Painting resulted a sensitivity of 62.5% and specificity of 99.3% on images from unseen batches. We provide a breakdown of the feature profiles by channel and feature type to understand the potential and limitations of label-free morphological profiling. We demonstrate that label-free Cell Painting has the potential to be used for downstream analyses and could allow for repurposing imaging channels for other non-generic fluorescent stains of more targeted biological interest.

Senescence-associated morphological profiles (SAMPs): an image-based phenotypic profiling method for evaluating the inter and intra model heterogeneity of senescence

Senescence occurs in response to a number of damaging stimuli to limit oncogenic transformation and cancer development. As no single, universal senescence marker has been discovered, the confident classification of senescence induction requires the parallel assessment of a series of hallmarks. Therefore, there is a growing need for “first-pass” tools of senescence identification to streamline experimental workflows and complement conventional markers.

Here, we utilise a high content, multidimensional phenotypic profiling-based approach, to assess the morphological profiles of senescent cells induced via a range of stimuli. In the context of senescence, we refer to these as senescence-associated morphological profiles (SAMPs), as they facilitate distinction between senescent and proliferating cells. The complexity of the profiles generated also allows exploration of the heterogeneity both between models of senescence and within an individual senescence model, providing a level of insight at the single cell level. Furthermore, we also demonstrate that these models are applicable to the assessment of senescence in vivo, which remains a key challenge for the field. Therefore, we believe SAMPs has the potential to serve as a useful addition in the repertoire of senescence researchers, either as a first-pass tool or as part of the established senescence hallmarks.

A multiparametric activity profiling platform for neuron disease phenotyping and drug screening

Patient stem cell-derived models enable imaging of complex disease phenotypes and the development of scalable drug discovery platforms. Current preclinical methods for assessing cellular activity do not, however, capture the full intricacies of disease-induced disturbances and instead typically focus on a single parameter, which impairs both the understanding of disease and the discovery of effective therapeutics. Here, we describe a cloud-based image processing and analysis platform that captures the intricate activity profile revealed by GCaMP fluorescence recordings of intracellular calcium changes and enables the discovery of molecules that correct 153 parameters that define the amyotrophic lateral sclerosis motor neuron disease phenotype. In a high-throughput screen, we identified compounds that revert the multiparametric disease profile to that found in healthy cells, a novel and robust measure of therapeutic potential quite distinct from unidimensional screening. This platform can guide the development of therapeutics that counteract the multifaceted pathological features of diseased cellular activity.

Synergy between machine learning and natural products cheminformatics: Application to the lead discovery of anthraquinone derivatives

Cheminformatics utilizing machine learning (ML) techniques have opened up a new horizon in drug discovery. This is owing to vast chemical space expansion with rocketing numbers of expected hits and lead compounds that match druggable macromolecular targets, in particular from natural compounds. Due to the natural products' (NP) structural complexity, uniqueness, and diversity, they could occupy a bigger space in pharmaceuticals, allowing the industry to pursue more selective leads in the nanomolar range of binding affinity. ML is an essential part of each step of the drug design pipeline, such as target prediction, compound library preparation, and lead optimization. Notably, molecular mechanic and dynamic simulations, induced docking, and free energy perturbations are essential in predicting best binding poses, binding free energy values, and molecular mechanics force fields. Those applications have leveraged from artificial intelligence (AI), which decreases the computational costs required for such costly simulations. This review aimed to describe chemical space and compound libraries related to NPs. High-throughput screening utilized for fractionating NPs and high-throughput virtual screening and their strategies, and significance, are reviewed. Particular emphasis was given to AI approaches, ML tools, algorithms, and techniques, especially in drug discovery of macrocyclic compounds and approaches in computer-aided and ML-based drug discovery. Anthraquinone derivatives were discussed as a source of new lead compounds that can be developed using ML tools for diverse medicinal uses such as cancer, infectious diseases, and metabolic disorders. Furthermore, the power of principal component analysis in understanding relevant protein conformations, and molecular modeling of protein-ligand interaction were also presented. Apart from being a concise reference for cheminformatics, this review is a useful text to understand the application of ML-based algorithms to molecular dynamics simulation and in silico absorption, distribution, metabolism, excretion, and toxicity prediction.

Integrating deep learning and unbiased automated high-content screening to identify complex disease signatures in human fibroblasts

Drug discovery for diseases such as Parkinson's disease are impeded by the lack of screenable cellular phenotypes. We present an unbiased phenotypic profiling platform that combines automated cell culture, high-content imaging, Cell Painting, and deep learning. We applied this platform to primary fibroblasts from 91 Parkinson's disease patients and matched healthy controls, creating the largest publicly available Cell Painting image dataset to date at 48 terabytes. We use fixed weights from a convolutional deep neural network trained on ImageNet to generate deep embeddings from each image and train machine learning models to detect morphological disease phenotypes. Our platform's robustness and sensitivity allow the detection of individual-specific variation with high fidelity across batches and plate layouts. Lastly, our models confidently separate LRRK2 and sporadic Parkinson's disease lines from healthy controls (receiver operating characteristic area under curve 0.79 (0.08 standard deviation)), supporting the capacity of this platform for complex disease modeling and drug screening applications.

A multiplexed epitope barcoding strategy that enables dynamic cellular phenotypic screens

Pooled genetic libraries have improved screening throughput for mapping genotypes to phenotypes. However, selectable phenotypes are limited, restricting screening to outcomes with a low spatiotemporal resolution. Here, we integrated live-cell imaging with pooled library-based screening. To enable intracellular multiplexing, we developed a method called EPICode that uses a combination of short epitopes, which can also appear in various subcellular locations. EPICode thus enables the use of live-cell microscopy to characterize a phenotype of interest over time, including after sequential stimulatory/inhibitory manipulations, and directly connects behavior to the cellular genotype. To test EPICode's capacity against an important milestone-engineering and optimizing dynamic, live-cell reporters-we developed a live-cell PKA kinase translocation reporter with improved sensitivity and specificity. The use of epitopes as fluorescent barcodes introduces a scalable strategy for high-throughput screening broadly applicable to protein engineering and drug discovery settings where image-based phenotyping is desired.

OpenCell: Endogenous tagging for the cartography of human cellular organization

Elucidating the wiring diagram of the human cell is a central goal of the postgenomic era. We combined genome engineering, confocal live-cell imaging, mass spectrometry, and data science to systematically map the localization and interactions of human proteins. Our approach provides a data-driven description of the molecular and spatial networks that organize the proteome. Unsupervised clustering of these networks delineates functional communities that facilitate biological discovery. We found that remarkably precise functional information can be derived from protein localization patterns, which often contain enough information to identify molecular interactions, and that RNA binding proteins form a specific subgroup defined by unique interaction and localization properties. Paired with a fully interactive website (opencell.czbiohub.org), our work constitutes a resource for the quantitative cartography of human cellular organization.

BioProfiling.jl: profiling biological perturbations with high-content imaging in single cells and heterogeneous populations

MOTIVATION

High-content imaging screens provide a cost-effective and scalable way to assess cell states across diverse experimental conditions. The analysis of the acquired microscopy images involves assembling and curating raw cellular measurements into morphological profiles suitable for testing biological hypotheses. Despite being a critical step, general-purpose and adaptable tools for morphological profiling are lacking and no solution is available for the high-performance Julia programming language.

RESULTS

Here, we introduce BioProfiling.jl, an efficient end-to-end solution for compiling and filtering informative morphological profiles in Julia. The package contains all the necessary data structures to curate morphological measurements and helper functions to transform, normalize and visualize profiles. Robust statistical distances and permutation tests enable quantification of the significance of the observed changes despite the high fraction of outliers inherent to high-content screens. This package also simplifies visual artifact diagnostics, thus streamlining a bottleneck of morphological analyses. We showcase the features of the package by analyzing a chemical imaging screen, in which the morphological profiles prove to be informative about the compounds' mechanisms of action and can be conveniently integrated with the network localization of molecular targets.

AVAILABILITY AND IMPLEMENTATION

The Julia package is available on GitHub: https://github.com/menchelab/BioProfiling.jl. We also provide Jupyter notebooks reproducing our analyses: https://github.com/menchelab/BioProfilingNotebooks. The data underlying this article are available from FigShare, at https://doi.org/10.6084/m9.figshare.14784678.v2.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Image-Based Annotation of Chemogenomic Libraries for Phenotypic Screening

Phenotypical screening is a widely used approach in drug discovery for the identification of small molecules with cellular activities. However, functional annotation of identified hits often poses a challenge. The development of small molecules with narrow or exclusive target selectivity such as chemical probes and chemogenomic (CG) libraries, greatly diminishes this challenge, but non-specific effects caused by compound toxicity or interference with basic cellular functions still pose a problem to associate phenotypic readouts with molecular targets. Hence, each compound should ideally be comprehensively characterized regarding its effects on general cell functions. Here, we report an optimized live-cell multiplexed assay that classifies cells based on nuclear morphology, presenting an excellent indicator for cellular responses such as early apoptosis and necrosis. This basic readout in combination with the detection of other general cell damaging activities of small molecules such as changes in cytoskeletal morphology, cell cycle and mitochondrial health provides a comprehensive time-dependent characterization of the effect of small molecules on cellular health in a single experiment. The developed high-content assay offers multi-dimensional comprehensive characterization that can be used to delineate generic effects regarding cell functions and cell viability, allowing an assessment of compound suitability for subsequent detailed phenotypic and mechanistic studies.

Computational analyses of mechanism of action (MoA): data, methods and integration

The elucidation of a compound's Mechanism of Action (MoA) is a challenging task in the drug discovery process, but it is important in order to rationalise phenotypic findings and to anticipate potential side-effects. Bioinformatic approaches, advances in machine learning techniques and the increasing deposition of high-throughput data in public databases have significantly contributed to recent advances in the field, but it is not straightforward to decide which data and methods are most suitable to use in a given case. In this review, we focus on these methods and data and their applications in generating MoA hypotheses for subsequent experimental validation. We discuss compound-specific data such as -omics, cell morphology and bioactivity data, as well as commonly used supplementary prior knowledge such as network and pathway data, and provide information on databases where this data can be accessed. In terms of methodologies, we discuss both well-established methods (connectivity mapping, pathway enrichment) as well as more developing methods (neural networks and multi-omics integration). Finally, we review case studies where the MoA of a compound was successfully suggested from computational analysis by incorporating multiple data modalities and/or methodologies. Our aim for this review is to provide researchers with insights into the benefits and drawbacks of both the data and methods in terms of level of understanding, biases and interpretation - and to highlight future avenues of investigation which we foresee will improve the field of MoA elucidation, including greater public access to -omics data and methodologies which are capable of data integration.

Predicting drug polypharmacology from cell morphology readouts using variational autoencoder latent space arithmetic

A variational autoencoder (VAE) is a machine learning algorithm, useful for generating a compressed and interpretable latent space. These representations have been generated from various biomedical data types and can be used to produce realistic-looking simulated data. However, standard vanilla VAEs suffer from entangled and uninformative latent spaces, which can be mitigated using other types of VAEs such as β-VAE and MMD-VAE. In this project, we evaluated the ability of VAEs to learn cell morphology characteristics derived from cell images. We trained and evaluated these three VAE variants-Vanilla VAE, β-VAE, and MMD-VAE-on cell morphology readouts and explored the generative capacity of each model to predict compound polypharmacology (the interactions of a drug with more than one target) using an approach called latent space arithmetic (LSA). To test the generalizability of the strategy, we also trained these VAEs using gene expression data of the same compound perturbations and found that gene expression provides complementary information. We found that the β-VAE and MMD-VAE disentangle morphology signals and reveal a more interpretable latent space. We reliably simulated morphology and gene expression readouts from certain compounds thereby predicting cell states perturbed with compounds of known polypharmacology. Inferring cell state for specific drug mechanisms could aid researchers in developing and identifying targeted therapeutics and categorizing off-target effects in the future.

Machine Learning Enables Accurate and Rapid Prediction of Active Molecules Against Breast Cancer Cells

Breast cancer (BC) has surpassed lung cancer as the most frequently occurring cancer, and it is the leading cause of cancer-related death in women. Therefore, there is an urgent need to discover or design new drug candidates for BC treatment. In this study, we first collected a series of structurally diverse datasets consisting of 33,757 active and 21,152 inactive compounds for 13 breast cancer cell lines and one normal breast cell line commonly used in in vitro antiproliferative assays. Predictive models were then developed using five conventional machine learning algorithms, including naïve Bayesian, support vector machine, k-Nearest Neighbors, random forest, and extreme gradient boosting, as well as five deep learning algorithms, including deep neural networks, graph convolutional networks, graph attention network, message passing neural networks, and Attentive FP. A total of 476 single models and 112 fusion models were constructed based on three types of molecular representations including molecular descriptors, fingerprints, and graphs. The evaluation results demonstrate that the best model for each BC cell subtype can achieve high predictive accuracy for the test sets with AUC values of 0.689–0.993. Moreover, important structural fragments related to BC cell inhibition were identified and interpreted. To facilitate the use of the model, an online webserver called ChemBC (http://chembc.idruglab.cn/) and its local version software (https://github.com/idruglab/ChemBC) were developed to predict whether compounds have potential inhibitory activity against BC cells.

Tools for Decoding Ubiquitin Signaling in DNA Repair

The maintenance of genome stability requires dedicated DNA repair processes and pathways that are essential for the faithful duplication and propagation of chromosomes. These DNA repair mechanisms counteract the potentially deleterious impact of the frequent genotoxic challenges faced by cells from both exogenous and endogenous agents. Intrinsic to these mechanisms, cells have an arsenal of protein factors that can be utilised to promote repair processes in response to DNA lesions. Orchestration of the protein factors within the various cellular DNA repair pathways is performed, in part, by post-translational modifications, such as phosphorylation, ubiquitin, SUMO and other ubiquitin-like modifiers (UBLs). In this review, we firstly explore recent advances in the tools for identifying factors involved in both DNA repair and ubiquitin signaling pathways. We then expand on this by evaluating the growing repertoire of proteomic, biochemical and structural techniques available to further understand the mechanistic basis by which these complex modifications regulate DNA repair. Together, we provide a snapshot of the range of methods now available to investigate and decode how ubiquitin signaling can promote DNA repair and maintain genome stability in mammalian cells.

Development of a chemogenomics library for phenotypic screening

With the development of advanced technologies in cell-based phenotypic screening, phenotypic drug discovery (PDD) strategies have re-emerged as promising approaches in the identification and development of novel and safe drugs. However, phenotypic screening does not rely on knowledge of specific drug targets and needs to be combined with chemical biology approaches to identify therapeutic targets and mechanisms of actions induced by drugs and associated with an observable phenotype. In this study, we developed a system pharmacology network integrating drug-target-pathway-disease relationships as well as morphological profile from an existing high content imaging-based high-throughput phenotypic profiling assay known as “Cell Painting”. Furthermore, from this network, a chemogenomic library of 5000 small molecules that represent a large and diverse panel of drug targets involved in diverse biological effects and diseases has been developed. Such a platform and a chemogenomic library could assist in the target identification and mechanism deconvolution of some phenotypic assays. The usefulness of the platform is illustrated through examples.

Plating human iPSC lines on micropatterned substrates reveals role for ITGB1 nsSNV in endoderm formation

Quantitative analysis of human induced pluripotent stem cell (iPSC) lines from healthy donors is a powerful tool for uncovering the relationship between genetic variants and cellular behavior. We previously identified rare, deleterious non-synonymous single nucleotide variants (nsSNVs) in cell adhesion genes that are associated with outlier iPSC phenotypes in the pluripotent state. Here, we generated micropatterned colonies of iPSCs to test whether nsSNVs influence patterning of radially ordered germ layers. Using a custom-built image analysis pipeline, we quantified the differentiation phenotypes of 13 iPSC lines that harbor nsSNVs in genes related to cell adhesion or germ layer development. All iPSC lines differentiated into the three germ layers; however, there was donor-specific variation in germ layer patterning. We identified one line that presented an outlier phenotype of expanded endodermal differentiation, which was associated with a nsSNV in ITGB1. Our study establishes a platform for investigating the impact of nsSNVs on differentiation.

High-throughput screening for drug discovery targeting the cancer cell-microenvironment interactions in hematological cancers

INTRODUCTION

The interactions between leukemic blasts and cells within the bone marrow environment affect oncogenesis, cancer stem cell survival, as well as drug resistance in hematological cancers. The importance of this interaction is increasingly being recognized as a potentially important target for future drug discoveries and developments. Recent innovations in the high throughput drug screening-related technologies, novel ex-vivo disease-models, and freely available machine-learning algorithms are advancing the drug discovery process by targeting earlier undruggable proteins, complex pathways, as well as physical interactions (e.g. leukemic cell-bone microenvironment interaction).

AREA COVERED

In this review, the authors discuss the recent methodological advancements and existing challenges to target specialized hematopoietic niches within the bone marrow during leukemia and suggest how such methods can be used to identify drugs targeting leukemic cell-bone microenvironment interactions.

EXPERT OPINION

The recent development in cell-cell communication scoring technology and culture conditions can speed up the drug discovery by targeting the cell-microenvironment interaction. However, to accelerate this process, collecting clinical-relevant patient tissues, developing culture model systems, and implementing computational algorithms, especially trained to predict drugs and their combination targeting the cancer cell-bone microenvironment interaction are needed.

Artificial Intelligence and Quantum Computing as the Next Pharma Disruptors

Artificial intelligence (AI) consists of a synergistic assembly of enhanced optimization strategies with wide application in drug discovery and development, providing advanced tools for promoting cost-effectiveness throughout drug life cycle. Specifically, AI brings together the potential to improve drug approval rates, reduce development costs, get medications to patients faster, and help patients complying with their treatments. Accelerated pharmaceutical development and drug product approval rates can further benefit from the quantum computing (QC) technology, which will ultimately enable larger profits from patent-protected market exclusivity.Key pharma stakeholders are endorsing cutting-edge technologies based on AI and QC , covering drug discovery, preclinical and clinical development, and postapproval activities. Indeed, AI-QC applications are expected to become standard in the pharma operating model over the next 5-10 years. Generalizing scalability to larger pharmaceutical problems instead of specialization is now the main principle for transforming pharmaceutical tasks on multiple fronts, for which systematic and cost-effective solutions have benefited in areas such as molecular screening, synthetic pathway design, and drug discovery and development.The information generated by coupling the life cycle of drugs and AI and/or QC through data-driven analysis, neural network prediction, and chemical system monitoring will enable (1) better understanding of the complexity of process data, (2) streamlining the design of experiments, (3) discovering new molecular targets and materials, and also (4) planning or rethinking upcoming pharmaceutical challenges The power of AI-QC makes accessible a range of different pharmaceutical problems and their rationalization that have not been previously addressed due to a lack of appropriate analytical tools, demonstrating the breadth of potential applications of these emerging multidimensional approaches. In this context, creating the right AI-QC strategy often involves a steep learning path, especially given the embryonic stage of the industry development and the relative lack of case studies documenting success. As such, a comprehensive knowledge of the underlying pillars is imperative to extend the landscape of applications across the drug life cycle.The topics enclosed in this chapter will focus on AI-QC methods applied to drug discovery and development, with emphasis on the most recent advances in this field.

Success stories of AI in drug discovery - where do things stand?

INTRODUCTION

Artificial intelligence (AI) in drug discovery and development (DDD) has gained more traction in the past few years. Many scientific reviews have already been made available in this area. Thus, in this review, the authors have focused on the success stories of AI-driven drug candidates and the scientometric analysis of the literature in this field.

AREA COVERED

The authors explore the literature to compile the success stories of AI-driven drug candidates that are currently being assessed in clinical trials or have investigational new drug (IND) status. The authors also provide the reader with their expert perspectives for future developments and their opinions on the field.

EXPERT OPINION

Partnerships between AI companies and the pharma industry are booming. The early signs of the impact of AI on DDD are encouraging, and the pharma industry is hoping for breakthroughs. AI can be a promising technology to unveil the greatest successes, but it has yet to be proven as AI is still at the embryonic stage.

Applying Machine Learning to Stem Cell Culture and Differentiation

Machine learning techniques are increasingly becoming incorporated into biological research workflows in a variety of disciplines, most notably cancer research and drug discovery. Efforts in stem cell research comparatively lag behind. We detail key paradigms in machine learning, with a focus on equipping stem cell biologists with the understanding necessary to begin conceptualizing and designing machine learning workflows within their own domain of expertise. Supervised approaches in both regression and classification as well as unsupervised clustering techniques are all covered, with examples from across the biological sciences. High-throughput, high-content, multiplex assays for data acquisition are also discussed in the form of single-cell RNA sequencing and image-based approaches. Lastly, potential applications in stem cell biology, including the development of novel cell types, and improving model maturation are also discussed. Machine learning approaches applied in stem cell biology show promise in accelerating progress in developmental biology, drug screening, disease modeling, and personalized medicine. © 2021 Wiley Periodicals LLC.

Combining Automated Organoid Workflows With Artificial Intelligence-Based Analyses: Opportunities to Build a New Generation of Interdisciplinary High-Throughput Screens for Parkinson's Disease and Beyond

Parkinson's disease (PD) is the second most common neurodegenerative disease and primarily characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta of the midbrain. Despite decades of research and the development of various disease model systems, there is no curative treatment. This could be due to current model systems, including cell culture and animal models, not adequately recapitulating human PD etiology. More complex human disease models, including human midbrain organoids, are maturing technologies that increasingly enable the strategic incorporation of the missing components needed to model PD in vitro. The resulting organoid-based biological complexity provides new opportunities and challenges in data analysis of rich multimodal data sets. Emerging artificial intelligence (AI) capabilities can take advantage of large, broad data sets and even correlate results across disciplines. Current organoid technologies no longer lack the prerequisites for large-scale high-throughput screening (HTS) and can generate complex yet reproducible data suitable for AI-based data mining. We have recently developed a fully scalable and HTS-compatible workflow for the generation, maintenance, and analysis of three-dimensional (3D) microtissues mimicking key characteristics of the human midbrain (called "automated midbrain organoids," AMOs). AMOs build a reproducible, scalable foundation for creating next-generation 3D models of human neural disease that can fuel mechanism-agnostic phenotypic drug discovery in human in vitro PD models and beyond. Here, we explore the opportunities and challenges resulting from the convergence of organoid HTS and AI-driven data analytics and outline potential future avenues toward the discovery of novel mechanisms and drugs in PD research. © 2021 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Interpretable deep learning uncovers cellular properties in label-free live cell images that are predictive of highly metastatic melanoma

Deep learning has emerged as the technique of choice for identifying hidden patterns in cell imaging data but is often criticized as "black box." Here, we employ a generative neural network in combination with supervised machine learning to classify patient-derived melanoma xenografts as "efficient" or "inefficient" metastatic, validate predictions regarding melanoma cell lines with unknown metastatic efficiency in mouse xenografts, and use the network to generate in silico cell images that amplify the critical predictive cell properties. These exaggerated images unveiled pseudopodial extensions and increased light scattering as hallmark properties of metastatic cells. We validated this interpretation using live cells spontaneously transitioning between states indicative of low and high metastatic efficiency. This study illustrates how the application of artificial intelligence can support the identification of cellular properties that are predictive of complex phenotypes and integrated cell functions but are too subtle to be identified in the raw imagery by a human expert. A record of this paper's transparent peer review process is included in the supplemental information. VIDEO ABSTRACT.

High-content approaches to anthelmintic drug screening

Most anthelmintics were discovered through in vivo screens using animal models of infection. Developing in vitro assays for parasitic worms presents several challenges. The lack of in vitro life cycle culture protocols requires harvesting worms from vertebrate hosts or vectors, limiting assay throughput. Once worms are removed from the host environment, established anthelmintics often show no obvious phenotype - raising concerns about the predictive value of many in vitro assays. However, with recent progress in understanding how anthelmintics subvert host-parasite interactions, and breakthroughs in high-content imaging and machine learning, in vitro assays have the potential to discern subtle cryptic parasite phenotypes. These may prove better endpoints than conventional in vitro viability assays.

Machine learning for perturbational single-cell omics

Cell biology is fundamentally limited in its ability to collect complete data on cellular phenotypes and the wide range of responses to perturbation. Areas such as computer vision and speech recognition have addressed this problem of characterizing unseen or unlabeled conditions with the combined advances of big data, deep learning, and computing resources in the past 5 years. Similarly, recent advances in machine learning approaches enabled by single-cell data start to address prediction tasks in perturbation response modeling. We first define objectives in learning perturbation response in single-cell omics; survey existing approaches, resources, and datasets (https://github.com/theislab/sc-pert); and discuss how a perturbation atlas can enable deep learning models to construct an informative perturbation latent space. We then examine future avenues toward more powerful and explainable modeling using deep neural networks, which enable the integration of disparate information sources and an understanding of heterogeneous, complex, and unseen systems.

High Content Image Analysis of Spatiotemporal Proliferation and Differentiation Patterns in 3D Embryoid Body Differentiation Model

The 3D embryoid body (EB) differentiation model is a promising tool for fundamental cell biology and drug discovery studies assessing the compound effects on mammalian and human development. This 3D cell model allows for analyzing spatiotemporal changes during morphogenesis and differentiation. A combination of confocal microscopy with high content image analysis (HCIA) can significantly improve the study of spatiotemporal patterns of early embryonic lineages and compound efficacy and toxicity testing by enhancing the identification and quantification of various cell types. HCIA can be used to assess the EB architecture through quantitative and qualitative characteristics, such as viability and apoptosis, identification, localization, ratio and timing for various types of early embryonic cells, dimensions of compartments of proliferating and differentiating cells, changes in the size and shape of EBs, and translocation of individual cells and cell layers. This chapter describes a comprehensive framework for HCIA for 3D EB differentiation model that allows investigators to analyze EB growth, differentiation, and morphogenetic dynamics.

Aquaglyceroporin-3's Expression and Cellular Localization Is Differentially Modulated by Hypoxia in Prostate Cancer Cell Lines

Aquaporins are required by cells to enable fast adaptation to volume and osmotic changes, as well as microenvironmental metabolic stimuli. Aquaglyceroporins play a crucial role in supplying cancer cells with glycerol for metabolic needs. Here, we show that AQP3 is differentially expressed in cells of a prostate cancer panel. AQP3 is located at the cell membrane and cytoplasm of LNCaP cell while being exclusively expressed in the cytoplasm of Du145 and PC3 cells. LNCaP cells show enhanced hypoxia growth; Du145 and PC3 cells display stress factors, indicating a crucial role for AQP3 at the plasma membrane in adaptation to hypoxia. Hypoxia, both acute and chronic affected AQP3′s cellular localization. These outcomes were validated using a machine learning classification approach of the three cell lines and of the six normoxic or hypoxic conditions. Classifiers trained on morphological features derived from cytoskeletal and nuclear labeling alongside corresponding texture features could uniquely identify each individual cell line and the corresponding hypoxia exposure. Cytoskeletal features were 70–90% accurate, while nuclear features allowed for 55–70% accuracy. Cellular texture features (73.9% accuracy) were a stronger predictor of the hypoxic load than the AQP3 distribution (60.3%).

Data science in cell imaging

Cell imaging has entered the 'Big Data' era. New technologies in light microscopy and molecular biology have led to an explosion in high-content, dynamic and multidimensional imaging data. Similar to the 'omics' fields two decades ago, our current ability to process, visualize, integrate and mine this new generation of cell imaging data is becoming a critical bottleneck in advancing cell biology. Computation, traditionally used to quantitatively test specific hypotheses, must now also enable iterative hypothesis generation and testing by deciphering hidden biologically meaningful patterns in complex, dynamic or high-dimensional cell image data. Data science is uniquely positioned to aid in this process. In this Perspective, we survey the rapidly expanding new field of data science in cell imaging. Specifically, we highlight how data science tools are used within current image analysis pipelines, propose a computation-first approach to derive new hypotheses from cell image data, identify challenges and describe the next frontiers where we believe data science will make an impact. We also outline steps to ensure broad access to these powerful tools - democratizing infrastructure availability, developing sensitive, robust and usable tools, and promoting interdisciplinary training to both familiarize biologists with data science and expose data scientists to cell imaging.

Histopathological and Immune Prognostic Factors in Colo-Rectal Liver Metastases

Simple Summary

Clinical management of colo-rectal liver metastasis would benefit from a refined stratification of patients in prognostic groups, in order to identify the best therapeutic option. Efforts are ongoing in the definition of parameters associated with clinical behaviors, which could help classifying patients in clinically relevant groups. Here we aimed at discussing the recent advances in this field, and we introduced current and new promising candidates, such as morphological tumor features and immune components, which have been showing significant association with survival. Some of these parameters are slowly reaching the clinic and further efforts are ongoing in the attempt to combine them in multiparametric scores.

Abstract

Prognostic studies are increasingly providing new tools to stratify colo-rectal liver metastasis patients into clinical subgroups, with remarkable implications in terms of clinical management and therapeutic choice. Here, the strengths and hurdles of current prognostic tools in colo-rectal liver metastasis are discussed. Alongside more classic histopathological parameters, which capture features related to the tumor component, such as tumor invasion, tumor growth pattern and regression score, we will discuss immune mediators, which are starting to be considered important features. Their objective quantification has shown significant results in prognostication studies, with most of the work focused on adaptive immune cells, namely T cells. As for macrophages, they are only starting to be appreciated and we will present recent advances in evaluation of macrophage morphological features. Deeper knowledge acquired by multiparametric analyses is rapidly uncovering the variety of immune players that should be assessed. The future projection is to implement deep-learning histopathological tools and to integrate histopathological and immune metrics in multiparametric scores, with the ultimate objective to achieve a deeper resolution of the tumor features and their relevance for colo-rectal liver metastasis.

Harnessing machine learning for development of microbiome therapeutics

The last twenty years of seminal microbiome research has uncovered microbiota's intrinsic relationship with human health. Studies elucidating the relationship between an unbalanced microbiome and disease are currently published daily. As such, microbiome big data have become a reality that provide a mine of information for the development of new therapeutics. Machine learning (ML), a branch of artificial intelligence, offers powerful techniques for big data analysis and prediction-making, that are out of reach of human intellect alone. This review will explore how ML can be applied for the development of microbiome-targeted therapeutics. A background on ML will be given, followed by a guide on where to find reliable microbiome big data. Existing applications and opportunities will be discussed, including the use of ML to discover, design, and characterize microbiome therapeutics. The use of ML to optimize advanced processes, such as 3D printing and in silico prediction of drug-microbiome interactions, will also be highlighted. Finally, barriers to adoption of ML in academic and industrial settings will be examined, concluded by a future outlook for the field.