Single-cell gene set scoring with nearest neighbor graph smoothed data (gssnng).

Summary: Gene set scoring (or enrichment) is a common dimension reduction task in bioinformatics that can be focused on the differences between groups or at the single sample level. Gene sets can represent biological functions, molecular pathways, cell identities, and more. Gene set scores are context dependent values that are useful for interpreting biological changes following experiments or perturbations. Single sample scoring produces a set of scores, one for each member of a group, which can be analyzed with statistical models that can include additional clinically important factors such as gender or age. However, the sparsity and technical noise of single-cell expression measures create difficulties for these methods, which were originally designed for bulk expression profiling (microarrays, RNAseq). This can be greatly remedied by first applying a smoothing transformation that shares gene measure information within transcriptomic neighborhoods. In this work, we use the nearest neighbor graph of cells for matrix smoothing to produce high quality gene set scores on a per-cell, per-group, level which is useful for visualization and statistical analysis. Availability and implementation: The gssnng software is available using the python package index (PyPI) and works with Scanpy AnnData objects. It can be installed using "pip install gssnng." More information and demo notebooks: see https://github.com/IlyaLab/gssnng.

Concerted epithelial and stromal changes during progression of Barrett's Esophagus to invasive adenocarcinoma exposed by multi-scale, multi-omics analysis.

Esophageal adenocarcinoma arises from Barrett's esophagus, a precancerous metaplastic replacement of squamous by columnar epithelium in response to chronic inflammation. Multi-omics profiling, integrating single-cell transcriptomics, extracellular matrix proteomics, tissue-mechanics and spatial proteomics of 64 samples from 12 patients' paths of progression from squamous epithelium through metaplasia, dysplasia to adenocarcinoma, revealed shared and patient-specific progression characteristics. The classic metaplastic replacement of epithelial cells was paralleled by metaplastic changes in stromal cells, ECM and tissue stiffness. Strikingly, this change in tissue state at metaplasia was already accompanied by appearance of fibroblasts with characteristics of carcinoma-associated fibroblasts and of an NK cell-associated immunosuppressive microenvironment. Thus, Barrett's esophagus progresses as a coordinated multi-component system, supporting treatment paradigms that go beyond targeting cancerous cells to incorporating stromal reprogramming.

Cancer cell population growth kinetics at low densities deviate from the exponential growth model and suggest an Allee effect.

Most models of cancer cell population expansion assume exponential growth kinetics at low cell densities, with deviations to account for observed slowing of growth rate only at higher densities due to limited resources such as space and nutrients. However, recent preclinical and clinical observations of tumor initiation or recurrence indicate the presence of tumor growth kinetics in which growth rates scale positively with cell numbers. These observations are analogous to the cooperative behavior of species in an ecosystem described by the ecological principle of the Allee effect. In preclinical and clinical models, however, tumor growth data are limited by the lower limit of detection (i.e., a measurable lesion) and confounding variables, such as tumor microenvironment, and immune responses may cause and mask deviations from exponential growth models. In this work, we present alternative growth models to investigate the presence of an Allee effect in cancer cells seeded at low cell densities in a controlled in vitro setting. We propose a stochastic modeling framework to disentangle expected deviations due to small population size stochastic effects from cooperative growth and use the moment approach for stochastic parameter estimation to calibrate the observed growth trajectories. We validate the framework on simulated data and apply this approach to longitudinal cell proliferation data of BT-474 luminal B breast cancer cells. We find that cell population growth kinetics are best described by a model structure that considers the Allee effect, in that the birth rate of tumor cells increases with cell number in the regime of small population size. This indicates a potentially critical role of cooperative behavior among tumor cells at low cell densities with relevance to early stage growth patterns of emerging and relapsed tumors.

Increasing Neural Stem Cell Division Asymmetry and Quiescence Are Predicted to Contribute to the Age-Related Decline in Neurogenesis.

Adult murine neural stem cells (NSCs) generate neurons in drastically declining numbers with age. How cellular dynamics sustain neurogenesis and how alterations with age may result in this decline are unresolved issues. We therefore clonally traced NSC lineages using confetti reporters in young and middle-aged adult mice. To understand the underlying mechanisms, we derived mathematical models that explain observed clonal cell type abundances. The best models consistently show self-renewal of transit-amplifying progenitors and rapid neuroblast cell cycle exit. In middle-aged mice, we identified an increased probability of asymmetric stem cell divisions at the expense of symmetric differentiation, accompanied by an extended persistence of quiescence between activation phases. Our model explains existing longitudinal population data and identifies particular cellular properties underlying adult NSC homeostasis and the aging of this stem cell compartment.

Lineage marker synchrony in hematopoietic genealogies refutes the PU.1/GATA1 toggle switch paradigm.

Molecular regulation of cell fate decisions underlies health and disease. To identify molecules that are active or regulated during a decision, and not before or after, the decision time point is crucial. However, cell fate markers are usually delayed and the time of decision therefore unknown. Fortunately, dividing cells induce temporal correlations in their progeny, which allow for retrospective inference of the decision time point. We present a computational method to infer decision time points from correlated marker signals in genealogies and apply it to differentiating hematopoietic stem cells. We find that myeloid lineage decisions happen generations before lineage marker onsets. Inferred decision time points are in agreement with data from colony assay experiments. The levels of the myeloid transcription factor PU.1 do not change during, but long after the predicted lineage decision event, indicating  that the PU.1/GATA1 toggle switch paradigm cannot explain the initiation of early myeloid lineage choice.

cgCorrect: a method to correct for confounding cell-cell variation due to cell growth in single-cell transcriptomics.

Accessing gene expression at a single-cell level has unraveled often large heterogeneity among seemingly homogeneous cells, which remains obscured when using traditional population-based approaches. The computational analysis of single-cell transcriptomics data, however, still imposes unresolved challenges with respect to normalization, visualization and modeling the data. One such issue is differences in cell size, which introduce additional variability into the data and for which appropriate normalization techniques are needed. Otherwise, these differences in cell size may obscure genuine heterogeneities among cell populations and lead to overdispersed steady-state distributions of mRNA transcript numbers. We present cgCorrect, a statistical framework to correct for differences in cell size that are due to cell growth in single-cell transcriptomics data. We derive the probability for the cell-growth-corrected mRNA transcript number given the measured, cell size-dependent mRNA transcript number, based on the assumption that the average number of transcripts in a cell increases proportionally to the cell's volume during the cell cycle. cgCorrect can be used for both data normalization and to analyze the steady-state distributions used to infer the gene expression mechanism. We demonstrate its applicability on both simulated data and single-cell quantitative real-time polymerase chain reaction (PCR) data from mouse blood stem and progenitor cells (and to quantitative single-cell RNA-sequencing data obtained from mouse embryonic stem cells). We show that correcting for differences in cell size affects the interpretation of the data obtained by typically performed computational analysis.

Early myeloid lineage choice is not initiated by random PU.1 to GATA1 protein ratios.

The mechanisms underlying haematopoietic lineage decisions remain disputed. Lineage-affiliated transcription factors with the capacity for lineage reprogramming, positive auto-regulation and mutual inhibition have been described as being expressed in uncommitted cell populations. This led to the assumption that lineage choice is cell-intrinsically initiated and determined by stochastic switches of randomly fluctuating cross-antagonistic transcription factors. However, this hypothesis was developed on the basis of RNA expression data from snapshot and/or population-averaged analyses. Alternative models of lineage choice therefore cannot be excluded. Here we use novel reporter mouse lines and live imaging for continuous single-cell long-term quantification of the transcription factors GATA1 and PU.1 (also known as SPI1). We analyse individual haematopoietic stem cells throughout differentiation into megakaryocytic-erythroid and granulocytic-monocytic lineages. The observed expression dynamics are incompatible with the assumption that stochastic switching between PU.1 and GATA1 precedes and initiates megakaryocytic-erythroid versus granulocytic-monocytic lineage decision-making. Rather, our findings suggest that these transcription factors are only executing and reinforcing lineage choice once made. These results challenge the current prevailing model of early myeloid lineage choice.

Inference of spatiotemporal effects on cellular state transitions from time-lapse microscopy.

BACKGROUND: Time-lapse microscopy allows to monitor cell state transitions in a spatiotemporal context. Combined with single cell tracking and appropriate cell state markers, transition events can be observed within the genealogical relationship of a proliferating population. However, to infer the correlations between the spatiotemporal context and cell state transitions, statistical analysis with an appropriately large number of samples is required. RESULTS: Here, we present a method to infer spatiotemporal features predictive of the state transition events observed in time-lapse microscopy data. We first formulate a generative model, simulate different scenarios, such as time-dependent or local cell density-dependent transitions, and illustrate how to estimate univariate transition rates. Second, we formulate the problem in a machine-learning language using regularized linear models. This allows for a multivariate analysis and to disentangle indirect dependencies via feature selection. We find that our method can accurately recover the relevant features and reconstruct the underlying interaction kernels if a critical number of samples is available. Finally, we explicitly use the tree structure of the data to validate if the estimated model is sufficient to explain correlated transition events of sister cells. CONCLUSIONS: Using synthetic cellular genealogies, we prove that our method is able to correctly identify features predictive of state transitions and we moreover validate the chosen model. Our approach allows to estimate the number of cellular genealogies required for the proposed spatiotemporal statistical analysis, and we thus provide an important tool for the experimental design of challenging single cell time-lapse microscopy assays.