Computational prediction of neural progenitor cell fates

Chapter 17: bioimage informatics for systems pharmacology

Recent advances in automated high-resolution fluorescence microscopy and robotic handling have made the systematic and cost effective study of diverse morphological changes within a large population of cells possible under a variety of perturbations, e.g., drugs, compounds, metal catalysts, RNA interference (RNAi). Cell population-based studies deviate from conventional microscopy studies on a few cells, and could provide stronger statistical power for drawing experimental observations and conclusions. However, it is challenging to manually extract and quantify phenotypic changes from the large amounts of complex image data generated. Thus, bioimage informatics approaches are needed to rapidly and objectively quantify and analyze the image data. This paper provides an overview of the bioimage informatics challenges and approaches in image-based studies for drug and target discovery. The concepts and capabilities of image-based screening are first illustrated by a few practical examples investigating different kinds of phenotypic changes caEditorsused by drugs, compounds, or RNAi. The bioimage analysis approaches, including object detection, segmentation, and tracking, are then described. Subsequently, the quantitative features, phenotype identification, and multidimensional profile analysis for profiling the effects of drugs and targets are summarized. Moreover, a number of publicly available software packages for bioimage informatics are listed for further reference. It is expected that this review will help readers, including those without bioimage informatics expertise, understand the capabilities, approaches, and tools of bioimage informatics and apply them to advance their own studies.

New insights into mechanisms of stem cell daughter fate determination in regenerative tissues

Stem cells can self-renew and differentiate over extended periods of time. Understanding how stem cells acquire their fates is a central question in stem cell biology. Early work in Drosophila germ line and neuroblast showed that fate choice is achieved by strict asymmetric divisions that can generate each time one stem and one differentiated cell. More recent work suggests that during homeostasis, some stem cells can divide symmetrically to generate two differentiated cells or two identical stem cells to compensate for stem cell loss that occurred by direct differentiation or apoptosis. The interplay of all these factors ensures constant tissue regeneration and the maintenance of stem cell pool size. This interplay can be modeled as a population-deterministic dynamics that, at least in some systems, may be described as stochastic behavior. Here, we overview recent progress made on the characterization of stem cell dynamics in regenerative tissues.

A multiple model probability hypothesis density tracker for time-lapse cell microscopy sequences

Quantitative analysis of the dynamics of tiny cellular and subcellular structures in time-lapse cell microscopy sequences requires the development of a reliable multi-target tracking method capable of tracking numerous similar targets in the presence of high levels of noise, high target density, maneuvering motion patterns and intricate interactions. The linear Gaussian jump Markov system probability hypothesis density (LGJMS-PHD) filter is a recent Bayesian tracking filter that is well-suited for this task. However, the existing recursion equations for this filter do not consider a state-dependent transition probability matrix. As required in many biological applications, we propose a new closed-form recursion that incorporates this assumption and introduce a general framework for particle tracking using the proposed filter. We apply our scheme to multi-target tracking in total internal reflection fluorescence microscopy (TIRFM) sequences and evaluate the performance of our filter against the existing LGJMS-PHD and IMM-JPDA filters.

Advances in tracking hematopoiesis at the single-cell level

PURPOSE OF REVIEW

Studying heterogeneous populations, such as hematopoietic stem cells (HSCs), requires continuous long-term observation of living cells at the single-cell level. The purpose of this review is to discuss recent advances in technologies required for continuous single-cell analysis and the contribution of this approach to find answers in hematopoiesis research.

RECENT FINDINGS

Continuous long-term imaging at the single-cell level still requires custom-made hardware, software and manual in-depth analysis of large amounts of data. Despite these technical difficulties, continuous time-lapse imaging and single-cell tracking are increasingly used in hematopoiesis research. It has already contributed to answering decades-old questions.

SUMMARY

Continuous long-term single-cell analysis is indispensable for a comprehensive analysis of dynamic processes in heterogeneous cell populations. Despite many remaining technological hurdles, this approach is increasingly used in hematopoiesis research.

Morphology-based features for adaptive mitosis detection of in vitro stem cell tracking data

OBJECTIVES

The cultivation of adherently growing cell populations is a major task in the field of adult stem cell production used for drug discovery and in the field of regenerative medicine. To assessthe quality of a cell population, a crucial event is the mitotic cell division: the precise knowledge of these events enables the reconstruction of lineages and accurate proliferation curves as well as a detailed analysis of cell cycles. To serve in an autonomous cell farming framework, such a detector requires to work reliably and unsupervised.

METHODS

We introduce a mitosis detector that is using a maximum likelihood (ML) estimator based on morphological cell features (cell area, brightness, length, compactness). It adapts to the 3 phases of cell growth (lag, log and stationary phase). As a concurrent model, we compared ML with kernel SVMs using linear, quadratic and Gaussian kernel functions. All approaches are evaluated for their ability to distinguish between mitotic and non-mitotic events. The large, publicly available benchmark data CeTReS (reference data set A with >240,000 segmented cells, >2,000 mitotic events) is used for this evaluation.

RESULTS

The adaptive (unsupervised) ML approach clearly outperforms previously published non-adaptive approaches and the linear SVM. Furthermore, it robustly reaches a performance comparable to quadratic and Gaussian SVM.

CONCLUSIONS

The proposed simple and label free adaptive variant might be the method of choice when it comes to autonomous cell farming. Hereby, it is essential to have reliable and unsupervised mitosis detection that covers all phases of cell growth.

Bovine mammary stem cells: cell biology meets production agriculture

Mammary stem cells (MaSC) provide for net growth, renewal and turnover of mammary epithelial cells, and are therefore potential targets for strategies to increase production efficiency. Appropriate regulation of MaSC can potentially benefit milk yield, persistency, dry period management and tissue repair. Accordingly, we and others have attempted to characterize and alter the function of bovine MaSC. In this review, we provide an overview of current knowledge of MaSC gained from studies using mouse and human model systems and present research on bovine MaSC within that context. Recent data indicate that MaSC retain labeled DNA for extended periods because of their selective segregation of template DNA strands during mitosis. Relying on this long-term retention of bromodeoxyuridine-labeled DNA, we identified putative bovine MaSC. These label-retaining epithelial cells (LREC) are in low abundance within mammary epithelium (<1%). They are predominantly estrogen receptor (ER)-negative and localized in a basal or suprabasal layer of the epithelium throughout the gland. Thus, the response of MaSC to estrogen, the major mitogen in mammary gland, is likely mediated by paracrine factors released by cells that are ER-positive. This is consistent with considerable evidence for cross-talk within and between epithelial cells and surrounding stromal cells. Excision of classes of cells by laser microdissection and subsequent microarray analysis will hopefully provide markers for MaSC and insights into their regulation. Preliminary analyses of gene expression in laser-microdissected LREC and non-LREC are consistent with the concept that LREC represent populations of stem cells and progenitor cells that differ with regard to their properties and location within the epithelial layer. We have attempted to modulate the MaSC number by infusing a solution of xanthosine through the teat canal and into the ductal network of the mammary glands of prepubertal heifers. This treatment increased the number of putative stem cells, as evidenced by an increase in the percentage of LREC and increased telomerase activity within the tissue. The exciting possibility that stem cell expansion can influence milk production is currently under investigation.

Time-lapse microscopy and classification of 2D human mesenchymal stem cells based on cell shape picks up myogenic from osteogenic and adipogenic differentiation

Current methods to characterize mesenchymal stem cells (MSCs) are limited to CD marker expression, plastic adherence and their ability to differentiate into adipogenic, osteogenic and chondrogenic precursors. It seems evident that stem cells undergoing differentiation should differ in many aspects, such as morphology and possibly also behaviour; however, such a correlation has not yet been exploited for fate prediction of MSCs. Primary human MSCs from bone marrow were expanded and pelleted to form high-density cultures and were then randomly divided into four groups to differentiate into adipogenic, osteogenic chondrogenic and myogenic progenitor cells. The cells were expanded as heterogeneous and tracked with time-lapse microscopy to record cell shape, using phase-contrast microscopy. The cells were segmented using a custom-made image-processing pipeline. Seven morphological features were extracted for each of the segmented cells. Statistical analysis was performed on the seven-dimensional feature vectors, using a tree-like classification method. Differentiation of cells was monitored with key marker genes and histology. Cells in differentiation media were expressing the key genes for each of the three pathways after 21 days, i.e. adipogenic, osteogenic and chondrogenic, which was also confirmed by histological staining. Time-lapse microscopy data were obtained and contained new evidence that two cell shape features, eccentricity and filopodia (= 'fingers') are highly informative to classify myogenic differentiation from all others. However, no robust classifiers could be identified for the other cell differentiation paths. The results suggest that non-invasive automated time-lapse microscopy could potentially be used to predict the stem cell fate of hMSCs for clinical application, based on morphology for earlier time-points. The classification is challenged by cell density, proliferation and possible unknown donor-specific factors, which affect the performance of morphology-based approaches.

Loss of Llgl1 in retinal neuroepithelia reveals links between apical domain size, Notch activity and neurogenesis

To gain insights into the cellular mechanisms of neurogenesis, we analyzed retinal neuroepithelia deficient for Llgl1, a protein implicated in apicobasal cell polarity, asymmetric cell division, cell shape and cell cycle exit. We found that vertebrate retinal neuroepithelia deficient for Llgl1 retained overt apicobasal polarity, but had expanded apical domains. Llgl1 retinal progenitors also had increased Notch activity and reduced rates of neurogenesis. Blocking Notch function by depleting Rbpj restored normal neurogenesis. Experimental expansion of the apical domain, through inhibition of Shroom3, also increased Notch activity and reduced neurogenesis. Significantly, in wild-type retina, neurogenic retinal progenitors had smaller apical domains compared with proliferative neuroepithelia. As nuclear position during interkinetic nuclear migration (IKNM) has been previously linked with cell cycle exit, we analyzed this phenomenon in cells depleted of Llgl1. We found that although IKNM was normal, the relationship between nuclear position and neurogenesis was shifted away from the apical surface, consistent with increased pro-proliferative and/or anti-neurogenic signals associated with the apical domain. These data, in conjunction with other findings, suggest that, in retinal neuroepithelia, the size of the apical domain modulates the strength of polarized signals that influence neurogenesis.

Online 3-D tracking of suspension living cells imaged with phase-contrast microscopy

Neural stem cells/neural progenitors (NSCs/NPs) are cells that give rise to the main cell types of the nervous system: oligodendrocytes, neurons, and astrocytes. Studying NSCs/NPs with time-lapse microscopy is critical to the understanding of the biology of these cells. However, NSCs/NPs are very sensitive to phototoxic damage, and therefore, fluorescent dyes cannot be used to follow these cells. Also, since in most of NSC/NP-related experiments, a large number of cells neesd to be monitored. Consequently, the acquisition of a huge amount of images is required. An additional difficulty is related to our original suspension living, tracking objective, behavior much closer to the natural, in vivo, way of development of the cells. Indeed, unlike adherent cells, suspension cells float freely in a liquid solution, thus, making their dynamics very different from that of adherent cells. As a result, existing visual tracking algorithms that have primarily been developed to track adherent cells are no longer adequate to tackle living cells in suspension. This paper presents a novel automated 3-D visual tracking of suspension living cells for time-lapse image acquisition using phase-contrast microscopy. This new tracking method can potentially strongly impact on current 3-D video microscopy methods, paving the way for innovative analysis of NSCs/NPs and as a result, on the study of neurodegenerative diseases.

Axonal transport analysis using Multitemporal Association Tracking

Multitemporal Association Tracking (MAT) is a new graph-based method for multitarget tracking in biological applications that reduces the error rate and implementation complexity compared to approaches based on bipartite matching. The data association problem is solved over a window of future detection data using a graph-based cost function that approximates the Bayesian a posteriori association probability. MAT has been applied to hundreds of image sequences, tracking organelle and vesicles to quantify the deficiencies in axonal transport that can accompany neurodegenerative disorders such as Huntington's Disease and Multiple Sclerosis and to quantify changes in transport in response to therapeutic interventions.

Extracting fluorescent reporter time courses of cell lineages from high-throughput microscopy at low temporal resolution

The extraction of fluorescence time course data is a major bottleneck in high-throughput live-cell microscopy. Here we present an extendible framework based on the open-source image analysis software ImageJ, which aims in particular at analyzing the expression of fluorescent reporters through cell divisions. The ability to track individual cell lineages is essential for the analysis of gene regulatory factors involved in the control of cell fate and identity decisions. In our approach, cell nuclei are identified using Hoechst, and a characteristic drop in Hoechst fluorescence helps to detect dividing cells. We first compare the efficiency and accuracy of different segmentation methods and then present a statistical scoring algorithm for cell tracking, which draws on the combination of various features, such as nuclear intensity, area or shape, and importantly, dynamic changes thereof. Principal component analysis is used to determine the most significant features, and a global parameter search is performed to determine the weighting of individual features. Our algorithm has been optimized to cope with large cell movements, and we were able to semi-automatically extract cell trajectories across three cell generations. Based on the MTrackJ plugin for ImageJ, we have developed tools to efficiently validate tracks and manually correct them by connecting broken trajectories and reassigning falsely connected cell positions. A gold standard consisting of two time-series with 15,000 validated positions will be released as a valuable resource for benchmarking. We demonstrate how our method can be applied to analyze fluorescence distributions generated from mouse stem cells transfected with reporter constructs containing transcriptional control elements of the Msx1 gene, a regulator of pluripotency, in mother and daughter cells. Furthermore, we show by tracking zebrafish PAC2 cells expressing FUCCI cell cycle markers, our framework can be easily adapted to different cell types and fluorescent markers.

Cell biological regulation of division fate in vertebrate neuroepithelial cells

The developing nervous system derives from neuroepithelial progenitor cells that divide to generate all of the mature neuronal types. For the proper complement of cell types to form, the progenitors must produce postmitotic cells, yet also replenish the progenitor pool. Progenitor divisions can be classified into three general types: symmetric proliferative (producing two progenitors), asymmetric neurogenic (producing one progenitor and one postmitotic cell), and symmetric neurogenic (producing two postmitotic cells). The appropriate ratios for these modes of cell division require intrinsic polarity, which is one of the characteristics that define neuroepithelial progenitor cells. The type of division an individual progenitor undergoes can be influenced by cellular features, or behaviors, which are heterogeneous within the population of progenitors. Here we review three key cellular parameters, asymmetric inheritance, cell cycle kinetics, and interkinetic nuclear migration, and the possible mechanisms for how these features influence progenitor fates.

Vertebrate neural stem cell segmentation, tracking and lineaging with validation and editing

This protocol and the accompanying software program called LEVER (lineage editing and validation) enable quantitative automated analysis of phase-contrast time-lapse images of cultured neural stem cells. Images are captured at 5-min intervals over a period of 5-15 d as the cells proliferate and differentiate. LEVER automatically segments, tracks and generates lineage trees of the stem cells from the image sequence. In addition to generating lineage trees capturing the population dynamics of clonal development, LEVER extracts quantitative phenotypic measurements of cell location, shape, movement and size. When available, the system can include biomolecular markers imaged using fluorescence. It then displays the results to the user for highly efficient inspection and editing to correct any errors in the segmentation, tracking or lineaging. To enable high-throughput inspection, LEVER incorporates features for rapid identification of errors and for learning from user-supplied corrections to automatically identify and correct related errors.

Mammalian cell dedifferentiation as a possible outcome of stress

Differentiation cascades are arranged hierarchically; stem cells positioned at the top of the hierarchy generate committed progenitors that, in turn, proliferate and further differentiate stepwise into mature progeny. This rigid, irreversible structure ensures the phenotypic stability of adult tissues. However, such rigidity may be problematic under conditions of tissue damage when reconstitution is required. Although it may seem unlikely that the restrictions on changes in cell phenotypes would be lifted to enable tissue reconstitution, it is nevertheless possible that mammalian tissues are endowed with sufficient flexibility to enable their adaptation to extreme conditions.

A novel validation algorithm allows for automated cell tracking and the extraction of biologically meaningful parameters

Automated microscopy is currently the only method to non-invasively and label-free observe complex multi-cellular processes, such as cell migration, cell cycle, and cell differentiation. Extracting biological information from a time-series of micrographs requires each cell to be recognized and followed through sequential microscopic snapshots. Although recent attempts to automatize this process resulted in ever improving cell detection rates, manual identification of identical cells is still the most reliable technique. However, its tedious and subjective nature prevented tracking from becoming a standardized tool for the investigation of cell cultures. Here, we present a novel method to accomplish automated cell tracking with a reliability comparable to manual tracking. Previously, automated cell tracking could not rival the reliability of manual tracking because, in contrast to the human way of solving this task, none of the algorithms had an independent quality control mechanism; they missed validation. Thus, instead of trying to improve the cell detection or tracking rates, we proceeded from the idea to automatically inspect the tracking results and accept only those of high trustworthiness, while rejecting all other results. This validation algorithm works independently of the quality of cell detection and tracking through a systematic search for tracking errors. It is based only on very general assumptions about the spatiotemporal contiguity of cell paths. While traditional tracking often aims to yield genealogic information about single cells, the natural outcome of a validated cell tracking algorithm turns out to be a set of complete, but often unconnected cell paths, i.e. records of cells from mitosis to mitosis. This is a consequence of the fact that the validation algorithm takes complete paths as the unit of rejection/acceptance. The resulting set of complete paths can be used to automatically extract important biological parameters with high reliability and statistical significance. These include the distribution of life/cycle times and cell areas, as well as of the symmetry of cell divisions and motion analyses. The new algorithm thus allows for the quantification and parameterization of cell culture with unprecedented accuracy. To evaluate our validation algorithm, two large reference data sets were manually created. These data sets comprise more than 320,000 unstained adult pancreatic stem cells from rat, including 2592 mitotic events. The reference data sets specify every cell position and shape, and assign each cell to the correct branch of its genealogic tree. We provide these reference data sets for free use by others as a benchmark for the future improvement of automated tracking methods.

Generation of Rab-based transgenic lines for in vivo studies of endosome biology in zebrafish

The Rab family of small GTPases function as molecular switches regulating membrane and protein trafficking. Individual Rab isoforms define and are required for specific endosomal compartments. To facilitate in vivo investigation of specific Rab proteins, and endosome biology in general, we have generated transgenic zebrafish lines to mark and manipulate Rab proteins. We also developed software to track and quantify endosome dynamics within time-lapse movies. The established transgenic lines ubiquitously express EGFP fusions of Rab5c (early endosomes), Rab11a (recycling endosomes), and Rab7 (late endosomes) to study localization and dynamics during development. Additionally, we generated UAS-based transgenic lines expressing constitutive active (CA) and dominant-negative (DN) versions for each of these Rab proteins. Predicted localization and functional consequences for each line were verified through a variety of assays, including lipophilic dye uptake and Crumbs2a localization. In summary, we have established a toolset for in vivo analyses of endosome dynamics and functions.

The ePetri dish, an on-chip cell imaging platform based on subpixel perspective sweeping microscopy (SPSM)

We report a chip-scale lensless wide-field-of-view microscopy imaging technique, subpixel perspective sweeping microscopy, which can render microscopy images of growing or confluent cell cultures autonomously. We demonstrate that this technology can be used to build smart Petri dish platforms, termed ePetri, for cell culture experiments. This technique leverages the recent broad and cheap availability of high performance image sensor chips to provide a low-cost and automated microscopy solution. Unlike the two major classes of lensless microscopy methods, optofluidic microscopy and digital in-line holography microscopy, this new approach is fully capable of working with cell cultures or any samples in which cells may be contiguously connected. With our prototype, we demonstrate the ability to image samples of area 6 mm × 4 mm at 660-nm resolution. As a further demonstration, we showed that the method can be applied to image color stained cell culture sample and to image and track cell culture growth directly within an incubator. Finally, we showed that this method can track embryonic stem cell differentiations over the entire sensor surface. Smart Petri dish based on this technology can significantly streamline and improve cell culture experiments by cutting down on human labor and contamination risks.

Eyes wide open: a critical review of sphere-formation as an assay for stem cells

Sphere-forming assays have been widely used to retrospectively identify stem cells based on their reported capacity to evaluate self-renewal and differentiation at the single-cell level in vitro. The discovery of markers that allow the prospective isolation of stem cells and their progeny from their in vivo niche allows the functional properties of purified populations to be defined. We provide a historical perspective of the evolution of the neurosphere assay and highlight limitations in the use of sphere-forming assays in the context of neurospheres. We discuss theoretical and technical considerations of experimental design and interpretation that surround the use of this assay with any tissue.

The cell monolayer trajectory from the system state point of view

Time-lapse microscopic movies are being increasingly utilized for understanding the derivation of cell states and predicting cell future. Often, fluorescence and other types of labeling are not available or desirable, and cell state-definitions based on observable structures must be used. We present the methodology for cell behavior recognition and prediction based on the short term cell recurrent behavior analysis. This approach has theoretical justification in non-linear dynamics theory. The methodology is based on the general stochastic systems theory which allows us to define the cell states, trajectory and the system itself. We introduce the usage of a novel image content descriptor based on information contribution (gain) by each image point for the cell state characterization as the first step. The linkage between the method and the general system theory is presented as a general frame for cell behavior interpretation. We also discuss extended cell description, system theory and methodology for future development. This methodology may be used for many practical purposes, ranging from advanced, medically relevant, precise cell culture diagnostics to very utilitarian cell recognition in a noisy or uneven image background. In addition, the results are theoretically justified.

Continuous live imaging of adult neural stem cell division and lineage progression in vitro

Little is known about the intrinsic specification of adult neural stem cells (NSCs) and to what extent they depend on their local niche. To observe adult NSC division and lineage progression independent of their niche, we isolated cells from the adult mouse subependymal zone (SEZ) and cultured them at low density without growth factors. We demonstrate here that SEZ cells in this culture system are primarily neurogenic and that adult NSCs progress through stereotypic lineage trees consisting of asymmetric stem cell divisions, symmetric transit-amplifying divisions and final symmetric neurogenic divisions. Stem cells, identified by their astro/radial glial identity and their slow-dividing nature, were observed to generate asymmetrically and fast-dividing cells that maintained an astro/radial glia identity. These, in turn, gave rise to symmetrically and fast-dividing cells that lost glial hallmarks, but had not yet acquired neuronal features. The number of amplifying divisions was limited to a maximum of five in this system. Moreover, we found that cell growth correlated with the number of subsequent divisions of SEZ cells, with slow-dividing astro/radial glia exhibiting the most substantial growth prior to division. The fact that in the absence both of exogenously supplied growth factors and of signals provided by the local niche neurogenic lineage progression takes place in such stereotypic fashion, suggests that lineage progression is, to a significant degree, cell intrinsic or pre-programmed at the beginning of the lineage.

Long-term single-cell imaging of mammalian stem cells

Continuous long-term single-cell observation provides insight into the molecular control of cell fate. This is particularly important for rare and heterogeneous populations of cells, such as mammalian stem cells. The current lack of usable off-the-shelf hardware and software for such experiments makes their implementation technically challenging. Here I discuss the need for continuous single-cell quantification to understand molecular cell fate control as well as organizational and technical solutions for long-term imaging and tracking of stem cells.

Reconstruction of rat retinal progenitor cell lineages in vitro reveals a surprising degree of stochasticity in cell fate decisions

In vivo cell lineage-tracing studies in the vertebrate retina have revealed that the sizes and cellular compositions of retinal clones are highly variable. It has been challenging to ascertain whether this variability reflects distinct but reproducible lineages among many different retinal progenitor cells (RPCs) or is the product of stochastic fate decisions operating within a population of more equivalent RPCs. To begin to distinguish these possibilities, we developed a method for long-term videomicroscopy to follow the lineages of rat perinatal RPCs cultured at clonal density. In such cultures, cell-cell interactions between two different clones are eliminated and the extracellular environment is kept constant, allowing us to study the cell-intrinsic potential of a given RPC. Quantitative analysis of the reconstructed lineages showed that the mode of division of RPCs is strikingly consistent with a simple stochastic pattern of behavior in which the decision to multiply or differentiate is set by fixed probabilities. The variability seen in the composition and order of cell type genesis within clones is well described by assuming that each of the four different retinal cell types generated at this stage is chosen stochastically by differentiating neurons, with relative probabilities of each type set by their abundance in the mature retina. Although a few of the many possible combinations of cell types within clones occur at frequencies that are incompatible with a fully stochastic model, our results support the notion that stochasticity has a major role during retinal development and therefore possibly in other parts of the central nervous system.

Non-invasive imaging of human embryos before embryonic genome activation predicts development to the blastocyst stage

We report studies of preimplantation human embryo development that correlate time-lapse image analysis and gene expression profiling. By examining a large set of zygotes from in vitro fertilization (IVF), we find that success in progression to the blastocyst stage can be predicted with >93% sensitivity and specificity by measuring three dynamic, noninvasive imaging parameters by day 2 after fertilization, before embryonic genome activation (EGA). These parameters can be reliably monitored by automated image analysis, confirming that successful development follows a set of carefully orchestrated and predictable events. Moreover, we show that imaging phenotypes reflect molecular programs of the embryo and of individual blastomeres. Single-cell gene expression analysis reveals that blastomeres develop cell autonomously, with some cells advancing to EGA and others arresting. These studies indicate that success and failure in human embryo development is largely determined before EGA. Our methods and algorithms may provide an approach for early diagnosis of embryo potential in assisted reproduction.