Three million images and morphological profiles of cells treated with matched chemical and genetic perturbations.

The identification of genetic and chemical perturbations with similar impacts on cell morphology can elucidate compounds' mechanisms of action or novel regulators of genetic pathways. Research on methods for identifying such similarities has lagged due to a lack of carefully designed and well-annotated image sets of cells treated with chemical and genetic perturbations. Here we create such a Resource dataset, CPJUMP1, in which each perturbed gene's product is a known target of at least two chemical compounds in the dataset. We systematically explore the directionality of correlations among perturbations that target the same protein encoded by a given gene, and we find that identifying matches between chemical and genetic perturbations is a challenging task. Our dataset and baseline analyses provide a benchmark for evaluating methods that measure perturbation similarities and impact, and more generally, learn effective representations of cellular state from microscopy images. Such advancements would accelerate the applications of image-based profiling of cellular states, such as uncovering drug mode of action or probing functional genomics.

A versatile information retrieval framework for evaluating profile strength and similarity.

In profiling assays, thousands of biological properties are measured in a single test, yielding biological discoveries by capturing the state of a cell population, often at the single-cell level. However, for profiling datasets, it has been challenging to evaluate the phenotypic activity of a sample and the phenotypic consistency among samples, due to profiles' high dimensionality, heterogeneous nature, and non-linear properties. Existing methods leave researchers uncertain where to draw boundaries between meaningful biological response and technical noise. Here, we developed a statistical framework that uses the well-established mean average precision (mAP) as a single, data-driven metric to bridge this gap. We validated the mAP framework against established metrics through simulations and real-world data applications, revealing its ability to capture subtle and meaningful biological differences in cell state. Specifically, we used mAP to assess both phenotypic activity for a given perturbation (or a sample) as well as consistency within groups of perturbations (or samples) across diverse high-dimensional datasets. We evaluated the framework on different profile types (image, protein, and mRNA profiles), perturbation types (CRISPR gene editing, gene overexpression, and small molecules), and profile resolutions (single-cell and bulk). Our open-source software allows this framework to be applied to identify interesting biological phenomena and promising therapeutics from large-scale profiling data.

Reproducible image-based profiling with Pycytominer.

Technological advances in high-throughput microscopy have facilitated the acquisition of cell images at a rapid pace, and data pipelines can now extract and process thousands of image-based features from microscopy images. These features represent valuable single-cell phenotypes that contain information about cell state and biological processes. The use of these features for biological discovery is known as image-based or morphological profiling. However, these raw features need processing before use and image-based profiling lacks scalable and reproducible open-source software. Inconsistent processing across studies makes it difficult to compare datasets and processing steps, further delaying the development of optimal pipelines, methods, and analyses. To address these issues, we present Pycytominer, an open-source software package with a vibrant community that establishes an image-based profiling standard. Pycytominer has a simple, user-friendly Application Programming Interface (API) that implements image-based profiling functions for processing high-dimensional morphological features extracted from microscopy images of cells. Establishing Pycytominer as a standard image-based profiling toolkit ensures consistent data processing pipelines with data provenance, therefore minimizing potential inconsistencies and enabling researchers to confidently derive accurate conclusions and discover novel insights from their data, thus driving progress in our field.

Integrating inflammatory biomarker analysis and artificial-intelligence-enabled image-based profiling to identify drug targets for intestinal fibrosis.

Intestinal fibrosis, often caused by inflammatory bowel disease, can lead to intestinal stenosis and obstruction, but there are no approved treatments. Drug discovery has been hindered by the lack of screenable cellular phenotypes. To address this, we used a scalable image-based morphology assay called Cell Painting, augmented with machine learning algorithms, to identify small molecules that could reverse the activated fibrotic phenotype of intestinal myofibroblasts. We then conducted a high-throughput small molecule chemogenomics screen of approximately 5,000 compounds with known targets or mechanisms, which have achieved clinical stage or approval by the FDA. By integrating morphological analyses and AI using pathologically relevant cells and disease-relevant stimuli, we identified several compounds and target classes that are potentially able to treat intestinal fibrosis. This phenotypic screening platform offers significant improvements over conventional methods for identifying a wide range of drug targets.

SOCRAT: a Dynamic Web Toolbox for Interactive Data Processing, Analysis and Visualization.

Many systems for exploratory and visual data analytics require platform-dependent software installation, coding skills, and analytical expertise. The rapid advances in data-acquisition, web-based information, and communication and computation technologies promoted the explosive growth of online services and tools implementing novel solutions for interactive data exploration and visualization. However, web-based solutions for visual analytics remain scattered and relatively problem-specific. This leads to per-case re-implementations of common components, system architectures, and user interfaces, rather than focusing on innovation and building sophisticated applications for visual analytics. In this paper, we present the Statistics Online Computational Resource Analytical Toolbox (SOCRAT), a dynamic, flexible, and extensible web-based visual analytics framework. The SOCRAT platform is designed and implemented using multi-level modularity and declarative specifications. This enables easy integration of a number of components for data management, analysis, and visualization. SOCRAT benefits from the diverse landscape of existing in-browser solutions by combining them with flexible template modules into a unique, powerful, and feature-rich visual analytics toolbox. The platform integrates a number of independently developed tools for data import, display, storage, interactive visualization, statistical analysis, and machine learning. Various use cases demonstrate the unique features of SOCRAT for visual and statistical analysis of heterogeneous types of data.

3D convolutional neural networks for stalled brain capillary detection.

Adequate blood supply is critical for normal brain function. Brain vasculature dysfunctions, including stalled blood flow in cerebral capillaries, are associated with cognitive decline and pathogenesis in Alzheimer's disease. Recent advances in imaging technology enabled generation of high-quality 3D images that can be used to visualize stalled blood vessels. However, localization of stalled vessels in 3D images is often required as the first step for downstream analysis. When performed manually, this process is tedious, time-consuming, and error-prone. Here, we describe a deep learning-based approach for automatic detection of stalled capillaries in brain images based on 3D convolutional neural networks. Our approach includes custom 3D data augmentations and a weights transfer method that re-uses weights from 2D models pre-trained on natural images for initialization of 3D networks. We used an ensemble of several 3D models to produce the winning submission to the "Clog Loss: Advance Alzheimer's Research with Stall Catchers" machine learning competition that challenged the participants with classifying blood vessels in 3D image stacks as stalled or flowing. In this setting, our approach outperformed other methods and demonstrated state-of-the-art results, achieving 85% Matthews correlation coefficient, 85% sensitivity, and 99.3% specificity. The source code for our solution is publicly available.

Valproic acid-induced changes of 4D nuclear morphology in astrocyte cells.

Histone deacetylase inhibitors, such as valproic acid (VPA), have important clinical therapeutic and cellular reprogramming applications. They induce chromatin reorganization that is associated with altered cellular morphology. However, there is a lack of comprehensive characterization of VPA-induced changes of nuclear size and shape. Here, we quantify 3D nuclear morphology of primary human astrocyte cells treated with VPA over time (hence, 4D). We compared volumetric and surface-based representations and identified seven features that jointly discriminate between normal and treated cells with 85% accuracy on day 7. From day 3, treated nuclei were more elongated and flattened and then continued to morphologically diverge from controls over time, becoming larger and more irregular. On day 7, most of the size and shape descriptors demonstrated significant differences between treated and untreated cells, including a 24% increase in volume and 6% reduction in extent (shape regularity) for treated nuclei. Overall, we show that 4D morphometry can capture how chromatin reorganization modulates the size and shape of the nucleus over time. These nuclear structural alterations may serve as a biomarker for histone (de-)acetylation events and provide insights into mechanisms of astrocytes-to-neurons reprogramming.

Model-Based and Model-Free Techniques for Amyotrophic Lateral Sclerosis Diagnostic Prediction and Patient Clustering.

Amyotrophic lateral sclerosis (ALS) is a complex progressive neurodegenerative disorder with an estimated prevalence of about 5 per 100,000 people in the United States. In this study, the ALS disease progression is measured by the change of Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS) score over time. The study aims to provide clinical decision support for timely forecasting of the ALS trajectory as well as accurate and reproducible computable phenotypic clustering of participants. Patient data are extracted from DREAM-Phil Bowen ALS Prediction Prize4Life Challenge data, most of which are from the Pooled Resource Open-Access ALS Clinical Trials Database (PRO-ACT) archive. We employed model-based and model-free machine-learning methods to predict the change of the ALSFRS score over time. Using training and testing data we quantified and compared the performance of different techniques. We also used unsupervised machine learning methods to cluster the patients into separate computable phenotypes and interpret the derived subcohorts. Direct prediction of univariate clinical outcomes based on model-based (linear models) or model-free (machine learning based techniques - random forest and Bayesian adaptive regression trees) was only moderately successful. The correlation coefficients between clinically observed changes in ALSFRS scores relative to the model-based/model-free predicted counterparts were 0.427 (random forest) and 0.545(BART). The reliability of these results were assessed using internal statistical cross validation and well as external data validation. Unsupervised clustering generated very reliable and consistent partitions of the patient cohort into four computable phenotypic subgroups. These clusters were explicated by identifying specific salient clinical features included in the PRO-ACT archive that discriminate between the derived subcohorts. There are differences between alternative analytical methods in forecasting specific clinical phenotypes. Although predicting univariate clinical outcomes may be challenging, our results suggest that modern data science strategies are useful in clustering patients and generating evidence-based ALS hypotheses about complex interactions of multivariate factors. Predicting univariate clinical outcomes using the PRO-ACT data yields only marginal accuracy (about 70%). However, unsupervised clustering of participants into sub-groups generates stable, reliable and consistent (exceeding 95%) computable phenotypes whose explication requires interpretation of multivariate sets of features. HIGHLIGHTS: • Used a large ALS data archive of 8,000 patients consisting of 3 million records, including 200 clinical features tracked over 12 months. • Employed model-based and model-free methods to predict ALSFRS changes over time, cluster patients into cohorts, and derive computable phenotypes. • Research findings include stable, reliable, and consistent (95%) patient stratification into computable phenotypes. However, clinical explication of the results requires interpretation of multivariate information. Graphical Abstract ᅟ.

Publisher Correction: 3D Shape Modeling for Cell Nuclear Morphological Analysis and Classification.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

3D Shape Modeling for Cell Nuclear Morphological Analysis and Classification.

Quantitative analysis of morphological changes in a cell nucleus is important for the understanding of nuclear architecture and its relationship with pathological conditions such as cancer. However, dimensionality of imaging data, together with a great variability of nuclear shapes, presents challenges for 3D morphological analysis. Thus, there is a compelling need for robust 3D nuclear morphometric techniques to carry out population-wide analysis. We propose a new approach that combines modeling, analysis, and interpretation of morphometric characteristics of cell nuclei and nucleoli in 3D. We used robust surface reconstruction that allows accurate approximation of 3D object boundary. Then, we computed geometric morphological measures characterizing the form of cell nuclei and nucleoli. Using these features, we compared over 450 nuclei with about 1,000 nucleoli of epithelial and mesenchymal prostate cancer cells, as well as 1,000 nuclei with over 2,000 nucleoli from serum-starved and proliferating fibroblast cells. Classification of sets of 9 and 15 cells achieved accuracy of 95.4% and 98%, respectively, for prostate cancer cells, and 95% and 98% for fibroblast cells. To our knowledge, this is the first attempt to combine these methods for 3D nuclear shape modeling and morphometry into a highly parallel pipeline workflow for morphometric analysis of thousands of nuclei and nucleoli in 3D.

Hypothesis: Caco-2 cell rotational 3D mechanogenomic turing patterns have clinical implications to colon crypts.

Colon crypts are recognized as a mechanical and biochemical Turing patterning model. Colon epithelial Caco-2 cell monolayer demonstrated 2D Turing patterns via force analysis of apical tight junction live cell imaging which illuminated actomyosin meshwork linking the actomyosin network of individual cells. Actomyosin forces act in a mechanobiological manner that alters cell/nucleus/tissue morphology. We observed the rotational motion of the nucleus in Caco-2 cells that appears to be driven by actomyosin during the formation of a differentiated confluent epithelium. Single- to multi-cell ring/torus-shaped genomes were observed prior to complex fractal Turing patterns extending from a rotating torus centre in a spiral pattern consistent with a gene morphogen motif. These features may contribute to the well-described differentiation from stem cells at the crypt base to the luminal colon epithelium along the crypt axis. This observation may be useful to study the role of mechanogenomic processes and the underlying molecular mechanisms as determinants of cellular and tissue architecture in space and time, which is the focal point of the 4D nucleome initiative. Mathematical and bioengineer modelling of gene circuits and cell shapes may provide a powerful algorithm that will contribute to future precision medicine relevant to a number of common medical disorders.

Model-based and Model-free Machine Learning Techniques for Diagnostic Prediction and Classification of Clinical Outcomes in Parkinson's Disease.

In this study, we apply a multidisciplinary approach to investigate falls in PD patients using clinical, demographic and neuroimaging data from two independent initiatives (University of Michigan and Tel Aviv Sourasky Medical Center). Using machine learning techniques, we construct predictive models to discriminate fallers and non-fallers. Through controlled feature selection, we identified the most salient predictors of patient falls including gait speed, Hoehn and Yahr stage, postural instability and gait difficulty-related measurements. The model-based and model-free analytical methods we employed included logistic regression, random forests, support vector machines, and XGboost. The reliability of the forecasts was assessed by internal statistical (5-fold) cross validation as well as by external out-of-bag validation. Four specific challenges were addressed in the study: Challenge 1, develop a protocol for harmonizing and aggregating complex, multisource, and multi-site Parkinson's disease data; Challenge 2, identify salient predictive features associated with specific clinical traits, e.g., patient falls; Challenge 3, forecast patient falls and evaluate the classification performance; and Challenge 4, predict tremor dominance (TD) vs. posture instability and gait difficulty (PIGD). Our findings suggest that, compared to other approaches, model-free machine learning based techniques provide a more reliable clinical outcome forecasting of falls in Parkinson's patients, for example, with a classification accuracy of about 70-80%.

Opportunities and obstacles for deep learning in biology and medicine.

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems-patient classification, fundamental biological processes and treatment of patients-and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

Deep learning in pharmacogenomics: from gene regulation to patient stratification.

This Perspective provides examples of current and future applications of deep learning in pharmacogenomics, including: identification of novel regulatory variants located in noncoding domains of the genome and their function as applied to pharmacoepigenomics; patient stratification from medical records; and the mechanistic prediction of drug response, targets and their interactions. Deep learning encapsulates a family of machine learning algorithms that has transformed many important subfields of artificial intelligence over the last decade, and has demonstrated breakthrough performance improvements on a wide range of tasks in biomedicine. We anticipate that in the future, deep learning will be widely used to predict personalized drug response and optimize medication selection and dosing, using knowledge extracted from large and complex molecular, epidemiological, clinical and demographic datasets.

SOCRAT Platform Design: A Web Architecture for Interactive Visual Analytics Applications.

The modern web is a successful platform for large scale interactive web applications, including visualizations. However, there are no established design principles for building complex visual analytics (VA) web applications that could efficiently integrate visualizations with data management, computational transformation, hypothesis testing, and knowledge discovery. This imposes a time-consuming design and development process on many researchers and developers. To address these challenges, we consider the design requirements for the development of a module-based VA system architecture, adopting existing practices of large scale web application development. We present the preliminary design and implementation of an open-source platform for Statistics Online Computational Resource Analytical Toolbox (SOCRAT). This platform defines: (1) a specification for an architecture for building VA applications with multi-level modularity, and (2) methods for optimizing module interaction, re-usage, and extension. To demonstrate how this platform can be used to integrate a number of data management, interactive visualization, and analysis tools, we implement an example application for simple VA tasks including raw data input and representation, interactive visualization and analysis.

Probability Distributome: A Web Computational Infrastructure for Exploring the Properties, Interrelations, and Applications of Probability Distributions.

Probability distributions are useful for modeling, simulation, analysis, and inference on varieties of natural processes and physical phenomena. There are uncountably many probability distributions. However, a few dozen families of distributions are commonly defined and are frequently used in practice for problem solving, experimental applications, and theoretical studies. In this paper, we present a new computational and graphical infrastructure, the Distributome, which facilitates the discovery, exploration and application of diverse spectra of probability distributions. The extensible Distributome infrastructure provides interfaces for (human and machine) traversal, search, and navigation of all common probability distributions. It also enables distribution modeling, applications, investigation of inter-distribution relations, as well as their analytical representations and computational utilization. The entire Distributome framework is designed and implemented as an open-source, community-built, and Internet-accessible infrastructure. It is portable, extensible and compatible with HTML5 and Web2.0 standards (http://Distributome.org). We demonstrate two types of applications of the probability Distributome resources: computational research and science education. The Distributome tools may be employed to address five complementary computational modeling applications (simulation, data-analysis and inference, model-fitting, examination of the analytical, mathematical and computational properties of specific probability distributions, and exploration of the inter-distributional relations). Many high school and college science, technology, engineering and mathematics (STEM) courses may be enriched by the use of modern pedagogical approaches and technology-enhanced methods. The Distributome resources provide enhancements for blended STEM education by improving student motivation, augmenting the classical curriculum with interactive webapps, and overhauling the learning assessment protocols.

SOCR data dashboard: an integrated big data archive mashing medicare, labor, census and econometric information.

INTRODUCTION: Intuitive formulation of informative and computationally-efficient queries on big and complex datasets present a number of challenges. As data collection is increasingly streamlined and ubiquitous, data exploration, discovery and analytics get considerably harder. Exploratory querying of heterogeneous and multi-source information is both difficult and necessary to advance our knowledge about the world around us. RESEARCH DESIGN: We developed a mechanism to integrate dispersed multi-source data and service the mashed information via human and machine interfaces in a secure, scalable manner. This process facilitates the exploration of subtle associations between variables, population strata, or clusters of data elements, which may be opaque to standard independent inspection of the individual sources. This a new platform includes a device agnostic tool (Dashboard webapp, http://socr.umich.edu/HTML5/Dashboard/) for graphical querying, navigating and exploring the multivariate associations in complex heterogeneous datasets. RESULTS: The paper illustrates this core functionality and serviceoriented infrastructure using healthcare data (e.g., US data from the 2010 Census, Demographic and Economic surveys, Bureau of Labor Statistics, and Center for Medicare Services) as well as Parkinson's Disease neuroimaging data. Both the back-end data archive and the front-end dashboard interfaces are continuously expanded to include additional data elements and new ways to customize the human and machine interactions. CONCLUSIONS: A client-side data import utility allows for easy and intuitive integration of user-supplied datasets. This completely open-science framework may be used for exploratory analytics, confirmatory analyses, meta-analyses, and education and training purposes in a wide variety of fields.

Calmodulin 4 is dispensable for epidermal barrier formation and wound healing in mice.

Calcium-mediated signals play important roles in epidermal barrier formation, skin homoeostasis and wound repair. Calmodulin 4 (Calm4) is a small, Ca2+ -binding protein with strong expression in suprabasal keratinocytes. In mice, Calm4 first appears in the skin at the time of barrier formation, and its expression increases in response to epidermal barrier challenges. In this study, we report the generation of Calm4 knockout mice and provide evidence that Calm4 is dispensable for epidermal barrier formation, maintenance and repair.

Role of Scarf and its binding target proteins in epidermal calcium homeostasis.

The novel Ca2+-binding protein, Scarf (skin calmodulin-related factor) belongs to the calmodulin-like protein family and is expressed in the differentiated layers of the epidermis. To determine the roles of Scarf during stratification, we set out to identify the binding target proteins by affinity chromatography and subsequent analysis by mass spectrometry. Several binding factors, including 14-3-3s, annexins, calreticulin, ERp72 (endoplasmic reticulum protein 72), and nucleolin, were identified, and their interactions with Scarf were corroborated by co-immunoprecipitation and co-localization analyses. To further understand the functions of Scarf in epidermis in vivo, we altered the epidermal Ca2+ gradient by acute barrier disruption. The change in the expression levels of Scarf and its binding target proteins were determined by immunohistochemistry and Western blot analysis. The expression of Scarf, annexins, calreticulin, and ERp72 were up-regulated by Ca2+ gradient disruption, whereas the expression of 14-3-3s and nucleolin was reduced. Because annexins, calreticulin, and ERp72 have been implicated in Ca2+-induced cellular trafficking, including the secretion of lamellar bodies and Ca2+ homeostasis, we propose that the interaction of Scarf with these proteins might be crucial in the process of barrier restoration. On the other hand, down-regulation of 14-3-3s and nucleolin is potentially involved in the process of keratinocyte differentiation and growth inhibition. The calcium-dependent localization and up-regulation of Scarf and its binding target proteins were studied in mouse keratinocytes treated with ionomycin and during the wound-healing process. We found increased expression and nuclear presence of Scarf in the epidermis of the wound edge 4 and 7 days post-wounding, entailing the role of Scarf in barrier restoration. Our results suggest that Scarf plays a critical role as a Ca2+ sensor, potentially regulating the function of its binding target proteins in a Ca2+-dependent manner in the process of restoration of epidermal Ca2+ gradient as well as during epidermal barrier formation.