Optimal-Transport Analysis of Single-Cell Gene Expression Identifies Developmental Trajectories in Reprogramming

Chemical reprogramming of melanocytes to skeletal muscle cells

BACKGROUND

Direct cell-fate conversion by chemical reprogramming is promising for regenerative cell therapies. However, this process requires the reactivation of a set of master transcription factors (TFs) of the target cell type, which has proven challenging using only small molecules.

METHODS

We developed a novel small-molecule cocktail permitting robust skin cell to muscle cell conversion. By single cell sequencing analysis, we identified a Pax3 (Paired box 3)-expressing melanocyte population holding a superior myogenic potential outperforming other seven types of skin cells. We further validated the single cell sequencing analysis results using immunofluorescence staining, in situ hybridization and FACS sorting and confirmed the myogenic potential of melanocytes during chemical reprogramming. We used single cell RNA-seq that detect the potential converted cell type, uncovering a unique role of Pax3 in facilitating chemical reprogramming from melanocytes to muscle cells.

RESULTS

In this study, we demonstrated that the Pax3-expressing melanocytes to be a skin cell type for skeletal muscle cell fate conversion in chemical reprogramming. By developing a small-molecule cocktail, we showed an efficient melanocyte reprogramming to skeletal muscle cells (40%, P < 0.001). The endogenous expression of specific TFs may circumvent the additional requirement for TF reactivation and form a shortcut for cell fate conversion, suggesting a basic principle that could ease cell fate conversion.

CONCLUSIONS

Our study demonstrates the first report of melanocyte-to-muscle conversion by small molecules, suggesting a novel strategy for muscle regeneration. Furthermore, skin is one of the tissues closely located to skeletal muscle, and therefore, our results provide a promising foundation for therapeutic chemical reprogramming in vivo treating skeletal muscle degenerative diseases.

Roadmap of the Early Events of In Vivo Somatic Cell Reprogramming

Single-cell transcriptomics and in situ imaging of murine pancreas upon partial reprogramming in vivo reveal transcriptional dynamics upon Oct4, Sox2, Klf4, and cMyc (OSKM) induction. Interestingly, transcriptomic signatures of partial reprogramming observed in pancreas are shared by several tissues upon OSKM induction as well as during in vitro reprogramming of fibroblasts, pointing to the existence of conserved pathways critical for early reprogramming, regeneration, and rejuvenation.

Synthetic genetic circuits to uncover and enforce the OCT4 trajectories of successful reprogramming of human fibroblasts

Reprogramming human fibroblasts to induced pluripotent stem cells (iPSCs) is inefficient, with heterogeneity among transcription factor (TF) trajectories driving divergent cell states. Nevertheless, the impact of TF dynamics on reprogramming efficiency remains uncharted. Here, we identify the successful reprogramming trajectories of the core pluripotency TF, OCT4, and design a genetic controller that enforces such trajectories with high precision. By combining a genetic circuit that generates a wide range of OCT4 trajectories with live-cell imaging, we track OCT4 trajectories with clonal resolution and find that a distinct constant OCT4 trajectory is required for colony formation. We then develop a synthetic genetic circuit that yields a tight OCT4 distribution around the identified trajectory and outperforms in terms of reprogramming efficiency other circuits that less accurately regulate OCT4. Our synthetic biology approach is generalizable for identifying and enforcing TF dynamics for cell fate programming applications.

Understanding How Cells Probe the World: A Preliminary Step towards Modeling Cell Behavior?

Cell biologists have long aimed at quantitatively modeling cell function. Recently, the outstanding progress of high-throughput measurement methods and data processing tools has made this a realistic goal. The aim of this paper is twofold: First, to suggest that, while much progress has been done in modeling cell states and transitions, current accounts of environmental cues driving these transitions remain insufficient. There is a need to provide an integrated view of the biochemical, topographical and mechanical information processed by cells to take decisions. It might be rewarding in the near future to try to connect cell environmental cues to physiologically relevant outcomes rather than modeling relationships between these cues and internal signaling networks. The second aim of this paper is to review exogenous signals that are sensed by living cells and significantly influence fate decisions. Indeed, in addition to the composition of the surrounding medium, cells are highly sensitive to the properties of neighboring surfaces, including the spatial organization of anchored molecules and substrate mechanical and topographical properties. These properties should thus be included in models of cell behavior. It is also suggested that attempts at cell modeling could strongly benefit from two research lines: (i) trying to decipher the way cells encode the information they retrieve from environment analysis, and (ii) developing more standardized means of assessing the quality of proposed models, as was done in other research domains such as protein structure prediction.

Topological Speed Limit

Any physical system evolves at a finite speed that is constrained not only by the energetic cost but also by the topological structure of the underlying dynamics. In this Letter, by considering such structural information, we derive a unified topological speed limit for the evolution of physical states using an optimal transport approach. We prove that the minimum time required for changing states is lower bounded by the discrete Wasserstein distance, which encodes the topological information of the system, and the time-averaged velocity. The bound obtained is tight and applicable to a wide range of dynamics, from deterministic to stochastic, and classical to quantum systems. In addition, the bound provides insight into the design principles of the optimal process that attains the maximum speed. We demonstrate the application of our results to chemical reaction networks and interacting many-body quantum systems.

Single-cell and long-read sequencing to enhance modelling of splicing and cell-fate determination

Single-cell sequencing technologies have revolutionised the life sciences and biomedical research. Single-cell sequencing provides high-resolution data on cell heterogeneity, allowing high-fidelity cell type identification, and lineage tracking. Computational algorithms and mathematical models have been developed to make sense of the data, compensate for errors and simulate the biological processes, which has led to breakthroughs in our understanding of cell differentiation, cell-fate determination and tissue cell composition. The development of long-read (a.k.a. third-generation) sequencing technologies has produced powerful tools for investigating alternative splicing, isoform expression (at the RNA level), genome assembly and the detection of complex structural variants (at the DNA level). In this review, we provide an overview of the recent advancements in single-cell and long-read sequencing technologies, with a particular focus on the computational algorithms that help in correcting, analysing, and interpreting the resulting data. Additionally, we review some mathematical models that use single-cell and long-read sequencing data to study cell-fate determination and alternative splicing, respectively. Moreover, we highlight the emerging opportunities in modelling cell-fate determination that result from the combination of single-cell and long-read sequencing technologies.

Comprehensive single-cell transcriptional profiling defines shared and unique epithelial injury responses during kidney fibrosis

The underlying cellular events driving kidney fibrogenesis and metabolic dysfunction are incompletely understood. Here, we employed single-cell combinatorial indexing RNA sequencing to analyze 24 mouse kidneys from two fibrosis models. We profiled 309,666 cells in one experiment, representing 50 cell types/states encompassing epithelial, endothelial, immune, and stromal populations. Single-cell analysis identified diverse injury states of the proximal tubule, including two distinct early-phase populations with dysregulated lipid and amino acid metabolism, respectively. Lipid metabolism was defective in the chronic phase but was transiently activated in the very early stages of ischemia-induced injury, where we discovered increased lipid deposition and increased fatty acid β-oxidation. Perilipin 2 was identified as a surface marker of intracellular lipid droplets, and its knockdown in vitro disrupted cell energy state maintenance during lipid accumulation. Surveying epithelial cells across nephron segments identified shared and unique injury responses. Stromal cells exhibited high heterogeneity and contributed to fibrogenesis by epithelial-stromal crosstalk.

Highly efficient reprogrammable mouse lines with integrated reporters to track the route to pluripotency

Revealing the molecular events associated with reprogramming different somatic cell types to pluripotency is critical for understanding the characteristics of induced pluripotent stem cell (iPSC) therapeutic derivatives. Inducible reprogramming factor transgenic cells or animals-designated as secondary (2°) reprogramming systems-not only provide excellent experimental tools for such studies but also offer a strategy to study the variances in cellular reprogramming outcomes due to different in vitro and in vivo environments. To make such studies less cumbersome, it is desirable to have a variety of efficient reprogrammable mouse systems to induce successful mass reprogramming in somatic cell types. Here, we report the development of two transgenic mouse lines from which 2° cells reprogram with unprecedented efficiency. These systems were derived by exposing primary reprogramming cells containing doxycycline-inducible Yamanaka factor expression to a transient interruption in transgene expression, resulting in selection for a subset of clones with robust transgene response. These systems also include reporter genes enabling easy readout of endogenous Oct4 activation (GFP), indicative of pluripotency, and reprogramming transgene expression (mCherry). Notably, somatic cells derived from various fetal and adult tissues from these 2° mouse lines gave rise to highly efficient and rapid reprogramming, with transgene-independent iPSC colonies emerging as early as 1 wk after induction. These mouse lines serve as a powerful tool to explore sources of variability in reprogramming and the mechanistic underpinnings of efficient reprogramming systems.

A unified computational framework for single-cell data integration with optimal transport

Single-cell data integration can provide a comprehensive molecular view of cells. However, how to integrate heterogeneous single-cell multi-omics as well as spatially resolved transcriptomic data remains a major challenge. Here we introduce uniPort, a unified single-cell data integration framework that combines a coupled variational autoencoder (coupled-VAE) and minibatch unbalanced optimal transport (Minibatch-UOT). It leverages both highly variable common and dataset-specific genes for integration to handle the heterogeneity across datasets, and it is scalable to large-scale datasets. uniPort jointly embeds heterogeneous single-cell multi-omics datasets into a shared latent space. It can further construct a reference atlas for gene imputation across datasets. Meanwhile, uniPort provides a flexible label transfer framework to deconvolute heterogeneous spatial transcriptomic data using an optimal transport plan, instead of embedding latent space. We demonstrate the capability of uniPort by applying it to integrate a variety of datasets, including single-cell transcriptomics, chromatin accessibility, and spatially resolved transcriptomic data.

Reprogramming cell fates towards novel cancer immunotherapies

Recent advances in our understanding of host immune and cancer cells interactions have made immunotherapy a prominent choice in cancer treatment. Despite such promise, cell-based immunotherapies remain inapplicable to many patients due to severe limitations in the availability and quality of immune cells isolated from donors. Reprogramming technologies that facilitate the engineering of cell types of interest, are emerging as a putative solution to such challenges. Here we focus on the recent progress being made in reprogramming technologies with respect to the immune system and their potential for clinical applications.

Fluorescent Reporters Distinguish Stem Cell Colony Subtypes During Somatic Cell Reprogramming

Somatic cell reprogramming was first developed to create induced pluripotent stem (iPS) cells. Since that time, the highly dynamic and heterogeneous nature of the reprogramming process has come to be appreciated. Remarkably, a distinct type of stem cell, called induced extraembryonic endoderm (iXEN) stem cell, is also formed during reprogramming of mouse somatic cells by ectopic expression of the transcription factors, OCT4, SOX2, KLF4, and MYC (OSKM). The mechanisms leading somatic cells to adopt differing stem cell fates are challenging to resolve given that formation of either stem cell type is slow, stochastic, and rare. For these reasons, fluorescent gene expression reporters have provided an invaluable tool for revealing the path from the somatic state to pluripotency. However, no such reporters have been established for comparable studies of iXEN cell formation. In this study, we examined the expression of multiple fluorescent reporters, including Nanog, Oct4, and the endodermal genes, Gata4 and Gata6-alone and in combination, during reprogramming. We show that only simultaneous evaluation of Nanog and Gata4 reliably distinguishes iPS and iXEN cell colonies during reprogramming.

Canalized gene expression during development mediates caste differentiation in ants

Ant colonies are higher-level organisms consisting of specialized reproductive and non-reproductive individuals that differentiate early in development, similar to germ-soma segregation in bilateral Metazoa. Analogous to diverging cell lines, developmental differentiation of individual ants has often been considered in epigenetic terms but the sets of genes that determine caste phenotypes throughout larval and pupal development remain unknown. Here, we reconstruct the individual developmental trajectories of two ant species, Monomorium pharaonis and Acromyrmex echinatior, after obtaining >1,400 whole-genome transcriptomes. Using a new backward prediction algorithm, we show that caste phenotypes can be accurately predicted by genome-wide transcriptome profiling. We find that caste differentiation is increasingly canalized from early development onwards, particularly in germline individuals (gynes/queens) and that the juvenile hormone signalling pathway plays a key role in this process by regulating body mass divergence between castes. We quantified gene-specific canalization levels and found that canalized genes with gyne/queen-biased expression were enriched for ovary and wing functions while canalized genes with worker-biased expression were enriched in brain and behavioural functions. Suppression in gyne larvae of Freja, a highly canalized gyne-biased ovary gene, disturbed pupal development by inducing non-adaptive intermediate phenotypes between gynes and workers. Our results are consistent with natural selection actively maintaining canalized caste phenotypes while securing robustness in the life cycle ontogeny of ant colonies.

Intermediate cells of in vitro cellular reprogramming and in vivo tissue regeneration require desmoplakin

Amphibians and fish show considerable regeneration potential via dedifferentiation of somatic cells into blastemal cells. In terms of dedifferentiation, in vitro cellular reprogramming has been proposed to share common processes with in vivo tissue regeneration, although the details are elusive. Here, we identified the cytoskeletal linker protein desmoplakin (Dsp) as a common factor mediating both reprogramming and regeneration. Our analysis revealed that Dsp expression is elevated in distinct intermediate cells during in vitro reprogramming. Knockdown of Dsp impedes in vitro reprogramming into induced pluripotent stem cells and induced neural stem/progenitor cells as well as in vivo regeneration of zebrafish fins. Notably, reduced Dsp expression impairs formation of the intermediate cells during cellular reprogramming and tissue regeneration. These findings suggest that there is a Dsp-mediated evolutionary link between cellular reprogramming in mammals and tissue regeneration in lower vertebrates and that the intermediate cells may provide alternative approaches for mammalian regenerative therapy.

Lateral confined growth of cells activates Lef1 dependent pathways to regulate cell-state transitions

Long-term sustained mechano-chemical signals in tissue microenvironment regulate cell-state transitions. In recent work, we showed that laterally confined growth of fibroblasts induce dedifferentiation programs. However, the molecular mechanisms underlying such mechanically induced cell-state transitions are poorly understood. In this paper, we identify Lef1 as a critical somatic transcription factor for the mechanical regulation of de-differentiation pathways. Network optimization methods applied to time-lapse RNA-seq data identify Lef1 dependent signaling as potential regulators of such cell-state transitions. We show that Lef1 knockdown results in the down-regulation of fibroblast de-differentiation and that Lef1 directly interacts with the promoter regions of downstream reprogramming factors. We also evaluate the potential upstream activation pathways of Lef1, including the Smad4, Atf2, NFkB and Beta-catenin pathways, thereby identifying that Smad4 and Atf2 may be critical for Lef1 activation. Collectively, we describe an important mechanotransduction pathway, including Lef1, which upon activation, through progressive lateral cell confinement, results in fibroblast de-differentiation.

Actomyosin contractility as a mechanical checkpoint for cell state transitions

Cell state transitions induced by mechano-chemical cues result in a heterogeneous population of cell states. While much of the work towards understanding the origins of such heterogeneity has focused on the gene regulatory mechanisms, the contribution of intrinsic mechanical properties of cells remains unknown. In this paper, using a well-defined single cell platform to induce cell-state transitions, we reveal the importance of actomyosin contractile forces in regulating the heterogeneous cell-fate decisions. Temporal analysis of laterally confined growth of fibroblasts revealed sequential changes in the colony morphology which was tightly coupled to the progressive erasure of lineage-specific transcription programs. Pseudo-trajectory constructed using unsupervised diffusion analysis of the colony morphology features revealed a bifurcation event in which some cells undergo successful cell state transitions towards partial reprogramming. Importantly, inhibiting actomyosin contractility before the bifurcation event leads to more efficient dedifferentiation. Taken together, this study highlights the presence of mechanical checkpoints that contribute to the heterogeneity in cell state transitions.

Density-based detection of cell transition states to construct disparate and bifurcating trajectories

Tree- and linear-shaped cell differentiation trajectories have been widely observed in developmental biologies and can be also inferred through computational methods from single-cell RNA-sequencing datasets. However, trajectories with complicated topologies such as loops, disparate lineages and bifurcating hierarchy remain difficult to infer accurately. Here, we introduce a density-based trajectory inference method capable of constructing diverse shapes of topological patterns including the most intriguing bifurcations. The novelty of our method is a step to exploit overlapping probability distributions to identify transition states of cells for determining connectability between cell clusters, and another step to infer a stable trajectory through a base-topology guided iterative fitting. Our method precisely re-constructed various benchmark reference trajectories. As a case study to demonstrate practical usefulness, our method was tested on single-cell RNA sequencing profiles of blood cells of SARS-CoV-2-infected patients. We not only re-discovered the linear trajectory bridging the transition from IgM plasmablast cells to developing neutrophils, and also found a previously-undiscovered lineage which can be rigorously supported by differentially expressed gene analysis.

Reconstructing data-driven governing equations for cell phenotypic transitions: integration of data science and systems biology

Cells with the same genome can exist in different phenotypes and can change between distinct phenotypes when subject to specific stimuli and microenvironments. Some examples include cell differentiation during development, reprogramming for induced pluripotent stem cells and transdifferentiation, cancer metastasis and fibrosis progression. The regulation and dynamics of cell phenotypic conversion is a fundamental problem in biology, and has a long history of being studied within the formalism of dynamical systems. A main challenge for mechanism-driven modeling studies is acquiring sufficient amount of quantitative information for constraining model parameters. Advances in quantitative experimental approaches, especially high throughput single-cell techniques, have accelerated the emergence of a new direction for reconstructing the governing dynamical equations of a cellular system from quantitative single-cell data, beyond the dominant statistical approaches. Here I review a selected number of recent studies using live- and fixed-cell data and provide my perspective on future development.

Reconstructing data-driven governing equations for cell phenotypic transitions: integration of data science and systems biology

Cells with the same genome can exist in different phenotypes and can change between distinct phenotypes when subject to specific stimuli and microenvironments. Some examples include cell differentiation during development, reprogramming for induced pluripotent stem cells and transdifferentiation, cancer metastasis and fibrosis progression. The regulation and dynamics of cell phenotypic conversion is a fundamental problem in biology, and has a long history of being studied within the formalism of dynamical systems. A main challenge for mechanism-driven modeling studies is acquiring sufficient amount of quantitative information for constraining model parameters. Advances in quantitative experimental approaches, especially high throughput single-cell techniques, have accelerated the emergence of a new direction for reconstructing the governing dynamical equations of a cellular system from quantitative single-cell data, beyond the dominant statistical approaches. Here I review a selected number of recent studies using live- and fixed-cell data and provide my perspective on future development.

Comparative roadmaps of reprogramming and oncogenic transformation identify Bcl11b and Atoh8 as broad regulators of cellular plasticity

Coordinated changes of cellular plasticity and identity are critical for pluripotent reprogramming and oncogenic transformation. However, the sequences of events that orchestrate these intermingled modifications have never been comparatively dissected. Here, we deconvolute the cellular trajectories of reprogramming (via Oct4/Sox2/Klf4/c-Myc) and transformation (via Ras/c-Myc) at the single-cell resolution and reveal how the two processes intersect before they bifurcate. This approach led us to identify the transcription factor Bcl11b as a broad-range regulator of cell fate changes, as well as a pertinent marker to capture early cellular intermediates that emerge simultaneously during reprogramming and transformation. Multiomics characterization of these intermediates unveiled a c-Myc/Atoh8/Sfrp1 regulatory axis that constrains reprogramming, transformation and transdifferentiation. Mechanistically, we found that Atoh8 restrains cellular plasticity, independent of cellular identity, by binding a specific enhancer network. This study provides insights into the partitioned control of cellular plasticity and identity for both regenerative and cancer biology.

Huyghe, Furlan et al. compare pluripotent reprogramming with oncogenic transformation and identify Bcl11b and Atoh8 as regulators of cellular plasticity in both processes, thus offering a unifying theory on the factors constraining cell fate changes.

Towards reliable quantification of cell state velocities

A few years ago, it was proposed to use the simultaneous quantification of unspliced and spliced messenger RNA (mRNA) to add a temporal dimension to high-throughput snapshots of single cell RNA sequencing data. This concept can yield additional insight into the transcriptional dynamics of the biological systems under study. However, current methods for inferring cell state velocities from such data (known as RNA velocities) are afflicted by several theoretical and computational problems, hindering realistic and reliable velocity estimation. We discuss these issues and propose new solutions for addressing some of the current challenges in consistency of data processing, velocity inference and visualisation. We translate our computational conclusion in two velocity analysis tools: one detailed method κ-velo and one heuristic method eco-velo, each of which uses a different set of assumptions about the data.

Cellular and molecular mechanisms of plasticity in cancer

Cancer cells are plastic - they can assume a wide range of distinct phenotypes. Plasticity is integral to cancer initiation and progression, as well as to the emergence and maintenance of intratumoral heterogeneity. Furthermore, plastic cells can rapidly adapt to and evade therapy, which poses a challenge for effective cancer treatment. As such, targeting plasticity in cancer holds tremendous promise. Yet, the principles governing plasticity in cancer cells remain poorly understood. Here, we provide an overview of the fundamental molecular and cellular mechanisms that underlie plasticity in cancer and in other biological contexts, including development and regeneration. We propose a key role for high-plasticity cell states (HPCSs) as crucial nodes for cell state transitions and enablers of intra-tumoral heterogeneity.

Soft substrates promote direct chemical reprogramming of fibroblasts into neurons

Fibroblasts can be directly reprogrammed via a combination of small molecules to generate induced neurons (iNs), bypassing intermediate stages. This method holds great promise for regenerative medicine; however, it remains inefficient. Recently, studies have suggested that physical cues may improve the direct reprogramming of fibroblasts into neurons, but the underlying mechanisms remain to be further explored, and the physical factors reported to date do not exhibit the full properties of the extracellular matrix (ECM). Previous in vitro studies mainly used rigid polystyrene dishes, while one of the characteristics of the native in-vivo environment of neurons is the soft nature of brain ECM. The reported stiffness of brain tissue is very soft ranging between 100 Pa and 3 kPa, and the effect of substrate stiffness on direct neuronal reprogramming has not been explored. Here, we show for the first time that soft substrates substantially improved the production efficiency and quality of iNs, without needing to co-culture with glial cells during reprogramming, producing more glutamatergic neurons with electrophysiological functions in a shorter time. Transcriptome sequencing indicated that soft substrates might promote glutamatergic neuron reprogramming through integrins, actin cytoskeleton, Hippo signalling pathway, and regulation of mesenchymal-to-epithelial transition, and competing endogenous RNA network analysis provided new targets for neuronal reprogramming. We demonstrated that soft substrates may promote neuronal reprogramming by inhibiting microRNA-615-3p-targeting integrin subunit beta 4. Our findings can aid the development of regenerative therapies and help improve our understanding of neuronal reprogramming. STATEMENT OF SIGNIFICANCE: : First, we have shown that low stiffness promotes direct reprogramming on the basis of small molecule combinations. To the best of our knowledge, this is the first report on this type of method, which may greatly promote the progress of neural reprogramming. Second, we found that miR-615-3p may interact with ITGB4, and the soft substrates may promote neural reprogramming by inhibiting microRNA (miR)-615-3p targeting integrin subunit beta 4 (ITGB4). We are the first to report on this mechanism. Our findings will provide more functional neurons for subsequent basic and clinical research in neurological regenerative medicine, and will help to improve the overall understanding of neural reprogramming. This work also provides new ideas for the design of medical biomaterials for nerve regeneration.

Mapping Phenotypic Plasticity upon the Cancer Cell State Landscape Using Manifold Learning

ABSTRACT

Phenotypic plasticity describes the ability of cancer cells to undergo dynamic, nongenetic cell state changes that amplify cancer heterogeneity to promote metastasis and therapy evasion. Thus, cancer cells occupy a continuous spectrum of phenotypic states connected by trajectories defining dynamic transitions upon a cancer cell state landscape. With technologies proliferating to systematically record molecular mechanisms at single-cell resolution, we illuminate manifold learning techniques as emerging computational tools to effectively model cell state dynamics in a way that mimics our understanding of the cell state landscape. We anticipate that "state-gating" therapies targeting phenotypic plasticity will limit cancer heterogeneity, metastasis, and therapy resistance.

SIGNIFICANCE

Nongenetic mechanisms underlying phenotypic plasticity have emerged as significant drivers of tumor heterogeneity, metastasis, and therapy resistance. Herein, we discuss new experimental and computational techniques to define phenotypic plasticity as a scaffold to guide accelerated progress in uncovering new vulnerabilities for therapeutic exploitation.

The trophectoderm acts as a niche for the inner cell mass through C/EBPα-regulated IL-6 signaling

IL-6 has been shown to be required for somatic cell reprogramming into induced pluripotent stem cells (iPSCs). However, how Il6 expression is regulated and whether it plays a role during embryo development remains unknown. Here, we describe that IL-6 is necessary for C/EBPα-enhanced reprogramming of B cells into iPSCs but not for B cell to macrophage transdifferentiation. C/EBPα overexpression activates both Il6 and Il6ra genes in B cells and in PSCs. In embryo development, Cebpa is enriched in the trophectoderm of blastocysts together with Il6, while Il6ra is mostly expressed in the inner cell mass (ICM). In addition, Il6 expression in blastocysts requires Cebpa. Blastocysts secrete IL-6 and neutralization of the cytokine delays the morula to blastocyst transition. The observed requirement of C/EBPα-regulated IL-6 signaling for pluripotency during somatic cell reprogramming thus recapitulates a physiologic mechanism in which the trophectoderm acts as niche for the ICM through the secretion of IL-6.

•

IL-6 is required for the C/EBPα-enhanced B cell to iPSC reprogramming

•

C/EBPα regulates the IL-6 signaling pathway during pluripotency acquisition

•

A Cebpa-Il6 expression axis is conserved in mouse and human trophectoderm

•

IL-6 signals to the Il6ra-expressing ICM and facilitates blastocyst development

Diverse partial reprogramming strategies restore youthful gene expression and transiently suppress cell identity

Partial pluripotent reprogramming can reverse features of aging in mammalian cells, but the impact on somatic identity and the necessity of individual reprogramming factors remain unknown. Here, we used single-cell genomics to map the identity trajectory induced by partial reprogramming in multiple murine cell types and dissected the influence of each factor by screening all Yamanaka Factor subsets with pooled single-cell screens. We found that partial reprogramming restored youthful expression in adipogenic and mesenchymal stem cells but also temporarily suppressed somatic identity programs. Our pooled screens revealed that many subsets of the Yamanaka Factors both restore youthful expression and suppress somatic identity, but these effects were not tightly entangled. We also found that a multipotent reprogramming strategy inspired by amphibian regeneration restored youthful expression in myogenic cells. Our results suggest that various sets of reprogramming factors can restore youthful expression with varying degrees of somatic identity suppression. A record of this paper's Transparent Peer Review process is included in the supplemental information.

Quantifying information of intracellular signaling: progress with machine learning

Cells convey information about their extracellular environment to their core functional machineries. Studying the capacity of intracellular signaling pathways to transmit information addresses fundamental questions about living systems. Here, we review how information-theoretic approaches have been used to quantify information transmission by signaling pathways that are functionally pleiotropic and subject to molecular stochasticity. We describe how recent advances in machine learning have been leveraged to address the challenges of complex temporal trajectory datasets and how these have contributed to our understanding of how cells employ temporal coding to appropriately adapt to environmental perturbations.

Practical Implementation of Genetics: New Concepts in Immunogenomics to Predict, Prevent, and Diagnose Drug Hypersensitivity

Delayed drug hypersensitivities are CD8⁺ T cell-mediated reactions associated with up to 50% mortality. Human leukocyte antigen (HLA) alleles are known to predispose disease and are specific to drug, reaction, and patient ethnicity. Pretreatment screening is recommended for a handful of the strongest associations to identify and prevent drug use in high-risk patients. However, an incomplete predictive value implicates other HLA-imposed risk factors, and low carriage of many identified HLA-risk alleles combined with the high cost of sequence-based typing has limited economic viability for similar recommendation of screening across drugs and health care systems. For mitigation, an expanding armory of low-cost polymerase chain reaction-based screens is being developed, and HLA-imposed risk factors are being discovered. These include (1) polymorphic variants of metabolic and endoplasmic reticulum aminopeptidase enzymes toward multiallelic screening with increased predictivity; (2) regulation by immune checkpoint inhibitors, enabling detolerized animal models of human disease; and (3) immunodominant T cell receptors (TCR) on clonally expanded CD8⁺ T cells. For the latter, HLA risk-restricted TCR provides immunogenomic strategies and samples from a single patient to identify novel HLA-risk associations in underserved minority populations, tissue-relevant effector biomarkers toward earlier diagnosis and treatment, and HLA-TCR-presented immunogenic structures to aid future drug development.

Challenges and opportunities in the use of transcriptomics characterization for human iPSC-derived BBB models

The blood-brain barrier (BBB) is localized at the brain microvascular endothelial cells. These cells form a tight barrier, limiting the access of cells, pathogens, chemicals, and toxins to the brain due to tight junctions and efflux transporters. As the BBB plays a role in the assessment of neurotoxicity and brain uptake of drugs, human in vitro BBB models are highly needed. They allow to evaluate if compounds could reach the central nervous system across the BBB or can compromise its barrier function. Past decade, multiple induced pluripotent stem cell (iPSC)-derived BBB differentiation protocols emerged. These protocols can be divided in two groups, the one-step protocols, direct differentiation from iPSC to BBB cells, or the two-step protocols, differentiation for iPSC to endothelial (progenitor) cells and further induction of BBB characteristics. While the one-step differentiation protocols display good barrier properties, reports question their endothelial nature and maturation status. Therefore protocol characterization remains important. With transcriptomics becoming cheaper, this may support iPSC-derived model characterization. Because of the constraints in obtaining human brain tissue, good human reference data is scarce and would bear inter-individual variability. Additionally, comparison across studies might be challenging due to variations in sample preparation and analysis. Hopefully, increasing use of transcriptomics will allow in-depth characterization of the current iPSC-BBB models and guide researchers to generate more relevant human BBB models.

psupertime: supervised pseudotime analysis for time-series single-cell RNA-seq data

Motivation

Improvements in single-cell RNA-seq technologies mean that studies measuring multiple experimental conditions, such as time series, have become more common. At present, few computational methods exist to infer time series-specific transcriptome changes, and such studies have therefore typically used unsupervised pseudotime methods. While these methods identify cell subpopulations and the transitions between them, they are not appropriate for identifying the genes that vary coherently along the time series. In addition, the orderings they estimate are based only on the major sources of variation in the data, which may not correspond to the processes related to the time labels.

Results

We introduce psupertime, a supervised pseudotime approach based on a regression model, which explicitly uses time-series labels as input. It identifies genes that vary coherently along a time series, in addition to pseudotime values for individual cells, and a classifier that can be used to estimate labels for new data with unknown or differing labels. We show that psupertime outperforms benchmark classifiers in terms of identifying time-varying genes and provides better individual cell orderings than popular unsupervised pseudotime techniques. psupertime is applicable to any single-cell RNA-seq dataset with sequential labels (e.g. principally time series but also drug dosage and disease progression), derived from either experimental design and provides a fast, interpretable tool for targeted identification of genes varying along with specific biological processes.

Availability and implementation

R package available at github.com/wmacnair/psupertime and code for results reproduction at github.com/wmacnair/psupplementary.

Supplementary information

Supplementary data are available at Bioinformatics online.

Comparative parallel multi-omics analysis during the induction of pluripotent and trophectoderm states

Following fertilization, it is only at the 32-64-cell stage when a clear segregation between cells of the inner cell mass and trophectoderm is observed, suggesting a 'T'-shaped model of specification. Here, we examine whether the acquisition of these two states in vitro, by nuclear reprogramming, share similar dynamics/trajectories. Using a comparative parallel multi-omics analysis (i.e., bulk RNA-seq, scRNA-seq, ATAC-seq, ChIP-seq, RRBS and CNVs) on cells undergoing reprogramming to pluripotency and TSC state we show that each reprogramming system exhibits specific trajectories from the onset of the process, suggesting 'V'-shaped model. We describe in detail the various trajectories toward the two states and illuminate reprogramming stage-specific markers, blockers, facilitators and TSC subpopulations. Finally, we show that while the acquisition of the TSC state involves the silencing of embryonic programs by DNA methylation, during the acquisition of pluripotency these regions are initially defined but retain inactive by the elimination of H3K27ac.

Identification, discrimination and heterogeneity of fibroblasts

Fibroblasts, the principal cell type of connective tissue, secrete extracellular matrix components during tissue development, homeostasis, repair and disease. Despite this crucial role, the identification and distinction of fibroblasts from other cell types are challenging and laden with caveats. Rapid progress in single-cell transcriptomics now yields detailed molecular portraits of fibroblasts and other cell types in our bodies, which complement and enrich classical histological and immunological descriptions, improve cell class definitions and guide further studies on the functional heterogeneity of cell subtypes and states, origins and fates in physiological and pathological processes. In this review, we summarize and discuss recent advances in the understanding of fibroblast identification and heterogeneity and how they discriminate from other cell types.

In this review, the authors look at how recent progress in single-cell transcriptomics complement and enrich the classical, largely morphological, portraits of fibroblasts. The detailed molecular information now available provides new insights into fibroblast identity, heterogeneity and function.

Dynamical landscapes of cell fate decisions

The generation of cellular diversity during development involves differentiating cells transitioning between discrete cell states. In the 1940s, the developmental biologist Conrad Waddington introduced a landscape metaphor to describe this process. The developmental path of a cell was pictured as a ball rolling through a terrain of branching valleys with cell fate decisions represented by the branch points at which the ball decides between one of two available valleys. Here we discuss progress in constructing quantitative dynamical models inspired by this view of cellular differentiation. We describe a framework based on catastrophe theory and dynamical systems methods that provides the foundations for quantitative geometric models of cellular differentiation. These models can be fit to experimental data and used to make quantitative predictions about cellular differentiation. The theory indicates that cell fate decisions can be described by a small number of decision structures, such that there are only two distinct ways in which cells make a binary choice between one of two fates. We discuss the biological relevance of these mechanisms and suggest the approach is broadly applicable for the quantitative analysis of differentiation dynamics and for determining principles of developmental decisions.

Comparison of cell state models derived from single-cell RNA sequencing data: graph versus multi-dimensional space

Single-cell sequencing technologies have revolutionized molecular and cellular biology and stimulated the development of computational tools to analyze the data generated from these technology platforms. However, despite the recent explosion of computational analysis tools, relatively few mathematical models have been developed to utilize these data. Here we compare and contrast two cell state geometries for building mathematical models of cell state-transitions with single-cell RNA-sequencing data with hematopoeisis as a model system; (i) by using partial differential equations on a graph representing intermediate cell states between known cell types, and (ii) by using the equations on a multi-dimensional continuous cell state-space. As an application of our approach, we demonstrate how the calibrated models may be used to mathematically perturb normal hematopoeisis to simulate, predict, and study the emergence of novel cell states during the pathogenesis of acute myeloid leukemia. We particularly focus on comparing the strength and weakness of the graph model and multi-dimensional model.

Efficient Discrete Optimal Transport Algorithm by Accelerated Gradient Descent

Optimal transport (OT) plays an essential role in various areas like machine learning and deep learning. However, computing discrete OT for large scale problems with adequate accuracy and efficiency is highly challenging. Recently, methods based on the Sinkhorn algorithm add an entropy regularizer to the prime problem and obtain a trade off between efficiency and accuracy. In this paper, we propose a novel algorithm based on Nesterov's smoothing technique to further improve the efficiency and accuracy in computing OT. Basically, the non-smooth c-transform of the Kantorovich potential is approximated by the smooth Log-Sum-Exp function, which smooths the original non-smooth Kantorovich dual functional. The smooth Kantorovich functional can be efficiently optimized by a fast proximal gradient method, the fast iterative shrinkage thresholding algorithm (FISTA). Theoretically, the computational complexity of the proposed method is given by O ( n 5 2 log n ∕ ϵ ) , which is lower than current estimation of the Sinkhorn algorithm. Experimentally, compared with the Sinkhorn algorithm, our results demonstrate that the proposed method achieves faster convergence and better accuracy with the same parameter.

BSDE: barycenter single-cell differential expression for case-control studies

MOTIVATION

Single-cell sequencing brings about a revolutionarily high resolution for finding differentially expressed genes (DEGs) by disentangling highly heterogeneous cell tissues. Yet, such analysis is so far mostly focused on comparing between different cell types from the same individual. As single-cell sequencing becomes cheaper and easier to use, an increasing number of datasets from case-control studies are becoming available, which call for new methods for identifying differential expressions between case and control individuals.

RESULTS

To bridge this gap, we propose barycenter single-cell differential expression (BSDE), a nonparametric method for finding DEGs for case-control studies. Through the use of optimal transportation for aggregating distributions and computing their distances, our method overcomes the restrictive parametric assumptions imposed by standard mixed-effect-modeling approaches. Through simulations, we show that BSDE can accurately detect a variety of differential expressions while maintaining the type-I error at a prescribed level. Further, 1345 and 1568 cell type-specific DEGs are identified by BSDE from datasets on pulmonary fibrosis and multiple sclerosis, among which the top findings are supported by previous results from the literature.

AVAILABILITY AND IMPLEMENTATION

R package BSDE is freely available from doi.org/10.5281/zenodo.6332254. For real data analysis with the R package, see doi.org/10.5281/zenodo.6332566. These can also be accessed thorough GitHub at github.com/mqzhanglab/BSDE and github.com/mqzhanglab/BSDE_pipeline. The two single-cell sequencing datasets can be download with UCSC cell browser from cells.ucsc.edu/?ds=ms and cells.ucsc.edu/?ds=lung-pf-control.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Adversarial domain translation networks for integrating large-scale atlas-level single-cell datasets

The rapid emergence of large-scale atlas-level single-cell RNA-seq datasets presents remarkable opportunities for broad and deep biological investigations through integrative analyses. However, harmonizing such datasets requires integration approaches to be not only computationally scalable, but also capable of preserving a wide range of fine-grained cell populations. We have created Portal, a unified framework of adversarial domain translation to learn harmonized representations of datasets. When compared to other state-of-the-art methods, Portal achieves better performance for preserving biological variation during integration, while achieving the integration of millions of cells, in minutes, with low memory consumption. We show that Portal is widely applicable to integrating datasets across different samples, platforms and data types. We also apply Portal to the integration of cross-species datasets with limited shared information among them, elucidating biological insights into the similarities and divergences in the spermatogenesis process among mouse, macaque and human.

Epigenetics, Enhancer Function and 3D Chromatin Organization in Reprogramming to Pluripotency

Genome architecture, epigenetics and enhancer function control the fate and identity of cells. Reprogramming to induced pluripotent stem cells (iPSCs) changes the transcriptional profile and chromatin landscape of the starting somatic cell to that of the pluripotent cell in a stepwise manner. Changes in the regulatory networks are tightly regulated during normal embryonic development to determine cell fate, and similarly need to function in cell fate control during reprogramming. Switching off the somatic program and turning on the pluripotent program involves a dynamic reorganization of the epigenetic landscape, enhancer function, chromatin accessibility and 3D chromatin topology. Within this context, we will review here the current knowledge on the processes that control the establishment and maintenance of pluripotency during somatic cell reprogramming.

Metacell-2: a divide-and-conquer metacell algorithm for scalable scRNA-seq analysis

Scaling scRNA-seq to profile millions of cells is crucial for constructing high-resolution maps of transcriptional manifolds. Current analysis strategies, in particular dimensionality reduction and two-phase clustering, offer only limited scaling and sensitivity to define such manifolds. We introduce Metacell-2, a recursive divide-and-conquer algorithm allowing efficient decomposition of scRNA-seq datasets of any size into small and cohesive groups of cells called metacells. Metacell-2 improves outlier cell detection and rare cell type identification, as shown with human bone marrow cell atlas and mouse embryonic data. Metacell-2 is implemented over the scanpy framework for easy integration in any analysis pipeline.

Chemical reprogramming of human somatic cells to pluripotent stem cells

Cellular reprogramming can manipulate the identity of cells to generate the desired cell types^1-3. The use of cell intrinsic components, including oocyte cytoplasm and transcription factors, can enforce somatic cell reprogramming to pluripotent stem cells^4-7. By contrast, chemical stimulation by exposure to small molecules offers an alternative approach that can manipulate cell fate in a simple and highly controllable manner^8-10. However, human somatic cells are refractory to chemical stimulation owing to their stable epigenome^2,11,12 and reduced plasticity^13,14; it is therefore challenging to induce human pluripotent stem cells by chemical reprogramming. Here we demonstrate, by creating an intermediate plastic state, the chemical reprogramming of human somatic cells to human chemically induced pluripotent stem cells that exhibit key features of embryonic stem cells. The whole chemical reprogramming trajectory analysis delineated the induction of the intermediate plastic state at the early stage, during which chemical-induced dedifferentiation occurred, and this process was similar to the dedifferentiation process that occurs in axolotl limb regeneration. Moreover, we identified the JNK pathway as a major barrier to chemical reprogramming, the inhibition of which was indispensable for inducing cell plasticity and a regeneration-like program by suppressing pro-inflammatory pathways. Our chemical approach provides a platform for the generation and application of human pluripotent stem cells in biomedicine. This study lays foundations for developing regenerative therapeutic strategies that use well-defined chemicals to change cell fates in humans.

Chemical Pretreatment Activated a Plastic State Amenable to Direct Lineage Reprogramming

Somatic cells can be chemically reprogrammed into a pluripotent stem cell (CiPSC) state, mediated by an extraembryonic endoderm- (XEN-) like state. We found that the chemical cocktail applied in CiPSC generation initially activated a plastic state in mouse fibroblasts before transitioning into XEN-like cells. The plastic state was characterized by broadly activated expression of development-associated transcription factors (TFs), such as Sox17, Ascl1, Tbx3, and Nkx6-1, with a more accessible chromatin state indicating an enhanced capability of cell fate conversion. Intriguingly, introducing such a plastic state remarkably improved the efficiency of chemical reprogramming from fibroblasts to functional neuron-like cells with electrophysiological activity or beating skeletal muscles. Furthermore, the generation of chemically induced neuron-like cells or skeletal muscles from mouse fibroblasts was independent of the intermediate XEN-like state or the pluripotency state. In summary, our findings revealed a plastic chemically activated multi-lineage priming (CaMP) state at the onset of chemical reprogramming. This state enhanced the cells’ potential to adapt to other cell fates. It provides a general approach to empowering chemical reprogramming methods to obtain functional cell types bypassing inducing pluripotent stem cells.

TedSim: temporal dynamics simulation of single-cell RNA sequencing data and cell division history

Recently, lineage tracing technology using CRISPR/Cas9 genome editing has enabled simultaneous readouts of gene expressions and lineage barcodes, which allows for the reconstruction of the cell division tree and makes it possible to reconstruct ancestral cell types and trace the origin of each cell type. Meanwhile, trajectory inference methods are widely used to infer cell trajectories and pseudotime in a dynamic process using gene expression data of present-day cells. Here, we present TedSim (single-cell temporal dynamics simulator), which simulates the cell division events from the root cell to present-day cells, simultaneously generating two data modalities for each single cell: the lineage barcode and gene expression data. TedSim is a framework that connects the two problems: lineage tracing and trajectory inference. Using TedSim, we conducted analysis to show that (i) TedSim generates realistic gene expression and barcode data, as well as realistic relationships between these two data modalities; (ii) trajectory inference methods can recover the underlying cell state transition mechanism with balanced cell type compositions; and (iii) integrating gene expression and barcode data can provide more insights into the temporal dynamics in cell differentiation compared to using only one type of data, but better integration methods need to be developed.

Label-Free Imaging to Track Reprogramming of Human Somatic Cells

The process of reprogramming patient samples to human-induced pluripotent stem cells (iPSCs) is stochastic, asynchronous, and inefficient, leading to a heterogeneous population of cells. In this study, we track the reprogramming status of patient-derived erythroid progenitor cells (EPCs) at the single-cell level during reprogramming with label-free live-cell imaging of cellular metabolism and nuclear morphometry to identify high-quality iPSCs. EPCs isolated from human peripheral blood of three donors were used for our proof-of-principle study. We found distinct patterns of autofluorescence lifetime for the reduced form of nicotinamide adenine dinucleotide (phosphate) and flavin adenine dinucleotide during reprogramming. Random forest models classified iPSCs with ∼95% accuracy, which enabled the successful isolation of iPSC lines from reprogramming cultures. Reprogramming trajectories resolved at the single-cell level indicated significant reprogramming heterogeneity along different branches of cell states. This combination of micropatterning, autofluorescence imaging, and machine learning provides a unique, real-time, and nondestructive method to assess the quality of iPSCs in a biomanufacturing process, which could have downstream impacts in regenerative medicine, cell/gene therapy, and disease modeling.

A single-cell Arabidopsis root atlas reveals developmental trajectories in wild-type and cell identity mutants

In all multicellular organisms, transcriptional networks orchestrate organ development. The Arabidopsis root, with its simple structure and indeterminate growth, is an ideal model for investigating the spatiotemporal transcriptional signatures underlying developmental trajectories. To map gene expression dynamics across root cell types and developmental time, we built a comprehensive, organ-scale atlas at single-cell resolution. In addition to estimating developmental progressions in pseudotime, we employed the mathematical concept of optimal transport to infer developmental trajectories and identify their underlying regulators. To demonstrate the utility of the atlas to interpret new datasets, we profiled mutants for two key transcriptional regulators at single-cell resolution, shortroot and scarecrow. We report transcriptomic and in vivo evidence for tissue trans-differentiation underlying a mixed cell identity phenotype in scarecrow. Our results support the atlas as a rich community resource for unraveling the transcriptional programs that specify and maintain cell identity to regulate spatiotemporal organ development.

Visualization, benchmarking and characterization of nested single-cell heterogeneity as dynamic forest mixtures

A major topic of debate in developmental biology centers on whether development is continuous, discontinuous, or a mixture of both. Pseudo-time trajectory models, optimal for visualizing cellular progression, model cell transitions as continuous state manifolds and do not explicitly model real-time, complex, heterogeneous systems and are challenging for benchmarking with temporal models. We present a data-driven framework that addresses these limitations with temporal single-cell data collected at discrete time points as inputs and a mixture of dependent minimum spanning trees (MSTs) as outputs, denoted as dynamic spanning forest mixtures (DSFMix). DSFMix uses decision-tree models to select genes that account for variations in multimodality, skewness and time. The genes are subsequently used to build the forest using tree agglomerative hierarchical clustering and dynamic branch cutting. We first motivate the use of forest-based algorithms compared to single-tree approaches for visualizing and characterizing developmental processes. We next benchmark DSFMix to pseudo-time and temporal approaches in terms of feature selection, time correlation, and network similarity. Finally, we demonstrate how DSFMix can be used to visualize, compare and characterize complex relationships during biological processes such as epithelial-mesenchymal transition, spermatogenesis, stem cell pluripotency, early transcriptional response from hormones and immune response to coronavirus disease. Our results indicate that the expression of genes during normal development exhibits a high proportion of non-uniformly distributed profiles that are mostly right-skewed and multimodal; the latter being a characteristic of major steady states during development. Our study also identifies and validates gene signatures driving complex dynamic processes during somatic or germline differentiation.

Optimal transport improves cell-cell similarity inference in single-cell omics data

MOTIVATION

High-throughput single-cell molecular profiling is revolutionizing biology and medicine by unveiling the diversity of cell types and states contributing to development and disease. The identification and characterization of cellular heterogeneity are typically achieved through unsupervised clustering, which crucially relies on a similarity metric.

RESULTS

We here propose the use of Optimal Transport (OT) as a cell-cell similarity metric for single-cell omics data. OT defines distances to compare high-dimensional data represented as probability distributions. To speed up computations and cope with the high dimensionality of single-cell data, we consider the entropic regularization of the classical OT distance. We then extensively benchmark OT against state-of-the-art metrics over 13 independent datasets, including simulated, scRNA-seq, scATAC-seq and single-cell DNA methylation data. First, we test the ability of the metrics to detect the similarity between cells belonging to the same groups (e.g. cell types, cell lines of origin). Then, we apply unsupervised clustering and test the quality of the resulting clusters. OT is found to improve cell-cell similarity inference and cell clustering in all simulated and real scRNA-seq data, as well as in scATAC-seq and single-cell DNA methylation data.

AVAILABILITY AND IMPLEMENTATION

All our analyses are reproducible through the OT-scOmics Jupyter notebook available at https://github.com/ComputationalSystemsBiology/OT-scOmics.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Spatial components of molecular tissue biology

Methods for profiling RNA and protein expression in a spatially resolved manner are rapidly evolving, making it possible to comprehensively characterize cells and tissues in health and disease. To maximize the biological insights obtained using these techniques, it is critical to both clearly articulate the key biological questions in spatial analysis of tissues and develop the requisite computational tools to address them. Developers of analytical tools need to decide on the intrinsic molecular features of each cell that need to be considered, and how cell shape and morphological features are incorporated into the analysis. Also, optimal ways to compare different tissue samples at various length scales are still being sought. Grouping these biological problems and related computational algorithms into classes across length scales, thus characterizing common issues that need to be addressed, will facilitate further progress in spatial transcriptomics and proteomics.

A single-cell Arabidopsis root atlas reveals developmental trajectories in wild-type and cell identity mutants

In all multicellular organisms, transcriptional networks orchestrate organ development. The Arabidopsis root, with its simple structure and indeterminate growth, is an ideal model to investigate the spatiotemporal transcriptional signatures underlying developmental trajectories. To map gene expression dynamics across root cell types and developmental time, we built a comprehensive, organ-scale atlas at single cell resolution. In addition to estimating developmental progressions in pseudotime, we employed the mathematical concept of optimal transport to infer developmental trajectories and identify their underlying regulators. To demonstrate the utility of the atlas to interpret new datasets, we profiled mutants for two key transcriptional regulators at single cell resolution, shortroot and scarecrow. We report transcriptomic and in vivo evidence for tissue trans-differentiation underlying a mixed cell identity phenotype in scarecrow. Our results support the atlas as a rich community resource for unraveling the transcriptional programs that specify and maintain cell identity to regulate spatiotemporal organ development.

How do transcriptional networks regulate organ development? Using scRNA-seq, Shahan and Hsu et al. produced an Arabidopsis root atlas, revealing gradual gene expression changes underlying differentiation of cell types and candidate regulators of cell fate. The atlas enabled interpretation of smaller scRNA-seq datasets and revealed new phenotypes in developmental mutants.

Temporal modelling using single-cell transcriptomics

Methods for profiling genes at the single-cell level have revolutionized our ability to study several biological processes and systems including development, differentiation, response programmes and disease progression. In many of these studies, cells are profiled over time in order to infer dynamic changes in cell states and types, sets of expressed genes, active pathways and key regulators. However, time-series single-cell RNA sequencing (scRNA-seq) also raises several new analysis and modelling issues. These issues range from determining when and how deep to profile cells, linking cells within and between time points, learning continuous trajectories, and integrating bulk and single-cell data for reconstructing models of dynamic networks. In this Review, we discuss several approaches for the analysis and modelling of time-series scRNA-seq, highlighting their steps, key assumptions, and the types of data and biological questions they are most appropriate for.