Robust and Scalable Learning of Complex Intrinsic Dataset Geometry via ElPiGraph

A time-resolved single-cell roadmap of the logic driving anterior neural crest diversification from neural border to migration stages

Neural crest cells exemplify cellular diversification from a multipotent progenitor population. However, the full sequence of early molecular choices orchestrating the emergence of neural crest heterogeneity from the embryonic ectoderm remains elusive. Gene-regulatory-networks (GRN) govern early development and cell specification toward definitive neural crest. Here, we combine ultradense single-cell transcriptomes with machine-learning and large-scale transcriptomic and epigenomic experimental validation of selected trajectories, to provide the general principles and highlight specific features of the GRN underlying neural crest fate diversification from induction to early migration stages using Xenopus frog embryos as a model. During gastrulation, a transient neural border zone state precedes the choice between neural crest and placodes which includes multiple converging gene programs. During neurulation, transcription factor connectome, and bifurcation analyses demonstrate the early emergence of neural crest fates at the neural plate stage, alongside an unbiased multipotent-like lineage persisting until epithelial-mesenchymal transition stage. We also decipher circuits driving cranial and vagal neural crest formation and provide a broadly applicable high-throughput validation strategy for investigating single-cell transcriptomes in vertebrate GRNs in development, evolution, and disease.

Reconstruction of macroglia and adult neurogenesis evolution through cross-species single-cell transcriptomic analyses

Macroglia fulfill essential functions in the adult vertebrate brain, producing and maintaining neurons and regulating neuronal communication. However, we still know little about their emergence and diversification. We used the zebrafish D. rerio as a distant vertebrate model with moderate glial diversity as anchor to reanalyze datasets covering over 600 million years of evolution. We identify core features of adult neurogenesis and innovations in the mammalian lineage with a potential link to the rarity of radial glia-like cells in adult humans. Our results also suggest that functions associated with astrocytes originated in a multifunctional cell type fulfilling both neural stem cell and astrocytic functions before these diverged. Finally, we identify conserved elements of macroglial cell identity and function and their time of emergence during evolution.

Deciphering the spatial landscape and plasticity of immunosuppressive fibroblasts in breast cancer

Although heterogeneity of FAP+ Cancer-Associated Fibroblasts (CAF) has been described in breast cancer, their plasticity and spatial distribution remain poorly understood. Here, we analyze trajectory inference, deconvolute spatial transcriptomics at single-cell level and perform functional assays to generate a high-resolution integrated map of breast cancer (BC), with a focus on inflammatory and myofibroblastic (iCAF/myCAF) FAP+ CAF clusters. We identify 10 spatially-organized FAP+ CAF-related cellular niches, called EcoCellTypes, which are differentially localized within tumors. Consistent with their spatial organization, cancer cells drive the transition of detoxification-associated iCAF (Detox-iCAF) towards immunosuppressive extracellular matrix (ECM)-producing myCAF (ECM-myCAF) via a DPP4- and YAP-dependent mechanism. In turn, ECM-myCAF polarize TREM2+ macrophages, regulatory NK and T cells to induce immunosuppressive EcoCellTypes, while Detox-iCAF are associated with FOLR2+ macrophages in an immuno-protective EcoCellType. FAP+ CAF subpopulations accumulate differently according to the invasive BC status and predict invasive recurrence of ductal carcinoma in situ (DCIS), which could help in identifying low-risk DCIS patients eligible for therapeutic de-escalation.

A nasal cell atlas reveals heterogeneity of tuft cells and their role in directing olfactory stem cell proliferation

The olfactory neuroepithelium serves as a sensory organ for odors and forms part of the nasal mucosal barrier. Olfactory sensory neurons are surrounded and supported by epithelial cells. Among them, microvillous cells (MVCs) are strategically positioned at the apical surface, but their specific functions are enigmatic, and their relationship to the other specialized epithelial cells is unclear. Here, we establish that the family of MVCs comprises tuft cells and ionocytes in both mice and humans. Integrating analysis of the respiratory and olfactory epithelia, we define the distinct receptor expression of TRPM5+ tuft-MVCs compared with Gɑ-gustducinhigh respiratory tuft cells and characterize a previously undescribed population of glandular DCLK1+ tuft cells. To establish how allergen sensing by tuft-MVCs might direct olfactory mucosal responses, we used an integrated single-cell transcriptional and protein analysis. Inhalation of Alternaria induced mucosal epithelial effector molecules including Chil4 and a distinct pathway leading to proliferation of the quiescent olfactory horizontal basal stem cell (HBC) pool, both triggered in the absence of olfactory apoptosis. Alternaria- and ATP-elicited HBC proliferation was dependent on TRPM5+ tuft-MVCs, identifying these specialized epithelial cells as regulators of olfactory stem cell responses. Together, our data provide high-resolution characterization of nasal tuft cell heterogeneity and identify a function of TRPM5+ tuft-MVCs in directing the olfactory mucosal response to allergens.

Detection of PatIent-Level distances from single cell genomics and pathomics data with Optimal Transport (PILOT)

Although clinical applications represent the next challenge in single-cell genomics and digital pathology, we still lack computational methods to analyze single-cell or pathomics data to find sample-level trajectories or clusters associated with diseases. This remains challenging as single-cell/pathomics data are multi-scale, i.e., a sample is represented by clusters of cells/structures, and samples cannot be easily compared with each other. Here we propose PatIent Level analysis with Optimal Transport (PILOT). PILOT uses optimal transport to compute the Wasserstein distance between two individual single-cell samples. This allows us to perform unsupervised analysis at the sample level and uncover trajectories or cellular clusters associated with disease progression. We evaluate PILOT and competing approaches in single-cell genomics or pathomics studies involving various human diseases with up to 600 samples/patients and millions of cells or tissue structures. Our results demonstrate that PILOT detects disease-associated samples from large and complex single-cell or pathomics data. Moreover, PILOT provides a statistical approach to find changes in cell populations, gene expression, and tissue structures related to the trajectories or clusters supporting interpretation of predictions.

Imaging and multi-omics datasets converge to define different neural progenitor origins for ATRT-SHH subgroups

Atypical teratoid rhabdoid tumors (ATRT) are divided into MYC, TYR and SHH subgroups, suggesting diverse lineages of origin. Here, we investigate the imaging of human ATRT at diagnosis and the precise anatomic origin of brain tumors in the Rosa26-CreERT2::Smarcb1flox/flox model. This cross-species analysis points to an extra-cerebral origin for MYC tumors. Additionally, we clearly distinguish SHH ATRT emerging from the cerebellar anterior lobe (CAL) from those emerging from the basal ganglia (BG) and intra-ventricular (IV) regions. Molecular characteristics point to the midbrain-hindbrain boundary as the origin of CAL SHH ATRT, and to the ganglionic eminence as the origin of BG/IV SHH ATRT. Single-cell RNA sequencing on SHH ATRT supports these hypotheses. Trajectory analyses suggest that SMARCB1 loss induces a de-differentiation process mediated by repressors of the neuronal program such as REST, ID and the NOTCH pathway.

Detection of Amyotrophic Lateral Sclerosis (ALS) Comorbidity Trajectories Based on Principal Tree Model Analytics

The multifaceted nature and swift progression of Amyotrophic Lateral Sclerosis (ALS) pose considerable challenges to our understanding of its evolution and interplay with comorbid conditions. This study seeks to elucidate the temporal dynamics of ALS progression and its interaction with associated diseases. We employed a principal tree-based model to decipher patterns within clinical data derived from a population-based database in Taiwan. The disease progression was portrayed as branched trajectories, each path representing a series of distinct stages. Each stage embodied the cumulative occurrence of co-existing diseases, depicted as nodes on the tree, with edges symbolizing potential transitions between these linked nodes. Our model identified eight distinct ALS patient trajectories, unveiling unique patterns of disease associations at various stages of progression. These patterns may suggest underlying disease mechanisms or risk factors. This research re-conceptualizes ALS progression as a migration through diverse stages, instead of the perspective of a sequence of isolated events. This new approach illuminates patterns of disease association across different progression phases. The insights obtained from this study hold the potential to inform doctors regarding the development of personalized treatment strategies, ultimately enhancing patient prognosis and quality of life.

Cell-connectivity-guided trajectory inference from single-cell data

MOTIVATION

Single-cell RNA-sequencing enables cell-level investigation of cell differentiation, which can be modelled using trajectory inference methods. While tremendous effort has been put into designing these methods, inferring accurate trajectories automatically remains difficult. Therefore, the standard approach involves testing different trajectory inference methods and picking the trajectory giving the most biologically sensible model. As the default parameters are often suboptimal, their tuning requires methodological expertise.

RESULTS

We introduce Totem, an open-source, easy-to-use R package designed to facilitate inference of tree-shaped trajectories from single-cell data. Totem generates a large number of clustering results, estimates their topologies as minimum spanning trees, and uses them to measure the connectivity of the cells. Besides automatic selection of an appropriate trajectory, cell connectivity enables to visually pinpoint branching points and milestones relevant to the trajectory. Furthermore, testing different trajectories with Totem is fast, easy, and does not require in-depth methodological knowledge.

AVAILABILITY AND IMPLEMENTATION

Totem is available as an R package at https://github.com/elolab/Totem.

Transgenic ferret models define pulmonary ionocyte diversity and function

Speciation leads to adaptive changes in organ cellular physiology and creates challenges for studying rare cell-type functions that diverge between humans and mice. Rare cystic fibrosis transmembrane conductance regulator (CFTR)-rich pulmonary ionocytes exist throughout the cartilaginous airways of humans1,2, but limited presence and divergent biology in the proximal trachea of mice has prevented the use of traditional transgenic models to elucidate ionocyte functions in the airway. Here we describe the creation and use of conditional genetic ferret models to dissect pulmonary ionocyte biology and function by enabling ionocyte lineage tracing (FOXI1-CreERT2::ROSA-TG), ionocyte ablation (FOXI1-KO) and ionocyte-specific deletion of CFTR (FOXI1-CreERT2::CFTRL/L). By comparing these models with cystic fibrosis ferrets3,4, we demonstrate that ionocytes control airway surface liquid absorption, secretion, pH and mucus viscosity-leading to reduced airway surface liquid volume and impaired mucociliary clearance in cystic fibrosis, FOXI1-KO and FOXI1-CreERT2::CFTRL/L ferrets. These processes are regulated by CFTR-dependent ionocyte transport of Cl- and HCO3-. Single-cell transcriptomics and in vivo lineage tracing revealed three subtypes of pulmonary ionocytes and a FOXI1-lineage common rare cell progenitor for ionocytes, tuft cells and neuroendocrine cells during airway development. Thus, rare pulmonary ionocytes perform critical CFTR-dependent functions in the proximal airway that are hallmark features of cystic fibrosis airway disease. These studies provide a road map for using conditional genetics in the first non-rodent mammal to address gene function, cell biology and disease processes that have greater evolutionary conservation between humans and ferrets.

Ensuring Explainability and Dimensionality Reduction in a Multidimensional HSI World for Early XAI-Diagnostics of Plant Stress

This work is mostly devoted to the search for effective solutions to the problem of early diagnosis of plant stress (given an example of wheat and its drought stress), which would be based on explainable artificial intelligence (XAI). The main idea is to combine the benefits of two of the most popular agricultural data sources, hyperspectral images (HSI) and thermal infrared images (TIR), in a single XAI model. Our own dataset of a 25-day experiment was used, which was created via both (1) an HSI camera Specim IQ (400-1000 nm, 204, 512 × 512) and (2) a TIR camera Testo 885-2 (320 × 240, res. 0.1 °C). The HSI were a source of the k-dimensional high-level features of plants (k ≤ K, where K is the number of HSI channels) for the learning process. Such combination was implemented as a single-layer perceptron (SLP) regressor, which is the main feature of the XAI model and receives as input an HSI pixel-signature belonging to the plant mask, which then automatically through the mask receives a mark from the TIR. The correlation of HSI channels with the TIR image on the plant's mask on the days of the experiment was studied. It was established that HSI channel 143 (820 nm) was the most correlated with TIR. The problem of training the HSI signatures of plants with their corresponding temperature value via the XAI model was solved. The RMSE of plant temperature prediction is 0.2-0.3 °C, which is acceptable for early diagnostics. Each HSI pixel was represented in training by a number (k) of channels (k ≤ K = 204 in our case). The number of channels used for training was minimized by a factor of 25-30, from 204 to eight or seven, while maintaining the RMSE value. The model is computationally efficient in training; the average training time was much less than one minute (Intel Core i3-8130U, 2.2 GHz, 4 cores, 4 GB). This XAI model can be considered a research-aimed model (R-XAI), which allows the transfer of knowledge about plants from the TIR domain to the HSI domain, with their contrasting onto only a few from hundreds of HSI channels.

A Clustering Method Unifying Cell-Type Recognition and Subtype Identification for Tumor Heterogeneity Analysis

The rapid development of single-cell technology has opened up a whole new perspective for identifying cell types in multicellular organisms and understanding the relationships between them. Distinguishing different cell types and subtypes can identify the components of different immune cells and different tumor clones in the tumor microenvironment, which is the basic work of tumor heterogeneity analysis and can help researchers understand the mechanism of tumor immune escape. Existing algorithms treat both cell types and subtypes as populations of cells with specific gene expression patterns, which is not conducive to accurate cell typing. For that, we proposed a cell similarity metric that unifies cell type recognition and subtype identification (UCRSI), with the assumption that selectively expressed genes represent differences in underlying cell type with on/off manner, while differences in expression level represent different cell subtype with more/less manner. Our method calculates these two kinds of differences separately, and then combines them using a consensus adjacency matrix, and finally cell typing is completed using spectral clustering algorithm. The results show that UCRSI can reconstruct expert annotation of single-cell RNA sequencing datasets more robustly than existing methods. And, UCRSI is useful for analyzing tumor heterogeneity and improving visualization of large-scale cell clustering.

scFates: a scalable python package for advanced pseudotime and bifurcation analysis from single-cell data

SUMMARY

scFates provides an extensive toolset for the analysis of dynamic trajectories comprising tree learning, feature association testing, branch differential expression and with a focus on cell biasing and fate splits at the level of bifurcations. It is meant to be fully integrated into the scanpy ecosystem for seamless analysis of trajectories from single-cell data of various modalities (e.g. RNA and ATAC).

AVAILABILITY AND IMPLEMENTATION

scFates is released as open-source software under the BSD 3-Clause 'New' License and is available from the Python Package Index at https://pypi.org/project/scFates/. The source code is available on GitHub at https://github.com/LouisFaure/scFates/. Code reproduction and tutorials on published datasets are available on GitHub at https://github.com/LouisFaure/scFates_notebooks.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Domain Adaptation Principal Component Analysis: Base Linear Method for Learning with Out-of-Distribution Data

Domain adaptation is a popular paradigm in modern machine learning which aims at tackling the problem of divergence (or shift) between the labeled training and validation datasets (source domain) and a potentially large unlabeled dataset (target domain). The task is to embed both datasets into a common space in which the source dataset is informative for training while the divergence between source and target is minimized. The most popular domain adaptation solutions are based on training neural networks that combine classification and adversarial learning modules, frequently making them both data-hungry and difficult to train. We present a method called Domain Adaptation Principal Component Analysis (DAPCA) that identifies a linear reduced data representation useful for solving the domain adaptation task. DAPCA algorithm introduces positive and negative weights between pairs of data points, and generalizes the supervised extension of principal component analysis. DAPCA is an iterative algorithm that solves a simple quadratic optimization problem at each iteration. The convergence of the algorithm is guaranteed, and the number of iterations is small in practice. We validate the suggested algorithm on previously proposed benchmarks for solving the domain adaptation task. We also show the benefit of using DAPCA in analyzing single-cell omics datasets in biomedical applications. Overall, DAPCA can serve as a practical preprocessing step in many machine learning applications leading to reduced dataset representations, taking into account possible divergence between source and target domains.

TATTOO-seq delineates spatial and cell type-specific regulatory programs in the developing limb

The coordinated differentiation of progenitor cells into specialized cell types and their spatial organization into distinct domains is central to embryogenesis. Here, we developed and applied an unbiased spatially resolved single-cell transcriptomics method to identify the genetic programs underlying the emergence of specialized cell types during mouse limb development and their spatial integration. We identify multiple transcription factors whose expression patterns are predominantly associated with cell type specification or spatial position, suggesting two parallel yet highly interconnected regulatory systems. We demonstrate that the embryonic limb undergoes a complex multiscale reorganization upon perturbation of one of its spatial organizing centers, including the loss of specific cell populations, alterations of preexisting cell states' molecular identities, and changes in their relative spatial distribution. Our study shows how multidimensional single-cell, spatially resolved molecular atlases can allow the deconvolution of spatial identity and cell fate and reveal the interconnected genetic networks that regulate organogenesis and its reorganization upon genetic alterations.

Regularization of Mixture Models for Robust Principal Graph Learning

A regularized version of Mixture Models is proposed to learn a principal graph from a distribution of D-dimensional datapoints. In the particular case of manifold learning for ridge detection, we assume that the underlying structure can be modeled as a graph acting like a topological prior for the Gaussian clusters turning the problem into a maximum a posteriori estimation. Parameters of the model are iteratively estimated through an Expectation-Maximization procedure making the learning of the structure computationally efficient with guaranteed convergence for any graph prior in a polynomial time. We also embed in the formalism a natural way to make the algorithm robust to outliers of the pattern and heteroscedasticity of the manifold sampling coherently with the graph structure. The method uses a graph prior given by the minimum spanning tree that we extend using random sub-samplings of the dataset to take into account cycles that can be observed in the spatial distribution.

Rosenblatt's First Theorem and Frugality of Deep Learning

The Rosenblatt's first theorem about the omnipotence of shallow networks states that elementary perceptrons can solve any classification problem if there are no discrepancies in the training set. Minsky and Papert considered elementary perceptrons with restrictions on the neural inputs: a bounded number of connections or a relatively small diameter of the receptive field for each neuron at the hidden layer. They proved that under these constraints, an elementary perceptron cannot solve some problems, such as the connectivity of input images or the parity of pixels in them. In this note, we demonstrated Rosenblatt's first theorem at work, showed how an elementary perceptron can solve a version of the travel maze problem, and analysed the complexity of that solution. We also constructed a deep network algorithm for the same problem. It is much more efficient. The shallow network uses an exponentially large number of neurons on the hidden layer (Rosenblatt's A-elements), whereas for the deep network, the second-order polynomial complexity is sufficient. We demonstrated that for the same complex problem, the deep network can be much smaller and reveal a heuristic behind this effect.

Entropy as a High-Level Feature for XAI-Based Early Plant Stress Detection

This article is devoted to searching for high-level explainable features that can remain explainable for a wide class of objects or phenomena and become an integral part of explainable AI (XAI). The present study involved a 25-day experiment on early diagnosis of wheat stress using drought stress as an example. The state of the plants was periodically monitored via thermal infrared (TIR) and hyperspectral image (HSI) cameras. A single-layer perceptron (SLP)-based classifier was used as the main instrument in the XAI study. To provide explainability of the SLP input, the direct HSI was replaced by images of six popular vegetation indices and three HSI channels (R₆₃₀, G₅₅₀, and B₄₈₀; referred to as indices), along with the TIR image. Furthermore, in the explainability analysis, each of the 10 images was replaced by its 6 statistical features: min, max, mean, std, max-min, and the entropy. For the SLP output explainability, seven output neurons corresponding to the key states of the plants were chosen. The inner layer of the SLP was constructed using 15 neurons, including 10 corresponding to the indices and 5 reserved neurons. The classification possibilities of all 60 features and 10 indices of the SLP classifier were studied. Study result: Entropy is the earliest high-level stress feature for all indices; entropy and an entropy-like feature (max-min) paired with one of the other statistical features can provide, for most indices, 100% accuracy (or near 100%), serving as an integral part of XAI.

Single-cell and single-nuclei RNA sequencing as powerful tools to decipher cellular heterogeneity and dysregulation in neurodegenerative diseases

Neurodegenerative diseases affect millions of people worldwide and there are currently no cures. Two types of common neurodegenerative diseases are Alzheimer's (AD) and Parkinson's disease (PD). Single-cell and single-nuclei RNA sequencing (scRNA-seq and snRNA-seq) have become powerful tools to elucidate the inherent complexity and dynamics of the central nervous system at cellular resolution. This technology has allowed the identification of cell types and states, providing new insights into cellular susceptibilities and molecular mechanisms underlying neurodegenerative conditions. Exciting research using high throughput scRNA-seq and snRNA-seq technologies to study AD and PD is emerging. Herein we review the recent progress in understanding these neurodegenerative diseases using these state-of-the-art technologies. We discuss the fundamental principles and implications of single-cell sequencing of the human brain. Moreover, we review some examples of the computational and analytical tools required to interpret the extensive amount of data generated from these assays. We conclude by highlighting challenges and limitations in the application of these technologies in the study of AD and PD.

Cutting-Edge AI Technologies Meet Precision Medicine to Improve Cancer Care

To provide precision medicine for better cancer care, researchers must work on clinical patient data, such as electronic medical records, physiological measurements, biochemistry, computerized tomography scans, digital pathology, and the genetic landscape of cancer tissue. To interpret big biodata in cancer genomics, an operational flow based on artificial intelligence (AI) models and medical management platforms with high-performance computing must be set up for precision cancer genomics in clinical practice. To work in the fast-evolving fields of patient care, clinical diagnostics, and therapeutic services, clinicians must understand the fundamentals of the AI tool approach. Therefore, the present article covers the following four themes: (i) computational prediction of pathogenic variants of cancer susceptibility genes; (ii) AI model for mutational analysis; (iii) single-cell genomics and computational biology; (iv) text mining for identifying gene targets in cancer; and (v) the NVIDIA graphics processing units, DRAGEN field programmable gate arrays systems and AI medical cloud platforms in clinical next-generation sequencing laboratories. Based on AI medical platforms and visualization, large amounts of clinical biodata can be rapidly copied and understood using an AI pipeline. The use of innovative AI technologies can deliver more accurate and rapid cancer therapy targets.

Single-cell RNA-sequencing analysis of the developing mouse inner ear identifies molecular logic of auditory neuron diversification

Different types of spiral ganglion neurons (SGNs) are essential for auditory perception by transmitting complex auditory information from hair cells (HCs) to the brain. Here, we use deep, single cell transcriptomics to study the molecular mechanisms that govern their identity and organization in mice. We identify a core set of temporally patterned genes and gene regulatory networks that may contribute to the diversification of SGNs through sequential binary decisions and demonstrate a role for NEUROD1 in driving specification of a Ic-SGN phenotype. We also find that each trajectory of the decision tree is defined by initial co-expression of alternative subtype molecular controls followed by gradual shifts toward cell fate resolution. Finally, analysis of both developing SGN and HC types reveals cell-cell signaling potentially playing a role in the differentiation of SGNs. Our results indicate that SGN identities are drafted prior to birth and reveal molecular principles that shape their differentiation and will facilitate studies of their development, physiology, and dysfunction.

Neural network learning defines glioblastoma features to be of neural crest perivascular or radial glia lineages

Glioblastoma is believed to originate from nervous system cells; however, a putative origin from vessel-associated progenitor cells has not been considered. We deeply single-cell RNA-sequenced glioblastoma progenitor cells of 18 patients and integrated 710 bulk tumors and 73,495 glioma single cells of 100 patients to determine the relation of glioblastoma cells to normal brain cell types. A novel neural network-based projection of the developmental trajectory of normal brain cells uncovered two principal cell-lineage features of glioblastoma, neural crest perivascular and radial glia, carrying defining methylation patterns and survival differences. Consistently, introducing tumorigenic alterations in naïve human brain perivascular cells resulted in brain tumors. Thus, our results suggest that glioblastoma can arise from the brains' vasculature, and patients with such glioblastoma have a significantly poorer outcome.

Modeling Progression of Single Cell Populations Through the Cell Cycle as a Sequence of Switches

Cell cycle is a biological process underlying the existence and propagation of life in time and space. It has been an object for mathematical modeling for long, with several alternative mechanistic modeling principles suggested, describing in more or less details the known molecular mechanisms. Recently, cell cycle has been investigated at single cell level in snapshots of unsynchronized cell populations, exploiting the new methods for transcriptomic and proteomic molecular profiling. This raises a need for simplified semi-phenomenological cell cycle models, in order to formalize the processes underlying the cell cycle, at a higher abstracted level. Here we suggest a modeling framework, recapitulating the most important properties of the cell cycle as a limit trajectory of a dynamical process characterized by several internal states with switches between them. In the simplest form, this leads to a limit cycle trajectory, composed by linear segments in logarithmic coordinates describing some extensive (depending on system size) cell properties. We prove a theorem connecting the effective embedding dimensionality of the cell cycle trajectory with the number of its linear segments. We also develop a simplified kinetic model with piecewise-constant kinetic rates describing the dynamics of lumps of genes involved in S-phase and G2/M phases. We show how the developed cell cycle models can be applied to analyze the available single cell datasets and simulate certain properties of the observed cell cycle trajectories. Based on our model, we can predict with good accuracy the cell line doubling time from the length of cell cycle trajectory.

scShaper: an ensemble method for fast and accurate linear trajectory inference from single-cell RNA-seq data

MOTIVATION

Computational models are needed to infer a representation of the cells, i.e. a trajectory, from single-cell RNA-sequencing data that model cell differentiation during a dynamic process. Although many trajectory inference methods exist, their performance varies greatly depending on the dataset and hence there is a need to establish more accurate, better generalizable methods.

RESULTS

We introduce scShaper, a new trajectory inference method that enables accurate linear trajectory inference. The ensemble approach of scShaper generates a continuous smooth pseudotime based on a set of discrete pseudotimes. We demonstrate that scShaper is able to infer accurate trajectories for a variety of trigonometric trajectories, including many for which the commonly used principal curves method fails. A comprehensive benchmarking with state-of-the-art methods revealed that scShaper achieved superior accuracy of the cell ordering and, in particular, the differentially expressed genes. Moreover, scShaper is a fast method with few hyperparameters, making it a promising alternative to the principal curves method for linear pseudotemporal ordering.

AVAILABILITY AND IMPLEMENTATION

scShaper is available as an R package at https://github.com/elolab/scshaper. The test data are available at https://doi.org/10.5281/zenodo.5734488.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

OCT4-mediated inflammation induces cell reprogramming at the origin of cardiac valve development and calcification

Cell plasticity plays a key role in embryos by maintaining the differentiation potential of progenitors. Whether postnatal somatic cells revert to an embryonic-like naïve state regaining plasticity and redifferentiate into a cell type leading to a disease remains intriguing. Using genetic lineage tracing and single-cell RNA sequencing, we reveal that Oct4 is induced by nuclear factor κB (NFκB) at embyronic day 9.5 in a subset of mouse endocardial cells originating from the anterior heart forming field at the onset of endocardial-to-mesenchymal transition. These cells acquired a chondro-osteogenic fate. OCT4 in adult valvular aortic cells leads to calcification of mouse and human valves. These calcifying cells originate from the Oct4 embryonic lineage. Genetic deletion of Pou5f1 (Pit-Oct-Unc, OCT4) in the endocardial cell lineage prevents aortic stenosis and calcification of ApoE^−/− mouse valve. We established previously unidentified self-cell reprogramming NFκB- and OCT4-mediated inflammatory pathway triggering a dose-dependent mechanism of valve calcification.

singlecellVR: Interactive Visualization of Single-Cell Data in Virtual Reality

Single-cell assays have transformed our ability to model heterogeneity within cell populations. As these assays have advanced in their ability to measure various aspects of molecular processes in cells, computational methods to analyze and meaningfully visualize such data have required matched innovation. Independently, Virtual Reality (VR) has recently emerged as a powerful technology to dynamically explore complex data and shows promise for adaptation to challenges in single-cell data visualization. However, adopting VR for single-cell data visualization has thus far been hindered by expensive prerequisite hardware or advanced data preprocessing skills. To address current shortcomings, we present singlecellVR, a user-friendly web application for visualizing single-cell data, designed for cheap and easily available virtual reality hardware (e.g., Google Cardboard, ∼$8). singlecellVR can visualize data from a variety of sequencing-based technologies including transcriptomic, epigenomic, and proteomic data as well as combinations thereof. Analysis modalities supported include approaches to clustering as well as trajectory inference and visualization of dynamical changes discovered through modelling RNA velocity. We provide a companion software package, scvr to streamline data conversion from the most widely-adopted single-cell analysis tools as well as a growing database of pre-analyzed datasets to which users can contribute.

Scikit-Dimension: A Python Package for Intrinsic Dimension Estimation

Dealing with uncertainty in applications of machine learning to real-life data critically depends on the knowledge of intrinsic dimensionality (ID). A number of methods have been suggested for the purpose of estimating ID, but no standard package to easily apply them one by one or all at once has been implemented in Python. This technical note introduces scikit-dimension, an open-source Python package for intrinsic dimension estimation. The scikit-dimension package provides a uniform implementation of most of the known ID estimators based on the scikit-learn application programming interface to evaluate the global and local intrinsic dimension, as well as generators of synthetic toy and benchmark datasets widespread in the literature. The package is developed with tools assessing the code quality, coverage, unit testing and continuous integration. We briefly describe the package and demonstrate its use in a large-scale (more than 500 datasets) benchmarking of methods for ID estimation for real-life and synthetic data.

High-Dimensional Separability for One- and Few-Shot Learning

This work is driven by a practical question: corrections of Artificial Intelligence (AI) errors. These corrections should be quick and non-iterative. To solve this problem without modification of a legacy AI system, we propose special 'external' devices, correctors. Elementary correctors consist of two parts, a classifier that separates the situations with high risk of error from the situations in which the legacy AI system works well and a new decision that should be recommended for situations with potential errors. Input signals for the correctors can be the inputs of the legacy AI system, its internal signals, and outputs. If the intrinsic dimensionality of data is high enough then the classifiers for correction of small number of errors can be very simple. According to the blessing of dimensionality effects, even simple and robust Fisher's discriminants can be used for one-shot learning of AI correctors. Stochastic separation theorems provide the mathematical basis for this one-short learning. However, as the number of correctors needed grows, the cluster structure of data becomes important and a new family of stochastic separation theorems is required. We refuse the classical hypothesis of the regularity of the data distribution and assume that the data can have a rich fine-grained structure with many clusters and corresponding peaks in the probability density. New stochastic separation theorems for data with fine-grained structure are formulated and proved. On the basis of these theorems, the multi-correctors for granular data are proposed. The advantages of the multi-corrector technology were demonstrated by examples of correcting errors and learning new classes of objects by a deep convolutional neural network on the CIFAR-10 dataset. The key problems of the non-classical high-dimensional data analysis are reviewed together with the basic preprocessing steps including the correlation transformation, supervised Principal Component Analysis (PCA), semi-supervised PCA, transfer component analysis, and new domain adaptation PCA.

Interpretable Feature Generation in ECG Using a Variational Autoencoder

We propose a method for generating an electrocardiogram (ECG) signal for one cardiac cycle using a variational autoencoder. Our goal was to encode the original ECG signal using as few features as possible. Using this method we extracted a vector of new 25 features, which in many cases can be interpreted. The generated ECG has quite natural appearance. The low value of the Maximum Mean Discrepancy metric, 3.83 × 10⁻³, indicates good quality of ECG generation too. The extracted new features will help to improve the quality of automatic diagnostics of cardiovascular diseases. Generating new synthetic ECGs will allow us to solve the issue of the lack of labeled ECG for using them in supervised learning.

Trajectories, bifurcations, and pseudo-time in large clinical datasets: applications to myocardial infarction and diabetes data

BACKGROUND

Large observational clinical datasets are becoming increasingly available for mining associations between various disease traits and administered therapy. These datasets can be considered as representations of the landscape of all possible disease conditions, in which a concrete disease state develops through stereotypical routes, characterized by "points of no return" and "final states" (such as lethal or recovery states). Extracting this information directly from the data remains challenging, especially in the case of synchronic (with a short-term follow-up) observations.

RESULTS

Here we suggest a semi-supervised methodology for the analysis of large clinical datasets, characterized by mixed data types and missing values, through modeling the geometrical data structure as a bouquet of bifurcating clinical trajectories. The methodology is based on application of elastic principal graphs, which can address simultaneously the tasks of dimensionality reduction, data visualization, clustering, feature selection, and quantifying the geodesic distances (pseudo-time) in partially ordered sequences of observations. The methodology allows a patient to be positioned on a particular clinical trajectory (pathological scenario) and the degree of progression along it to be characterized with a qualitative estimate of the uncertainty of the prognosis. We developed a tool ClinTrajan for clinical trajectory analysis implemented in the Python programming language. We test the methodology in 2 large publicly available datasets: myocardial infarction complications and readmission of diabetic patients data.

CONCLUSIONS

Our pseudo-time quantification-based approach makes it possible to apply the methods developed for dynamical disease phenotyping and illness trajectory analysis (diachronic data analysis) to synchronic observational data.

Minimum Spanning vs. Principal Trees for Structured Approximations of Multi-Dimensional Datasets

Construction of graph-based approximations for multi-dimensional data point clouds is widely used in a variety of areas. Notable examples of applications of such approximators are cellular trajectory inference in single-cell data analysis, analysis of clinical trajectories from synchronic datasets, and skeletonization of images. Several methods have been proposed to construct such approximating graphs, with some based on computation of minimum spanning trees and some based on principal graphs generalizing principal curves. In this article we propose a methodology to compare and benchmark these two graph-based data approximation approaches, as well as to define their hyperparameters. The main idea is to avoid comparing graphs directly, but at first to induce clustering of the data point cloud from the graph approximation and, secondly, to use well-established methods to compare and score the data cloud partitioning induced by the graphs. In particular, mutual information-based approaches prove to be useful in this context. The induced clustering is based on decomposing a graph into non-branching segments, and then clustering the data point cloud by the nearest segment. Such a method allows efficient comparison of graph-based data approximations of arbitrary topology and complexity. The method is implemented in Python using the standard scikit-learn library which provides high speed and efficiency. As a demonstration of the methodology we analyse and compare graph-based data approximation methods using synthetic as well as real-life single cell datasets.

Single cell RNA sequencing identifies early diversity of sensory neurons forming via bi-potential intermediates

Somatic sensation is defined by the existence of a diversity of primary sensory neurons with unique biological features and response profiles to external and internal stimuli. However, there is no coherent picture about how this diversity of cell states is transcriptionally generated. Here, we use deep single cell analysis to resolve fate splits and molecular biasing processes during sensory neurogenesis in mice. Our results identify a complex series of successive and specific transcriptional changes in post-mitotic neurons that delineate hierarchical regulatory states leading to the generation of the main sensory neuron classes. In addition, our analysis identifies previously undetected early gene modules expressed long before fate determination although being clearly associated with defined sensory subtypes. Overall, the early diversity of sensory neurons is generated through successive bi-potential intermediates in which synchronization of relevant gene modules and concurrent repression of competing fate programs precede cell fate stabilization and final commitment.

Random Parametric Perturbations of Gene Regulatory Circuit Uncover State Transitions in Cell Cycle

Many biological processes involve precise cellular state transitions controlled by complex gene regulation. Here, we use budding yeast cell cycle as a model system and explore how a gene regulatory circuit encodes essential information of state transitions. We present a generalized random circuit perturbation method for circuits containing heterogeneous regulation types and its usage to analyze both steady and oscillatory states from an ensemble of circuit models with random kinetic parameters. The stable steady states form robust clusters with a circular structure that are associated with cell cycle phases. This circular structure in the clusters is consistent with single-cell RNA sequencing data. The oscillatory states specify the irreversible state transitions along cell cycle progression. Furthermore, we identify possible mechanisms to understand the irreversible state transitions from the steady states. We expect this approach to be robust and generally applicable to unbiasedly predict dynamical transitions of a gene regulatory circuit.