Biology subjects

Dumitrascu, B.

Publications and source records attributed to Dumitrascu, B..

15 recordsLinked to original sources

bioRxiv2026

Differentiable Design for Morphogenesis I: Simulation and Simulacra

Cells build tissues through local exchanges of force and information, yet the rules governing these interactions are difficult to infer from sparse observations. Here, we introduce waxMorph, a differentiable cell-based framework for generating and reconstructing three-dimensional morphogenesis. In synthetic and biological data, waxMorph reproduced established mechanochemical shape programs, inferred continuous trajectories from static tissue volumes, and recovered spatially organized latent signals. In a developing mouse myocardium dataset, it reconstructed unobserved intermediate geometries more accurately than optimal-transport interpolation, while in forelimbs it distinguished related developmental trajectories. By varying the capacity and spatial organization of the latent cues available to cells, waxMorph also provides a model-based way to quantify the complexity of shape assembly. waxMorph is built within the spatial-computing ecosystem of NVIDIA Warp. It provides an open-source, Python-native, GPU-accelerated, hybrid physics-AI framework for learning how local cellular interactions give rise to biological form.

systems biology↗

bioRxiv2026

A quantitative coordinate system for developmental dynamics

Quantitative comparison of morphogenesis across individuals remains a fundamental challenge, as developing embryos vary in shape, orientation and developmental tempo. Moreover, real-time three-dimensional imaging generates large, heterogeneous four-dimensional datasets that are difficult to directly align. As a result, developmental variability is typically described qualitatively rather than measured. Here we introduce STERN, a quantitative framework that learns continuous spatiotemporal representations of morphogenesis directly from in vivo 4D imaging data. By embedding embryos into a shared spatiotemporal space, STERN defines a quantitative developmental coordinate system that enables direct comparison of developmental trajectories across individuals without requiring explicit registration or staging. Applied to mouse embryogenesis, STERN reveals that embryos follow conserved developmental trajectories while progressing at distinct temporal rates, providing a quantitative measure of developmental heterochrony. Extending this framework to zebrafish neural crest light-sheet timelapse imaging, we further show that developmental order is preserved across distinct imaging views even with altered anatomical coverage, supporting the generality of the learned representation across vertebrate imaging contexts. Finally, in developing mouse hearts, where morphogenesis proceeds through subtle and continuously evolving structural changes, STERN resolves fine-scale developmental dynamics at minute-scale temporal resolution that are difficult to localize reproducibly using human experts or general-purpose multimodal AI. Together, these results establish a shared quantitative coordinate system for morphogenesis, in which developmental trajectories become directly comparable across individuals and developmental variability becomes a measurable property.

bioinformatics↗

bioRxiv2026

Wayfarer: A multiscale framework for spatial analysis of tumor progression

Spatial biology spans multiple length scales, from intracellular organization to tissue-level architecture. Spatial transcriptomics captures this structure, yet most analyses operate at a single spatial resolution, implicitly assuming that biological organization is scale-consistent. In practice, spatial autocorrelation and co-localization are functions of scale, and conclusions can depend on arbitrary aggregation choices. Here we present Wayfarer, a multiscale framework for spatial -omics that tracks how spatial association metrics evolve across nested spatial aggregations, enabling statistical comparison of multiscale structure across biological conditions. Using Xenium data from lung adenocarcinoma (LUAD), we show that spatial patterns often co-exist at fine and coarse scales and that progression is accompanied by reproducible shifts in scale-response profiles. These include increased fine-scale coherence of ERBB2-high tumor regions and coarse-scale clustering of immune-associated markers that are not apparent at a single resolution. Wayfarer converts spatial aggregation from a confounder into a diagnostic signal and is implemented as an R package to be released through Bioconductor.

bioinformatics↗

bioRxiv2025

Single-cell morphodynamics predict cell fate decisions during mucociliary epithelial differentiation.

Cell state transitions underlie the emergence of diverse cell types and are traditionally defined by changes in gene expression. Yet these transitions also involve coordinated shifts in cell morphology and behaviour, which remain poorly characterized in densely packed epithelia. We developed a quantitative live-imaging and computational framework to track thousands of individual cells over time in the rapidly differentiating Xenopus mucociliary epithelium (MCE). From segmentations and trajectories, we extracted dynamic features-cell and nuclear shape, movement, position-to create a time-resolved morphodynamic dataset spanning the full course of differentiation. While single features showed high noise and low separability of ground-truth cell types, supervised machine learning revealed that integrating time-resolved features robustly predicts final cell fate. Gradient-boosted trees and multinomial logistic regression achieved moderate but consistent accuracy, especially for abundant epithelial lineages. Key discriminants included normalized Z-position, membrane-nucleus offset, and absolute experimental time, whereas movement contributed minimally to the results. Our results show that morphodynamic signatures encode predictive information about cell identity and provide a framework linking physical dynamics with molecular state.

developmental biology↗

bioRxiv2025

Predictable Engineering of Signal-DependentCis-Regulatory Elements

Cis-regulatory elements (CREs) control how genes respond to external signals, but the principles governing their structure and function remain poorly understood. While differential transcription factor binding is known to regulate gene expression, how CREs integrate the amount and combination of inputs to secure precise spatiotemporal profiles of gene expression remains unclear. Here, we developed a high-throughput combinatorial screening strategy, that we term NeMECiS, to investigate signal- dependent synthetic CREs (synCREs) in differentiating mammalian stem cells. By concatenating fragments of functional CREs from genes that respond to Sonic Hedgehog in the developing vertebrate neural tube, we found that CRE activity follows hierarchical design rules. While individual 200-base-pair fragments showed minimal activity, their combinations generated thousands of functional signal-responsive synCREs, many exceeding the activity of natural sequences. Statistical modelling revealed CRE function can be decomposed into specific quantitative contributions in which sequence fragments combine through a multiplicative rule, tuned by their relative positioning and spacing. These findings provide a predictive framework for CRE redesign, which we used to engineer synthetic CREs that alter the pattern of motor neuron differentiation in neural tissue. These findings establish quantitative principles for engineering synthetic regulatory elements with programmable signal responses to rewire genetic circuits and control stem cell differentiation, providing a basis for understanding developmental gene regulation and designing therapeutic gene expression systems.

developmental biology↗

bioRxiv2024

Patches: A Representation Learning framework for Decoding Shared and Condition-Specific Transcriptional Programs in Wound Healing

Single-cell genomics enables the study of cell states and cell state transitions across biological conditions like aging, drug treatment, or injury. However, existing computational methods often struggle to simultaneously disentangle shared and condition-specific transcriptional patterns, particularly in experimental designs with missing data, unmatched cell populations, or complex attribute combinations. To address these challenges, Patches identifies universal transcriptomic features alongside condition-dependent variations in scRNA-seq data. Using conditional subspace learning, Patches enables robust integration, cross-condition prediction, and biologically interpretable representations of gene expression. Unlike prior methods, Patches excels in experimental designs with multiple attributes, such as age, treatment, and temporal dynamics, distinguishing general cellular mechanisms from condition-dependent changes. We applied Patches to both simulated data and real transcriptomic datasets from skin injury models, focusing on the effects of aging and drug treatment. Patches revealed shared wound healing patterns and condition-specific changes in cell behavior and extracellular matrix remodeling. These insights deepen our understanding of tissue repair and can identify potential biomarkers for therapeutic interventions, particularly in contexts where the experimental design is complicated by missing or difficult-to-collect data.

genomics↗

bioRxiv2024

Dynamic cell fate plasticity and tissue integration drive functional synovial joint regeneration

Adult mammalian synovial joints have limited regenerative capacity, where injuries heal with mechanically inferior fibrotic tissues. Here we developed a unilateral whole-joint resection model in adult zebrafish to advance our understanding of how to stimulate regrowth of native synovial joint tissues. Using a combination of microCT, histological, live imaging, and single-cell RNA sequencing (scRNAseq) approaches after complete removal of all joint tissues, we find de novo regeneration of articular cartilage, ligament, and synovium into a functional joint. Clonal lineage tracing and scRNAseq implicate a multipotent, neural crest-derived population in the adult skeleton as a cell source for these regenerating tissues. Together, our findings reveal latent molecular and cellular programs within the adult skeleton that are deployed to regenerate a complex joint with lubricated articular cartilage.

developmental biology↗

bioRxiv2023

A computational pipeline for spatial mechano-transcriptomics

Advances in spatial profiling technologies are providing insights into how molecular programs are influenced by local signaling and environmental cues. However, cell fate specification and tissue patterning involve the interplay of biochemical and mechanical feedback. Here, we develop a computational framework that enables the joint statistical analysis of transcriptional and mechanical signals in the context of spatial transcriptomics. To illustrate the application and utility of the approach, we use spatial transcriptomics data from the developing mouse embryo to infer the forces acting on individual cells, and use these results to identify mechanical, morphometric, and gene expression signatures that are predictive of tissue compartment boundaries. In addition, we use geoadditive structural equation modeling to identify gene modules that predict the mechanical behavior of cells in an unbiased manner. This computational framework is easily generalized to other spatial profiling contexts, providing a generic scheme for exploring the interplay of biomolecular and mechanical cues in tissues.

genomics↗

bioRxiv2023

Gene-level alignment of single cell trajectories informs the progression of in vitro T cell differentiation

Single-cell data analysis can infer dynamic changes in cell populations, for example across time, space or in response to perturbation. To compare these dynamics between two conditions, trajectory alignment via dynamic programming (DP) optimization is frequently used, but is limited by assumptions such as a definite existence of a match. Here we describe Genes2Genes, a Bayesian information-theoretic DP framework for aligning single-cell trajectories. Genes2Genes overcomes current limitations and is able to capture sequential matches and mismatches between a reference and a query at single gene resolution, highlighting distinct clusters of genes with varying patterns of expression dynamics. Across both real world and simulated datasets, Genes2Genes accurately captured different alignment patterns, demonstrated its utility in disease cell state trajectory analysis, and revealed that T cells differentiated in vitro matched to an immature in vivo state while lacking expression of genes associated with TNF[a] signaling. This use case demonstrates that precise trajectory alignment can pinpoint divergence from the in vivo system, thus guiding the optimization of in vitro culture conditions.

bioinformatics↗

bioRxiv2022

Hypergraph factorisation for multi-tissue gene expression imputation

AO_SCPLOWBSTRACTC_SCPLOWIntegrating gene expression across scales and tissues is crucial for understanding the biological mechanisms that drive disease and characterise homeostasis. However, traditional multi-tissue integration methods cannot handle uncollected tissues or rely on genotype information, which is subject to privacy concerns and often unavailable. To address these challenges, we present HYFA (Hypergraph Factorisation), a novel method for joint imputation of multi-tissue and cell-type gene expression. HYFA imputes tissue-specific gene expression via a specialised graph neural network operating on a hypergraph of individuals, metagenes, and tissues. HYFA is genotype- agnostic, supports a variable number of collected tissues per individual, and imposes strong inductive biases to leverage the shared regulatory architecture of tissues. In performance comparison on data from the Genotype Tissue Expression project, HYFA achieves superior performance over existing transcriptome imputation methods, especially when multiple reference tissues are available. Through transfer learning on a paired single-nucleus RNA-seq (snRNA-seq) dataset, we further show that HYFA can accurately resolve cell-type signatures from bulk gene expression, highlighting the methods ability to leverage gene expression programs underlying cell-type identity, even in tissues that were never observed in the training set. Using Gene Set Enrichment Analysis, we find that the metagenes learned by HYFA capture information about known biological pathways. Notably, the HYFA-imputed dataset can be used to identify regulatory genetic variations (eQTLs), with substantial gains over the original incomplete dataset. Our framework can accelerate effective and scalable integration of tissue and cell-type gene expression biorepositories.

bioinformatics↗

bioRxiv2022

Power analysis for spatial omics

As spatially-resolved multiplex profiling of RNA and proteins becomes more prominent, it is increasingly important to understand the statistical power available to test specific hypotheses when designing and interpreting such experiments. Ideally, it would be possible to create an oracle that predicts sampling requirements for generalized spatial experiments. However, the unknown number of relevant spatial features and the complexity of spatial data analysis makes this challenging. Here, we enumerate multiple parameters of interest that should be considered in the design of a properly powered spatial omics. We introduce a method for tunable in silico tissue generation, and use it with spatial profiling datasets to construct an exploratory computational framework for single cell spatial power analysis. Finally, we demonstrate that our framework can be applied across diverse spatial data modalities and tissues of interest.

bioinformatics↗

bioRxiv2019

Optimal gene selection for cell type discrimination in single cell analyses

Single-cell technologies characterize complex cell populations across multiple data modalities at un-precedented scale and resolution. Multi-omic data for single cell gene expression, in situ hybridization, or single cell chromatin states are increasingly available across diverse tissue types. When isolating specific cell types from a sample of disassociated cells or performing in situ sequencing in collections of heterogeneous cells, one challenging task is to select a small set of informative markers to identify and differentiate specific cell types or cell states as precisely as possible. Given single cell RNA-seq data and a set of cellular labels to discriminate, scGene-Fit selects gene transcript markers that jointly optimize cell label recovery using label-aware compressive classification methods, resulting in a substantially more robust and less redundant set of markers than existing methods. When applied to a data set given a hierarchy of cell type labels, the markers found by our method enable the recovery of the label hierarchy through a computationally efficient and principled optimization.

genomics↗

bioRxiv2019

Causal Network Inference from Gene Transcriptional Time Series Response to Glucocorticoids

Gene regulatory network inference is essential to uncover complex relationships among gene pathways and inform downstream experiments, ultimately paving the way for regulatory network re-engineering. Network inference from transcriptional time series data requires accurate, interpretable, and efficient determination of causal relationships among thousands of genes. Here, we develop Bootstrap Elastic net regression from Time Series (BETS), a statistical framework based on Granger causality for the recovery of a directed gene network from transcriptional time series data. BETS uses elastic net regression and stability selection from bootstrapped samples to infer causal relationships among genes. BETS is highly parallelized, enabling efficient analysis of large transcriptional data sets. We show competitive accuracy on a community benchmark, the DREAM4 100-gene network inference challenge, where BETS is one of the fastest among methods of similar performance but additionally infers whether the causal effects are activating or inhibitory. We apply BETS to transcriptional time series data of 2, 768 differentially-expressed genes from A549 cells exposed to glucocorticoids over a period of 12 hours. We identify a network of 2, 768 genes and 31, 945 directed edges (FDR [≤] 0.2). We validate inferred causal network edges using two external data sources: overexpression experiments on the same glucocorticoid system, and genetic variants associated with inferred edges in primary lung tissue in the Genotype-Tissue Expression (GTEx) v6 project. BETS is freely available as an open source software package at https://github.com/lujonathanh/BETS.

bioinformatics↗

bioRxiv2019

netNMF-sc: Leveraging gene-gene interactions for imputation and dimensionality reduction in single-cell expression analysis

MotivationSingle-cell RNA-sequencing (scRNA-seq) enables high throughput measurement of RNA expression in individual cells. Due to technical limitations, scRNA-seq data often contain zero counts for many transcripts in individual cells. These zero counts, or dropout events, complicate the analysis of scRNA-seq data using standard analysis methods developed for bulk RNA-seq data. Current scRNA-seq analysis methods typically overcome dropout by combining information across cells, leveraging the observation that cells generally occupy a small number of RNA expression states. ResultsWe introduce netNMF-sc, an algorithm for scRNA-seq analysis that leverages information across both cells and genes. netNMF-sc combines network-regularized non-negative matrix factorization with a procedure for handling zero inflation in transcript count matrices. The matrix factorization results in a low-dimensional representation of the transcript count matrix, which imputes gene abundance for both zero and non-zero entries and can be used to cluster cells. The network regularization leverages prior knowledge of gene-gene interactions, encouraging pairs of genes with known interactions to be close in the low-dimensional representation. We show that netNMF-sc outperforms existing methods on simulated and real scRNA-seq data, with increasing advantage at higher dropout rates (e.g. above 60%). Furthermore, we show that the results from netNMF-sc - including estimation of gene-gene covariance - are robust to choice of network, with more representative networks leading to greater performance gains. AvailabilitynetNMF-sc is available at github.com/raphael-group/netNMF-sc. Contactbraphael@princeton.edu

genomics↗

bioRxiv2018

GT-TS: Experimental design for maximizing cell type discovery in single-cell data

We present the Good-Toulmin like estimator via Thompson sampling, a computational method for iterative experimental design in multi-tissue single-cell RNA-seq (scRNA-seq) data. Given a budget and modeling cell type information across tissues, GT-TS estimates how many cells are required for sampling from each tissue with the goal of maximizing cell type discovery across samples from multiple iterations. In both real and simulated data, we demonstrate the advantages of GT-TS in data collection planning when compared to a random strategy in the absence of experimental design.

genomics↗