A Generalized Gene-Regulatory Network Model of Stem Cell Differentiation for Predicting Lineage Specifiers

CRISPR-GEM: A Novel Machine Learning Model for CRISPR Genetic Target Discovery and Evaluation

CRISPR gene editing strategies are shaping cell therapies through precise and tunable control over gene expression. However, limitations in safely delivering high quantities of CRISPR machinery demand careful target gene selection to achieve reliable therapeutic effects. Informed target gene selection requires a thorough understanding of the involvement of target genes in gene regulatory networks (GRNs) and thus their impact on cell phenotype. Effective decoding of these complex networks has been achieved using machine learning models, but current techniques are limited to single cell types and focus mainly on transcription factors, limiting their applicability to CRISPR strategies. To address this, we present CRISPR-GEM, a multilayer perceptron (MLP) based synthetic GRN constructed to accurately predict the downstream effects of CRISPR gene editing. First, input and output nodes are identified as differentially expressed genes between defined experimental and target cell/tissue types, respectively. Then, MLP training learns regulatory relationships in a black-box approach allowing accurate prediction of output gene expression using only input gene expression. Finally, CRISPR-mimetic perturbations are made to each input gene individually, and the resulting model predictions are compared to those for the target group to score and assess each input gene as a CRISPR candidate. The top scoring genes provided by CRISPR-GEM therefore best modulate experimental group GRNs to motivate transcriptomic shifts toward a target group phenotype. This machine learning model is the first of its kind for predicting optimal CRISPR target genes and serves as a powerful tool for enhanced CRISPR strategies across a range of cell therapies.

The cell rejuvenation atlas: leveraging network biology to identify master regulators of rejuvenation strategies

Current rejuvenation strategies, which range from calorie restriction to in vivo partial reprogramming, only improve a few specific cellular processes. In addition, the molecular mechanisms underlying these approaches are largely unknown, which hinders the design of more holistic cellular rejuvenation strategies. To address this issue, we developed SINGULAR (Single-cell RNA-seq Investigation of Rejuvenation Agents and Longevity), a cell rejuvenation atlas that provides a unified system biology analysis of diverse rejuvenation strategies across multiple organs at single-cell resolution. In particular, we leverage network biology approaches to characterize and compare the effects of each strategy at the level of intracellular signaling, cell-cell communication, and transcriptional regulation. As a result, we identified master regulators orchestrating the rejuvenation response and propose that targeting a combination of them leads to a more holistic improvement of age-dysregulated cellular processes. Thus, the interactive database accompanying SINGULAR is expected to facilitate the future design of synthetic rejuvenation interventions.

Dissecting reversible and irreversible single cell state transitions from gene regulatory networks

Understanding cell state transitions and their governing regulatory mechanisms remains one of the fundamental questions in biology. We develop a computational method, state transition inference using cross-cell correlations (STICCC), for predicting reversible and irreversible cell state transitions at single-cell resolution by using gene expression data and a set of gene regulatory interactions. The method is inspired by the fact that the gene expression time delays between regulators and targets can be exploited to infer past and future gene expression states. From applications to both simulated and experimental single-cell gene expression data, we show that STICCC-inferred vector fields capture basins of attraction and irreversible fluxes. By connecting regulatory information with systems' dynamical behaviors, STICCC reveals how network interactions influence reversible and irreversible state transitions. Compared to existing methods that infer pseudotime and RNA velocity, STICCC provides complementary insights into the gene regulation of cell state transitions.

Expression and Functional Analysis of Immuno-Micro-RNAs mir-146a and mir-326 in Colorectal Cancer

Micro-RNAs (miRNAs) are non-coding RNAs with importance in the development of cancer. They are involved in both tumor development and immune processes in tumors. The present study aims to characterize the behavior of two miRNAs, the proinflammatory miR-326-5p and the anti-inflammatory miR-146a-5p, in colorectal cancer (CRC), to decipher the mechanisms that regulate their expression, and to study potential applications. Tissue levels of miR-326-5p and miR-146a-5p were determined by qrt-PCR (real-time quantitative reverse transcription polymerase chain reaction) in 45 patients with colorectal cancer in tumoral and normal adjacent tissue. Subsequent bioinformatic analysis was performed to characterize the transcriptional networks that control the expression of the two miRNAs. The biomarker potential of miRNAs was assessed. The expression of miR-325-5p and miR-146a-5p was decreased in tumors compared to normal tissue. The two miRNAs are regulated through a transcriptional network, which originates in the inflammatory and proliferative pathways and regulates a set of cellular functions related to immunity, proliferation, and differentiation. The miRNAs coordinate distinct modules in the network. There is good biomarker potential of miR-326 with an AUC (Area under the curve) of 0.827, 0.911 sensitivity (Sn), and 0.689 specificity (Sp), and of the combination miR-326-miR-146a, with an AUC of 0.845, Sn of 0.75, and Sp of 0.89. The miRNAs are downregulated in the tumor tissue. They are regulated by a transcriptional network in which they coordinate distinct modules. The structure of the network highlights possible therapeutic approaches. MiR-326 and the combination of the two miRNAs may serve as biomarkers in CRC.

CRISPR-GEM: A Novel Machine Learning Model for CRISPR Genetic Target Discovery and Evaluation

CRISPR gene editing strategies are shaping cell therapies through precise and tunable control over gene expression. However, achieving reliable therapeutic effects with improved safety and efficacy requires informed target gene selection. This depends on a thorough understanding of the involvement of target genes in gene regulatory networks (GRNs) that regulate cell phenotype and function. Machine learning models have been previously used for GRN reconstruction using RNA-seq data, but current techniques are limited to single cell types and focus mainly on transcription factors. This restriction overlooks many potential CRISPR target genes, such as those encoding extracellular matrix components, growth factors, and signaling molecules, thus limiting the applicability of these models for CRISPR strategies. To address these limitations, we have developed CRISPR-GEM, a multi-layer perceptron (MLP)-based synthetic GRN constructed to accurately predict the downstream effects of CRISPR gene editing. First, input and output nodes are identified as differentially expressed genes between defined experimental and target cell/tissue types respectively. Then, MLP training learns regulatory relationships in a black-box approach allowing accurate prediction of output gene expression using only input gene expression. Finally, CRISPR-mimetic perturbations are made to each input gene individually and the resulting model predictions are compared to those for the target group to score and assess each input gene as a CRISPR candidate. The top scoring genes provided by CRISPR-GEM therefore best modulate experimental group GRNs to motivate transcriptomic shifts towards a target group phenotype. This machine learning model is the first of its kind for predicting optimal CRISPR target genes and serves as a powerful tool for enhanced CRISPR strategies across a range of cell therapies.

BMP4-SMAD1/5/9-RUNX2 pathway activation inhibits neurogenesis and oligodendrogenesis in Alzheimer's patients' iPSCs in senescence-related conditions

In addition to increasing β-amyloid plaque deposition and tau tangle formation, inhibition of neurogenesis has recently been observed in Alzheimer's disease (AD). This study generated a cellular model that recapitulated neurogenesis defects observed in patients with AD, using induced pluripotent stem cell lines derived from sporadic and familial AD (AD iPSCs). AD iPSCs exhibited impaired neuron and oligodendrocyte generation when expression of several senescence markers was induced. Compound screening using these cellular models identified three drugs able to restore neurogenesis, and extensive morphological quantification revealed cell-line- and drug-type-dependent neuronal generation. We also found involvement of elevated Sma- and Mad-related protein 1/5/9 (SMAD1/5/9) phosphorylation and greater Runt-related transcription factor 2 (RUNX2) expression in neurogenesis defects in AD. Moreover, BMP4 was elevated in AD iPSC medium during neural differentiation and cerebrospinal fluid of patients with AD, suggesting a BMP4-SMAD1/5/9-RUNX2 signaling pathway contribution to neurogenesis defects in AD under senescence-related conditions.

A Group of Tumor-Suppressive micro-RNAs Changes Expression Coordinately in Colon Cancer

MicroRNAs (miRNAs) are molecules with a role in the post-transcriptional regulation of messenger RNA, being involved in a wide range of biological and pathological processes. In the present study, we aim to characterize the behavior of a few miRNAs with roles in the cell cycle and differentiation of colon cancer (CC) cells. The present work considers miRNAs as reflections of the complex cellular processes in which they are generated, their observed variations being used to characterize the molecular networks in which they are part and through which cell proliferation is achieved. Tumoral and adjacent normal tissue samples were obtained from 40 CC patients, and the expression of miR-29a, miR-146a, miR-215 and miR-449 were determined by qRT-PCR analysis. Subsequent bioinformatic analysis was performed to highlight the transcription factors (TFs) network that regulate the miRNAs and functionally characterizes this network. There was a significant decrease in the expression of all miRNAs in tumor tissue. All miRNAs were positively correlated with each other. The analysis of the TF network showed tightly connected functional modules related to the cell cycle and associated processes. The four miRNAs are downregulated in CC; they are strongly correlated, showing coherence within the cellular network that regulates them and highlighting possible approach strategies.

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.

Probabilistic boolean networks predict transcription factor targets to induce transdifferentiation

We developed a computational approach to find the best intervention to achieve transcription factor (TF) mediated transdifferentiation. We construct probabilistic Boolean networks (PBNs) from single-cell RNA sequencing data of two different cell states to model hematopoietic transcription factors cross-talk. This was achieved by a “sampled network” approach, which enabled us to construct large networks. The interventions to induce transdifferentiation consisted of permanently activating or deactivating each of the TFs and determining the probability mass transfer of steady-state probabilities from the departure to the destination cell type or state. Our findings support the common assumption that TFs that are differentially expressed between the two cell types are the best intervention points to achieve transdifferentiation. TFs whose interventions are found to transdifferentiate progenitor B cells into monocytes include EBF1 down-regulation, CEBPB up-regulation, TCF3 down-regulation, and STAT3 up-regulation.

•

Differentially expressed transcription factors are the best for transdifferentiation

•

Probabilistic Boolean networks (PBNs) are used to model transdifferentiation using the scRNAseq data at one time point

•

A new approach works for a large number of network nodes

Computational approaches for direct cell reprogramming: from the bulk omics era to the single cell era

Recent advances in direct cell reprogramming have made possible the conversion of one cell type to another cell type, offering a potential cell-based treatment to many major diseases. Despite much attention, substantial roadblocks remain including the inefficiency in the proportion of reprogrammed cells of current experiments, and the requirement of a significant amount of time and resources. To this end, several computational algorithms have been developed with the goal of guiding the hypotheses to be experimentally validated. These approaches can be broadly categorized into two main types: transcription factor identification methods which aim to identify candidate transcription factors for a desired cell conversion, and transcription factor perturbation methods which aim to simulate the effect of a transcription factor perturbation on a cell state. The transcription factor perturbation methods can be broken down into Boolean networks, dynamical systems and regression models. We summarize the contributions and limitations of each method and discuss the innovation that single cell technologies are bringing to these approaches and we provide a perspective on the future direction of this field.

scREMOTE: Using multimodal single cell data to predict regulatory gene relationships and to build a computational cell reprogramming model

Cell reprogramming offers a potential treatment to many diseases, by regenerating specialized somatic cells. Despite decades of research, discovering the transcription factors that promote cell reprogramming has largely been accomplished through trial and error, a time-consuming and costly method. A computational model for cell reprogramming, however, could guide the hypothesis formulation and experimental validation, to efficiently utilize time and resources. Current methods often cannot account for the heterogeneity observed in cell reprogramming, or they only make short-term predictions, without modelling the entire reprogramming process. Here, we present scREMOTE, a novel computational model for cell reprogramming that leverages single cell multiomics data, enabling a more holistic view of the regulatory mechanisms at cellular resolution. This is achieved by first identifying the regulatory potential of each transcription factor and gene to uncover regulatory relationships, then a regression model is built to estimate the effect of transcription factor perturbations. We show that scREMOTE successfully predicts the long-term effect of overexpressing two key transcription factors in hair follicle development by capturing higher-order gene regulations. Together, this demonstrates that integrating the multimodal processes governing gene regulation creates a more accurate model for cell reprogramming with significant potential to accelerate research in regenerative medicine.

Computational Methods to Identify Cell-Fate Determinants, Identity Transcription Factors, and Niche-Induced Signaling Pathways for Stem Cell Research

The large-scale development of high-throughput sequencing technologies has not only allowed the generation of reliable omics data related to various regulatory layers but also the development of novel computational models in the field of stem cell research. These computational approaches have enabled the disentangling of a complex interplay between these interrelated layers of regulation by interpreting large quantities of biomedical data in a systematic way. In the context of stem cell research, network modeling of complex gene-gene interactions has been successfully used for understanding the mechanisms underlying stem cell differentiation and cellular conversion. Notably, it has proven helpful for predicting cell-fate determinants and signaling molecules controlling such processes. This chapter will provide an overview of various computational approaches that rely on single-cell and/or bulk RNA sequencing data for elucidating the molecular underpinnings of cell subpopulation identities, lineage specification, and the process of cell-fate decisions. Furthermore, we discuss how these computational methods provide the right framework for computational modeling of biological systems in order to address long-standing challenges in the stem cell field by guiding experimental efforts in stem cell research and regenerative medicine.

Long Non-Coding RNA Lacuna Regulates Neuronal Differentiation of Neural Stem Cells During Brain Development

Although long non-coding RNAs (lncRNAs) is one of the most abundant classes of RNAs encoded within the mammalian genome and are highly expressed in the adult brain, they remain poorly characterized and their roles in the brain development are not well understood. Here we identify the lncRNA Lacuna (also catalogued as NONMMUT071331.2 in NONCODE database) as a negative regulator of neuronal differentiation in the neural stem/progenitor cells (NSCs) during mouse brain development. In particular, we show that Lacuna is transcribed from a genomic locus near to the Tbr2/Eomes gene, a key player in the transition of intermediate progenitor cells towards the induction of neuronal differentiation. Lacuna RNA expression peaks at the developmental time window between E14.5 and E16.5, consistent with a role in neural differentiation. Overexpression experiments in ex vivo cultured NSCs from murine cortex suggest that Lacuna is sufficient to inhibit neuronal differentiation, induce the number of Nestin+ and Olig2+ cells, without affecting proliferation or apoptosis of NSCs. CRISPR/dCas9-KRAB mediated knockdown of Lacuna gene expression leads to the opposite phenotype by inducing neuronal differentiation and suppressing Nestin+ and Olig2+ cells, again without any effect on proliferation or apoptosis of NSCs. Interestingly, despite the negative action of Lacuna on neurogenesis, its knockdown inhibits Eomes transcription, implying a simultaneous, but opposite, role in facilitating the Eomes gene expression. Collectively, our observations indicate a critical function of Lacuna in the gene regulation networks that fine tune the neuronal differentiation in the mammalian NSCs.

Noise distorts the epigenetic landscape and shapes cell-fate decisions

The Waddington epigenetic landscape has become an iconic representation of the cellular differentiation process. Recent single-cell transcriptomic data provide new opportunities for quantifying this originally conceptual tool, offering insight into the gene regulatory networks underlying cellular development. While many methods for constructing the landscape have been proposed, by far the most commonly employed approach is based on computing the landscape as the negative logarithm of the steady-state probability distribution. Here, we use simple models to highlight the complexities and limitations that arise when reconstructing the potential landscape in the presence of stochastic fluctuations. We consider how the landscape changes in accordance with different stochastic systems and show that it is the subtle interplay between the deterministic and stochastic components of the system that ultimately shapes the landscape. We further discuss how the presence of noise has important implications for the identifiability of the regulatory dynamics from experimental data. A record of this paper's transparent peer review process is included in the supplemental information.

Proneural genes define ground-state rules to regulate neurogenic patterning and cortical folding

Asymmetric neuronal expansion is thought to drive evolutionary transitions between lissencephalic and gyrencephalic cerebral cortices. We report that Neurog2 and Ascl1 proneural genes together sustain neurogenic continuity and lissencephaly in rodent cortices. Using transgenic reporter mice and human cerebral organoids, we found that Neurog2 and Ascl1 expression defines a continuum of four lineage-biased neural progenitor cell (NPC) pools. Double⁺ NPCs, at the hierarchical apex, are least lineage restricted due to Neurog2-Ascl1 cross-repression and display unique features of multipotency (more open chromatin, complex gene regulatory network, G₂ pausing). Strikingly, selectively eliminating double⁺ NPCs by crossing Neurog2-Ascl1 split-Cre mice with diphtheria toxin-dependent "deleter" strains locally disrupts Notch signaling, perturbs neurogenic symmetry, and triggers cortical folding. In support of our discovery that double⁺ NPCs are Notch-ligand-expressing "niche" cells that control neurogenic periodicity and cortical folding, NEUROG2, ASCL1, and HES1 transcript distribution is modular (adjacent high/low zones) in gyrencephalic macaque cortices, prefiguring future folds.

RGC-32 Acts as a Hub to Regulate the Transcriptomic Changes Associated With Astrocyte Development and Reactive Astrocytosis

Response Gene to Complement 32 (RGC-32) is an important mediator of the TGF-β signaling pathway, and an increasing amount of evidence implicates this protein in regulating astrocyte biology. We showed recently that spinal cord astrocytes in mice lacking RGC-32 display an immature phenotype reminiscent of progenitors and radial glia, with an overall elongated morphology, increased proliferative capacity, and increased expression of progenitor markers when compared to their wild-type (WT) counterparts that make them incapable of undergoing reactive changes during the acute phase of experimental autoimmune encephalomyelitis (EAE). Here, in order to decipher the molecular networks underlying RGC-32's ability to regulate astrocytic maturation and reactivity, we performed next-generation sequencing of RNA from WT and RGC-32 knockout (KO) neonatal mouse brain astrocytes, either unstimulated or stimulated with the pleiotropic cytokine TGF-β. Pathway enrichment analysis showed that RGC-32 is critical for the TGF-β-induced up-regulation of transcripts encoding proteins involved in brain development and tissue remodeling, such as axonal guidance molecules, transcription factors, extracellular matrix (ECM)-related proteins, and proteoglycans. Our next-generation sequencing of RNA analysis also demonstrated that a lack of RGC-32 results in a significant induction of WD repeat and FYVE domain-containing protein 1 (Wdfy1) and stanniocalcin-1 (Stc1). Immunohistochemical analysis of spinal cords isolated from normal adult mice and mice with EAE at the peak of disease showed that RGC-32 is necessary for the in vivo expression of ephrin receptor type A7 in reactive astrocytes, and that the lack of RGC-32 results in a higher number of homeodomain-only protein homeobox (HOPX)+ and CD133+ radial glia cells. Collectively, these findings suggest that RGC-32 plays a major role in modulating the transcriptomic changes in astrocytes that ultimately lead to molecular programs involved in astrocytic differentiation and reactive changes during neuroinflammation.

Exact Probability Landscapes of Stochastic Phenotype Switching in Feed-Forward Loops: Phase Diagrams of Multimodality

Feed-forward loops (FFLs) are among the most ubiquitously found motifs of reaction networks in nature. However, little is known about their stochastic behavior and the variety of network phenotypes they can exhibit. In this study, we provide full characterizations of the properties of stochastic multimodality of FFLs, and how switching between different network phenotypes are controlled. We have computed the exact steady-state probability landscapes of all eight types of coherent and incoherent FFLs using the finite-butter Accurate Chemical Master Equation (ACME) algorithm, and quantified the exact topological features of their high-dimensional probability landscapes using persistent homology. Through analysis of the degree of multimodality for each of a set of 10,812 probability landscapes, where each landscape resides over 10⁵–10⁶ microstates, we have constructed comprehensive phase diagrams of all relevant behavior of FFL multimodality over broad ranges of input and regulation intensities, as well as different regimes of promoter binding dynamics. In addition, we have quantified the topological sensitivity of the multimodality of the landscapes to regulation intensities. Our results show that with slow binding and unbinding dynamics of transcription factor to promoter, FFLs exhibit strong stochastic behavior that is very different from what would be inferred from deterministic models. In addition, input intensity play major roles in the phenotypes of FFLs: At weak input intensity, FFL exhibit monomodality, but strong input intensity may result in up to 6 stable phenotypes. Furthermore, we found that gene duplication can enlarge stable regions of specific multimodalities and enrich the phenotypic diversity of FFL networks, providing means for cells toward better adaptation to changing environment. Our results are directly applicable to analysis of behavior of FFLs in biological processes such as stem cell differentiation and for design of synthetic networks when certain phenotypic behavior is desired.

Transcription Factor Reprogramming in the Inner Ear: Turning on Cell Fate Switches to Regenerate Sensory Hair Cells

Non-mammalian vertebrates can restore their auditory and vestibular hair cells naturally by triggering the regeneration of adjacent supporting cells. The transcription factor ATOH1 is a key regulator of hair cell development and regeneration in the inner ear. Following the death of hair cells, supporting cells upregulate ATOH1 and give rise to new hair cells. However, in the mature mammalian cochlea, such natural regeneration of hair cells is largely absent. Transcription factor reprogramming has been used in many tissues to convert one cell type into another, with the long-term hope of achieving tissue regeneration. Reprogramming transcription factors work by altering the transcriptomic and epigenetic landscapes in a target cell, resulting in a fate change to the desired cell type. Several studies have shown that ATOH1 is capable of reprogramming cochlear non-sensory tissue into cells resembling hair cells in young animals. However, the reprogramming ability of ATOH1 is lost with age, implying that the potency of individual hair cell-specific transcription factors may be reduced or lost over time by mechanisms that are still not clear. To circumvent this, combinations of key hair cell transcription factors have been used to promote hair cell regeneration in older animals. In this review, we summarize recent findings that have identified and studied these reprogramming factor combinations for hair cell regeneration. Finally, we discuss the important questions that emerge from these findings, particularly the feasibility of therapeutic strategies using reprogramming factors to restore human hearing in the future.

Stability and Robustness of Unbalanced Genetic Toggle Switches in the Presence of Scarce Resources

While the vision of synthetic biology is to create complex genetic systems in a rational fashion, system-level behaviors are often perplexing due to the context-dependent dynamics of modules. One major source of context-dependence emerges due to the limited availability of shared resources, coupling the behavior of disconnected components. Motivated by the ubiquitous role of toggle switches in genetic circuits ranging from controlling cell fate differentiation to optimizing cellular performance, here we reveal how their fundamental dynamic properties are affected by competition for scarce resources. Combining a mechanistic model with nullcline-based stability analysis and potential landscape-based robustness analysis, we uncover not only the detrimental impacts of resource competition, but also how the unbalancedness of the switch further exacerbates them. While in general both of these factors undermine the performance of the switch (by pushing the dynamics toward monostability and increased sensitivity to noise), we also demonstrate that some of the unwanted effects can be alleviated by strategically optimized resource competition. Our results provide explicit guidelines for the context-aware rational design of toggle switches to mitigate our reliance on lengthy and expensive trial-and-error processes, and can be seamlessly integrated into the computer-aided synthesis of complex genetic systems.

New Insights Into the Intricacies of Proneural Gene Regulation in the Embryonic and Adult Cerebral Cortex

Historically, the mammalian brain was thought to lack stem cells as no new neurons were found to be made in adulthood. That dogma changed ∼25 years ago with the identification of neural stem cells (NSCs) in the adult rodent forebrain. However, unlike rapidly self-renewing mature tissues (e.g., blood, intestinal crypts, skin), the majority of adult NSCs are quiescent, and those that become 'activated' are restricted to a few neurogenic zones that repopulate specific brain regions. Conversely, embryonic NSCs are actively proliferating and neurogenic. Investigations into the molecular control of the quiescence-to-proliferation-to-differentiation continuum in the embryonic and adult brain have identified proneural genes encoding basic-helix-loop-helix (bHLH) transcription factors (TFs) as critical regulators. These bHLH TFs initiate genetic programs that remove NSCs from quiescence and drive daughter neural progenitor cells (NPCs) to differentiate into specific neural cell subtypes, thereby contributing to the enormous cellular diversity of the adult brain. However, new insights have revealed that proneural gene activities are context-dependent and tightly regulated. Here we review how proneural bHLH TFs are regulated, with a focus on the murine cerebral cortex, drawing parallels where appropriate to other organisms and neural tissues. We discuss upstream regulatory events, post-translational modifications (phosphorylation, ubiquitinylation), protein-protein interactions, epigenetic and metabolic mechanisms that govern bHLH TF expression, stability, localization, and consequent transactivation of downstream target genes. These tight regulatory controls help to explain paradoxical findings of changes to bHLH activity in different cellular contexts.

Directed differentiation and direct reprogramming: Applying stem cell technologies to hearing research

Hearing loss is the most widely spread sensory disorder in our society. In the majority of cases, it is caused by the loss or malfunctioning of cells in the cochlea: the mechanosensory hair cells, which act as primary sound receptors, and the connecting auditory neurons of the spiral ganglion, which relay the signal to upper brain centers. In contrast to other vertebrates, where damage to the hearing organ can be repaired through the activity of resident cells, acting as tissue progenitors, in mammals, sensory cell damage or loss is irreversible. The understanding of gene and cellular functions, through analysis of different animal models, has helped to identify causes of disease and possible targets for hearing restoration. Translation of these findings to novel therapeutics is, however, hindered by the lack of cellular assays, based on human sensory cells, to evaluate the conservation of molecular pathways across species and the efficacy of novel therapeutic strategies. In the last decade, stem cell technologies enabled to generate human sensory cell types in vitro, providing novel tools to study human inner ear biology, model disease, and validate therapeutics. This review focuses specifically on two technologies: directed differentiation of pluripotent stem cells and direct reprogramming of somatic cell types to sensory hair cells and neurons. Recent development in the field are discussed as well as how these tools could be implemented to become routinely adopted experimental models for hearing research.

Machine intelligence for nerve conduit design and production

Nerve guidance conduits (NGCs) have emerged from recent advances within tissue engineering as a promising alternative to autografts for peripheral nerve repair. NGCs are tubular structures with engineered biomaterials, which guide axonal regeneration from the injured proximal nerve to the distal stump. NGC design can synergistically combine multiple properties to enhance proliferation of stem and neuronal cells, improve nerve migration, attenuate inflammation and reduce scar tissue formation. The aim of most laboratories fabricating NGCs is the development of an automated process that incorporates patient-specific features and complex tissue blueprints (e.g. neurovascular conduit) that serve as the basis for more complicated muscular and skin grafts. One of the major limitations for tissue engineering is lack of guidance for generating tissue blueprints and the absence of streamlined manufacturing processes. With the rapid expansion of machine intelligence, high dimensional image analysis, and computational scaffold design, optimized tissue templates for 3D bioprinting (3DBP) are feasible. In this review, we examine the translational challenges to peripheral nerve regeneration and where machine intelligence can innovate bottlenecks in neural tissue engineering.

The Importance of Computational Modeling in Stem Cell Research

The generation of large amounts of omics data is increasingly enabling not only the processing and analysis of large data sets but also the development of computational models in the field of stem cell research. Although computational models have been proposed in recent decades, we believe that the stem cell community is not fully aware of the potentiality of computational modeling in guiding their experimental research. In this regard, we discuss how single-cell technologies provide the right framework for computational modeling at different scales of biological organization in order to address challenges in the stem cell field and to guide experimentalists in the design of new strategies for stem cell therapies and treatment of congenital disorders.

Intracellular Energy Variability Modulates Cellular Decision-Making Capacity

Cells generate phenotypic diversity both during development and in response to stressful and changing environments, aiding survival. Functionally vital cell fate decisions from a range of phenotypic choices are made by regulatory networks, the dynamics of which rely on gene expression and hence depend on the cellular energy budget (and particularly ATP levels). However, despite pronounced cell-to-cell ATP differences observed across biological systems, the influence of energy availability on regulatory network dynamics is often overlooked as a cellular decision-making modulator, limiting our knowledge of how energy budgets affect cell behaviour. Here, we consider a mathematical model of a highly generalisable, ATP-dependent, decision-making regulatory network, and show that cell-to-cell ATP variability changes the sets of decisions a cell can make. Our model shows that increasing intracellular energy levels can increase the number of supported stable phenotypes, corresponding to increased decision-making capacity. Model cells with sub-threshold intracellular energy are limited to a singular phenotype, forcing the adoption of a specific cell fate. We suggest that energetic differences between cells may be an important consideration to help explain observed variability in cellular decision-making across biological systems.

Identification and Classification of Hubs in microRNA Target Gene Networks in Human Neural Stem/Progenitor Cells following Japanese Encephalitis Virus Infection

RNA viruses are known to modulate host microRNA (miRNA) machinery for their own benefit. Japanese encephalitis virus (JEV), a neurotropic RNA virus, has been reported to manipulate several miRNAs in neurons or microglia. However, no report indicates a complete sketch of the miRNA profile of neural stem/progenitor cells (NSPCs), hence the focus of our current study. We used an miRNA array of 84 miRNAs in uninfected and JEV-infected human neuronal progenitor cells and primary neural precursor cells isolated from aborted fetuses. Severalfold downregulation of hsa-miR-9-5p, hsa-miR-22-3p, hsa-miR-124-3p, and hsa-miR-132-3p was found postinfection in both of the cell types compared to the uninfected cells. Subsequently, we screened for the target genes of these miRNAs and looked for the biological pathways that were significantly regulated by the genes. The target genes involved in two or more pathways were sorted out. Protein-protein interaction (PPI) networks of the miRNA target genes were formed based on their interaction patterns. A binary adjacency matrix for each gene network was prepared. Different modules or communities were identified in those networks by community detection algorithms. Mathematically, we identified the hub genes by analyzing their degree centrality and participation coefficient in the network. The hub genes were classified as either provincial (P < 0.4) or connector (P > 0.4) hubs. We validated the expression of hub genes in both cell line and primary cells through qRT-PCR after JEV infection and respective miR mimic transfection. Taken together, our findings highlight the importance of specific target gene networks of miRNAs affected by JEV infection in NSPCs.IMPORTANCE JEV damages the neural stem/progenitor cell population of the mammalian brain. However, JEV-induced alteration in the miRNA expression pattern of the cell population remains an open question, hence warranting our present study. In this study, we specifically address the downregulation of four miRNAs, and we prepared a protein-protein interaction network of miRNA target genes. We identified two types of hub genes in the PPI network, namely, connector hubs and provincial hubs. These two types of miRNA target hub genes critically influence the participation strength in the networks and thereby significantly impact up- and downregulation in several key biological pathways. Computational analysis of the PPI networks identifies key protein interactions and hubs in those modules, which opens up the possibility of precise identification and classification of host factors for viral infection in NSPCs.

Experimental and Computational Approaches to Direct Cell Reprogramming: Recent Advancement and Future Challenges

The process of direct cell reprogramming, also named transdifferentiation, permits for the conversion of one mature cell type directly into another, without returning to a dedifferentiated state. This makes direct reprogramming a promising approach for the development of several cellular and tissue engineering therapies. To achieve the change in the cell identity, direct reprogramming requires an arsenal of tools that combine experimental and computational techniques. In the recent years, several methods of transdifferentiation have been developed. In this review, we will introduce the concept of direct cell reprogramming and its background, and cover the recent developments in the experimental and computational prediction techniques with their applications. We also discuss the challenges of translating this technology to clinical setting, accompanied with potential solutions.

Single-cell analysis of cardiogenesis reveals basis for organ-level developmental defects

Organogenesis involves integration of myriad cell types, and dysregulation of cellular gene networks results in birth defects, affecting 5 per cent of live births. Congenital heart defects (CHD) are the most common malformations and result from disruption of discrete subsets of cardiac progenitor cells¹, yet the transcriptional changes in individual progenitors that lead to organ-level defects remain unknown. Here, we employed single-cell RNA sequencing (scRNA-seq) to interrogate early cardiac progenitor cells as they become specified during normal and abnormal cardiogenesis, revealing how dysregulation of specific cellular sub-populations has catastrophic consequences. A network-based computational method for scRNA-seq that predicts lineage-specifying transcription factors^2,3 identified Hand2 as a specifier of outflow tract cells but not right ventricular cells, despite failure of right ventricular formation in Hand2-null mice⁴. Temporal single-cell transcriptome analysis of Hand2-null embryos revealed failure of outflow tract myocardium specification, whereas right ventricular myocardium was specified but failed to properly differentiate and migrate. Loss of Hand2 also led to dysregulation of retinoic acid signaling and disruption of anterior-posterior patterning of cardiac progenitors. This work reveals transcriptional determinants that specify fate and differentiation in individual cardiac progenitor cells, and exposes mechanisms of disrupted cardiac development at single-cell resolution, providing a framework to investigate congenital heart defects.

Cell Reprogramming: The Many Roads to Success

Cellular reprogramming experiments from somatic cell types have demonstrated the plasticity of terminally differentiated cell states. Recent efforts in understanding the mechanisms of cellular reprogramming have begun to elucidate the differentiation trajectories along the reprogramming processes. In this review, we focus mainly on direct reprogramming strategies by transcription factors and highlight the variables that contribute to cell fate conversion outcomes. We review key studies that shed light on the cellular and molecular mechanisms by investigating differentiation trajectories and alternative cell states as well as transcription factor regulatory activities during cell fate reprogramming. Finally, we highlight a few concepts that we believe require attention, particularly when measuring the success of cell reprogramming experiments.

Stage-Specific Transcription Factors Drive Astrogliogenesis by Remodeling Gene Regulatory Landscapes

A broad molecular framework of how neural stem cells are specified toward astrocyte fate during brain development has proven elusive. Here we perform comprehensive and integrated transcriptomic and epigenomic analyses to delineate gene regulatory programs that drive the developmental trajectory from mouse embryonic stem cells to astrocytes. We report molecularly distinct phases of astrogliogenesis that exhibit stage- and lineage-specific transcriptomic and epigenetic signatures with unique primed and active chromatin regions, thereby revealing regulatory elements and transcriptional programs underlying astrocyte generation and maturation. By searching for transcription factors that function at these elements, we identified NFIA and ATF3 as drivers of astrocyte differentiation from neural precursor cells while RUNX2 promotes astrocyte maturation. These transcription factors facilitate stage-specific gene expression programs by switching the chromatin state of their target regulatory elements from primed to active. Altogether, these findings provide integrated insights into the genetic and epigenetic mechanisms steering the trajectory of astrogliogenesis.

SeesawPred: A Web Application for Predicting Cell-fate Determinants in Cell Differentiation

Cellular differentiation is a complex process where a less specialized cell evolves into a more specialized cell. Despite the increasing research effort, identification of cell-fate determinants (transcription factors (TFs) determining cell fates during differentiation) still remains a challenge, especially when closely related cell types from a common progenitor are considered. Here, we develop SeesawPred, a web application that, based on a gene regulatory network (GRN) model of cell differentiation, can computationally predict cell-fate determinants from transcriptomics data. Unlike previous approaches, it allows the user to upload gene expression data and does not rely on pre-compiled reference data sets, enabling its application to novel differentiation systems. SeesawPred correctly predicted known cell-fate determinants on various cell differentiation examples in both mouse and human, and also performed better compared to state-of-the-art methods. The application is freely available for academic, non-profit use at http://seesaw.lcsb.uni.lu.

Computational approaches for predicting key transcription factors in targeted cell reprogramming (Review)

There is a need for specific cell types in regenerative medicine and biological research. Frequently, specific cell types may not be easily obtained or the quantity obtained is insufficient for study. Therefore, reprogramming by the direct conversion (transdifferentiation) or re‑induction of induced pluripotent stem cells has been used to obtain cells expressing similar profiles to those of the desired types. Therefore, a specific cocktail of transcription factors (TFs) is required for induction. Nevertheless, identifying the correct combination of TFs is difficult. Although certain computational approaches have been proposed for this task, their methods are complex, and corresponding implementations are difficult to use and generalize for specific source or target cell types. In the present review four computational approaches that have been proposed to obtain likely TFs were compared and discussed. A simplified view of the computational complexity of these methods is provided that consists of three basic ideas: i) The definition of target and non‑target cell types; ii) the estimation of candidate TFs; and iii) filtering candidates. This simplified view was validated by analyzing a well‑documented cardiomyocyte differentiation. Subsequently, these reviewed methods were compared when applied to an unknown differentiation of corneal endothelial cells. The generated results may provide important insights for laboratory assays. Data and computer scripts that may assist with direct conversions in other cell types are also provided.

A stochastic and dynamical view of pluripotency in mouse embryonic stem cells

Pluripotent embryonic stem cells are of paramount importance for biomedical sciences because of their innate ability for self-renewal and differentiation into all major cell lines. The fateful decision to exit or remain in the pluripotent state is regulated by complex genetic regulatory networks. The rapid growth of single-cell sequencing data has greatly stimulated applications of statistical and machine learning methods for inferring topologies of pluripotency regulating genetic networks. The inferred network topologies, however, often only encode Boolean information while remaining silent about the roles of dynamics and molecular stochasticity inherent in gene expression. Herein we develop a framework for systematically extending Boolean-level network topologies into higher resolution models of networks which explicitly account for the promoter architectures and gene state switching dynamics. We show the framework to be useful for disentangling the various contributions that gene switching, external signaling, and network topology make to the global heterogeneity and dynamics of transcription factor populations. We find the pluripotent state of the network to be a steady state which is robust to global variations of gene switching rates which we argue are a good proxy for epigenetic states of individual promoters. The temporal dynamics of exiting the pluripotent state, on the other hand, is significantly influenced by the rates of genetic switching which makes cells more responsive to changes in extracellular signals.

Transcriptional and Genomic Targets of Neural Stem Cells for Functional Recovery after Hemorrhagic Stroke

Hemorrhagic stroke is a life-threatening disease characterized by a sudden rupture of cerebral blood vessels, and it is widely believed that neural cell death occurs after exposure to blood metabolites or subsequently damaged cells. Neural stem cells (NSCs), which maintain neurogenesis and are found in subgranular zone and subventricular zone, are thought to be an endogenous neuroprotective mechanism for these brain injuries. However, due to the complexity of NSCs and their microenvironment, current strategies cannot satisfactorily enhance functional recovery after hemorrhagic stroke. It is well known that transcriptional and genomic pathways play important roles in ensuring the normal functions of NSCs, including proliferation, migration, differentiation, and neural reconnection. Recently, emerging evidence from the use of new technologies such as next-generation sequencing and transcriptome profiling has provided insight into our understanding of genomic function and regulation of NSCs. In the present article, we summarize and present the current data on the control of NSCs at both the transcriptional and genomic levels. Using bioinformatics methods, we sought to predict novel therapeutic targets of endogenous neurogenesis and exogenous NSC transplantation for functional recovery after hemorrhagic stroke, which could also advance our understanding of its pathophysiology.