Biology subjects

Alon, U.

Publications and source records attributed to Alon, U..

At least 19 recordsLinked to original sources

bioRxiv2026

A unified framework links infant vulnerability with aging-related mortality dynamics

A central question in Geroscience is whether early-life mortality, which declines from birth to sexual maturity, and late-life mortality, which grows exponentially in time, can be understood within a shared conceptual framework. We show that stochastic threshold models can explain both phases by incorporating heterogeneity in neonatal vulnerability. Using U.S. National Center for Health Statistics data, we find that infant mortality risk is strongly associated with neonatal clinical markers such as Apgar scores, gestational age, and birth weight, suggesting that initial physiological differences persist across early life. We show that the [~]1/t mortality decline generically arises in stochastic threshold models via depletion of the most vulnerable, across a wide range of model specifications. Incorporating this mechanism into the Saturating-Removal model captures both the early decline and the later Gompertz acceleration, reproducing the full J-shaped mortality curve. Together, our findings link neonatal vulnerability to late-life mortality dynamics within a shared stochastic framework, supporting a life-course perspective on aging and longevity.

systems biology↗

bioRxiv2026

Dynamics of the end-of-life phase explained by the saturating removal model

End of life is characterized by a phase of rapid physiological decline and high morbidity, phenotypically observed as the "Smurf" phase in Drosophila, metabolic end-of-life dysregulation in mice, and end-stage frailty in humans. Existing two-phase aging models often conceptualize this end-of-life phase as a discrete biological state. Here, we demonstrate that a continuous stochastic model of damage accumulation, the saturating removal (SR) model, captures these multi-species morbidity dynamics. By defining the end-of-life phase as a stochastic crossing of a sub-lethal damage threshold, the SR model accurately reproduces empirical end-of-life dynamics across flies, mice, and humans. The model predicts a surprising temporary reduction in hazard shortly after entering the end-of-life phase consistent with empirical data in all three organisms. It also correctly predicts a shortening twilight phenomenon where the mean duration of the end-of-life phase decreases the later its onset. We conclude that end-of-life dynamics are consistent with universal features of a driver of aging crossing a threshold for end-of-life morbidity and then a threshold for death.

systems biology↗

bioRxiv2026

Cancer resistance to therapy by tissue-level homeostatic feedback

Cancer displays remarkable robustness, exemplified by its ability to develop resistance to virtually every therapy. Resistance has traditionally been explained by clonal selection of pre-existing mutations, but there is now abundant evidence for resistance by non-genetic pathways including signals from normal stromal and immune cells. It is largely unclear why normal cells help cancer cells overcome treatment. We propose that physiological circuits responsible for tissue homeostasis can explain why cells cooperate to produce pathological resistance to therapy. To show this, we construct mathematical models of physiological dynamics. We then simulate cancer treatments within the context of a functioning tissue. We find that classic examples of resistance to therapy can be explained by homeostatic feedback regulation - including BRAF inhibitors in melanoma and anti-angiogenic therapy. The homeostatic theory of resistance (HTOR) reframes resistance as a byproduct of tissue robustness, rather than solely tumor-specific adaptation. Finally, we analyze two large-scale single-cell RNAseq databases of normal and cancer samples: the Tabula Sapiens1 and the Curated Cancer Atlas2. We show that in multiple cancers (breast, colon, kidney, liver, lung, ovary, prostate, and skin), malignant cells preserve their tissue-specific homeostatic cell-signaling. We thus expect the robust feedback loops from healthy tissues to play a role in cancer.

cancer biology↗

bioRxiv2025

Maximal human lifespan in light of a mechanistic model of aging

Why has maximal human lifespan barely changed in the past two centuries? To understand this we make a mechanistic link between cellular damage, survival curves, and maximum lifespan using a validated stochastic model of damage accumulation and extensive human data. We show that maximal lifespan is set mainly by damage production and clearance rates, as in progeroid syndromes. In contrast, lifestyle factors such as exercise, nutrition, and sleep chiefly reduce stochastic noise and raise the damage level compatible with survival, shifting the median but not the maximum. Similar constraints arise in other mortality models. Our analysis predicts that lifestyle can extend maximal lifespan by at most [~]1 year; substantial gains will require directly perturbing damage production or removal, suggesting specific molecular targets.

systems biology↗

bioRxiv2025

Cell surface ATP6V1B2 marks a subset of persistent senescent cells with increased resistance to apoptosis

Accumulation of senescent cells promotes ageing and age-related diseases. While senescent cells are heterogenous and increasingly persistent in vivo with age, the mechanisms underlying their heterogeneity, resistance to apoptosis, and tissue accumulation remain insufficiently understood. Here we report that in response to DNA damage, a subset of senescent cells upregulates the v-type ATPase subunit, ATP6V1B2 (V1B2) on the cell surface. This upregulation is associated with altered lysosomal activity and changes in intracellular pH. Heterogeneity of senescent cells marked by cell surface V1B2 (csV1B2) is present in naturally occurring senescent cells within both ageing and fibrotic lungs. Senescent cells expressing csV1B2 show an age-independent transcriptional signature associated with DNA repair and resistance to apoptosis. Consistent with this, we show that csV1B2 expression correlates with senescent cell resistance to ABT-737-induced apoptosis in culture. Our study identifies a subset of senescent cells, marked by csV1B2, with a distinct signature of apoptosis resistance. Understanding the functional heterogeneity of senescent cells and the mechanisms accountable for persistence of specific subpopulations in tissues may facilitate the development of improved senotherapeutic strategies for age-related diseases.

cell biology↗

bioRxiv2025

Single-cell quantification of senescence burden reveals cell type-specific ageing dynamics across organs

Cellular senescence, a hallmark of ageing, drives tissue dysfunction by promoting inflammation and fuelling disease. Yet, the dynamics of senescent cell accumulation across tissues and their cell type identity remain poorly understood. Here, we introduce the first, single-cell, protein-level approach, combining multiple senescence markers for the identification and quantification of senescent cells across multiple tissues in mice and in human PBMCs. Applying this method, we reveal widespread but heterogeneous changes in senescence marker expression across cell types and tissues. The cells we identify as senescent displayed transcriptomic senescence signatures, providing a direct molecular link between protein- and mRNA-level detection of senescence. Importantly, senescence accumulation was strongly coordinated within organs but showed little correlation across them, supporting the idea of a tissue specific progression of ageing. These findings refine our understanding of the tissue-specific dynamics of senescence accumulation with age, and provide a framework for evaluating diverse therapeutic interventions.

cell biology↗

bioRxiv2025

A Mechanistic Model for the HPA Axis Cortisol Paradox in PTSD

Post-traumatic Stress Disorder (PTSD) is a debilitating psychiatric condition characterized by intrusive memories, hyperarousal, avoidance, and cognitive and mood disturbances. A longstanding biological paradox in PTSD is the observation of low basal cortisol levels, despite the expectation of elevated cortisol under chronic stress. This "low cortisol paradox" challenges traditional hypothalamic-pituitary-adrenal (HPA) axis regulation models. Individuals with PTSD also exhibit normal or near-normal adrenocorticotropic hormone (ACTH) levels despite reduced cortisol and blunted hormonal responses to acute stress. In this paper, we provide further evidence of reduced cortisol levels using a large medical database from thousands of individuals diagnosed with PTSD. To understand this dysregulation, we apply a systems-level mathematical model of HPA axis regulation that incorporates dynamic changes in gland functional mass, the pituitary corticotrophs and adrenal cortex, over weeks to months. Using this model, we demonstrate that enhanced glucocorticoid receptor (GR) sensitivity, a known risk factor for PTSD, can lead to a steady state with low cortisol and normal ACTH, reconciling key clinical observations. The model also recapitulates the blunted cortisol and ACTH responses to acute stress and the dexamethasone/ corticotropin-releasing hormone (DEX/CRH) test, reported in PTSD cohorts. Importantly, despite low cortisol levels, basal glucocorticoid receptor (GR) activity is higher than normal due to the reduced receptor affinity. Thus, individuals with PTSD effectively experience elevated cortisol signaling relative to their GR sensitivity. These findings provide a unified mechanistic explanation for HPA axis dysregulation in PTSD, grounded in the slow time scale of changes in gland functional mass and supported by literature and clinical data.

systems biology↗

bioRxiv2025

OSDR2.0 infers microenvironment-driven cell-state transitions and population dynamics from a single spatial biopsy

Cell populations in human tissues change over time by cell division, death and transitions between functional states. In the tumor microenvironment (TME), such dynamics are central to immune evasion, stromal remodeling and therapeutic response. However, it is difficult to measure such dynamics in vivo because usually only a single biopsy is available providing a static snapshot. To obtain cell population dynamics from a snapshot we previously developed One Shot Dynamic Reconstruction (OSDR1.0), an algorithm that reconstructs cell population dynamics over weeks to months from a spatial biopsy using cell-type information and a proliferation marker. OSDR1.0 however does not include transitions between cell states. Here we present OSDR2.0, an extension that incorporates transitions between cell states inferred from the local cellular neighborhood. The algorithm OSDR2.0 models the probability of a cell being in a given state (e.g., PD1+ vs. PD1- T cell, or cancer-associated vs. resting fibroblast) as a function of its surrounding cell types. These state-transition rules are then integrated into simulations of population dynamics. After the cell population is advanced by a timestep, the cell states are adjusted according to the new neighborhoods, using the fact that cell state transitions, which take hours, are typically much more rapid than cell population changes on the scale of weeks, and can thus be considered at quasi-steady-state. Applying OSDR2.0 to spatial proteomics data from triple-negative breast cancer (TNBC), we find that cell state is strongly associated with local microenvironment composition. Incorporating state transitions significantly improves the ability to predict treatment response - OSDR2.0 accurately separates responders from nonresponders to chemotherapy and immunotherapy based on early post-treatment biopsies, outperforming OSDR1.0 that does not include cell state transitions. This work highlights the importance of cell state plasticity in shaping tumor response and opens a way to infer both population dynamics and cell state transitions from static clinical samples.

cancer biology↗

bioRxiv2025

Excitability as a Design Principle in the Immune System

Growing datasets and mechanistic detail in immunology have outpaced the development of unifying concepts. Such concepts are required to explain the primary goals of immune circuits - strong response to pathogens, tolerance to self, and prevention of collateral damage. A principle that achieves these goals across diverse immune circuits could unify our understanding of the immune system. Here, we propose that excitability, a concept from dynamical systems, serves this role. We screen thousands of circuits to identify those that generate excitable dynamics, and find a single robust design. We scan the human immune network to find this circuit architecture in dozens of innate and adaptive subsystems. We provide evidence for excitability in data on longitudinal responses to SARS-CoV-2. Similar motifs underlie T cell activation, autoimmune flares, and tumor immune responses. This conserved motif provides therapeutic targets and suggests that excitability is a core design principle of immunity, bridging molecular and cellular levels.

immunology↗

bioRxiv2025

A fibroblast-centric network drives cold fibrosis in the tumor microenvironment of lung squamous cell carcinoma

The tumor microenvironment (TME) of chronic inflammation-associated cancers (CIACs) is shaped by cycles of injury and maladaptive repair, yet the principles organizing fibrotic stroma in these tumors remain unclear. Here, we applied the concept of hot versus cold fibrosis, originally credentialed in non-cancerous fibrosis of heart and kidney, to lung squamous cell carcinoma (LUSC), a prototypical CIAC. Single-cell transcriptomics of matched tumor and adjacent-normal tissue from 16 treatment-naive LUSC patients identified a cold fibrotic architecture in the LUSC TME: cancer-associated fibroblasts (CAFs) expanded and adopted myofibroblast and stress-response states, while macrophages were depleted. This macrophage-poor, CAF-rich stroma was maintained by CAF autocrine growth factor loops, including TIMP1, INHBA, TGFB1, and GMFB. In parallel, the immune compartment exhibited a hot tumor phenotype with abundant T and B cells, forming spatially distinct but molecularly engaged networks with CAFs. CAF gene programs typifying cold fibrosis in LUSC were conserved in other CIACs, including esophageal and gastric adenocarcinomas. These results redefine desmoplastic regions of tumors through the lens of a non-cancer fibrosis model, demonstrating that conserved stromal circuits constitute therapeutic vulnerabilities in CIACs.

cancer biology↗

bioRxiv2025

A Fibroblast State Choreographs an Epithelial YAP-dependent Regenerative Program Essential to (Pre)malignancy via ECM-mediated Mechanotransduction

Chronic lung injury generates metaplasia which occasionally, but ominously, progresses to squamous dysplasia and squamous lung cancer. To identify mechanisms through which disrupted tissue homeostasis contributes to malignant initiation and progression, we used in vivo and in vitro heterotypic recombinant models of human bronchial epithelial cells (hBECs) and fibroblasts. We demonstrate that injury-associated TGF-{beta} signaling creates a fibroblast state dependent upon HSP47 upregulation. These fibroblasts accumulated collagen, thus elevating tissue stiffness and activating mechanosignaling that sustained YAP-dependent embryonic-like, pro-malignant activities in adjacent hBECs. This Stress/Tension-Instructive Fibroblast (STIF) state, exhibited by stressed fibroblasts in premalignant and malignant lesions across multiple cancer types, was sufficient to reprogram disease-free hBECs to metaplasia and to drive hBECs with compromised tumor suppressor function to dysplasia, yet could be inhibited and reversed. STIFs suffice to activate epithelial phenotypes reminiscent of oncogene-mediated cell transformation and induce (pre)malignancy via increased force transmission, providing novel targets for prevention. Statement of significanceTissue injury creates a regenerative pro-tumorigenic Stress/Tension-Instructive Fibroblast (STIF) state which is sufficient to activate a YAP-dependent, pre-malignant program to induce or unmask pre-cancerous phenotypes in epithelial cells through mechanotransduction. Inhibition of STIF activity or mechanosignaling prevents metaplasia and progression to dysplasia. HighlightsO_LITissue injury creates a pro-tumorigenic Stress/Tension-Instructive Fibroblast (STIF) state in multiple organs that precedes and persists through cancer C_LIO_LISTIF signaling alone, working through fibroblasts and not epithelial cells, is sufficient to activate embryonic-like plasticity and induce epithelial pre-cancerous metaplastic lesions C_LIO_LISTIFs program (pre)malignant phenotypes in adjacent epithelial cells through mechanosignaling by activating YAP prior to tumor formation C_LIO_LIInhibiting STIFs or mechanosignaling prevents/reverts metaplasia and prevents progression to dysplasia C_LI

cancer biology↗

bioRxiv2025

Dynamic growth trajectories distinguish bacteriostatic and bactericidal antibiotics at subinhibitory concentrations

Subinhibitory antibiotic exposures are common in clinical and environmental contexts, yet their effects on bacterial growth dynamics remain incompletely understood. We studied the temporal response of Escherichia coli to a panel of bactericidal ("cidal") and bacteriostatic ("static") antibiotics at sub-minimum inhibitory concentrations (sub-MIC). We uncover a sharp dynamical distinction between the two classes. Bacteriostatic antibiotics reduce the initial growth rate in a dose-dependent manner, similar to a nutrient starvation response. In contrast, bactericidal antibiotics do not alter initial growth rates - cells continue to grow as fast as untreated cells - until an abrupt slowdown in growth rate. The onset of slowdown occurs earlier with increasing dose, suggesting a damage accumulation mechanism leading to a lethal threshold. Cidals also show a steeper dose-response curve. We propose that bacteria respond to cidal antibiotics with a "grow-fast-then-crash" strategy that is adaptive for transient lethal threats, whereas static antibiotics trigger stress adaptation and slower growth. While clinical outcomes of statics and cidals may be similar at full inhibitory doses, these sub-MIC dynamical signatures could influence resistance evolution and treatment outcomes in biofilms or partially resistant strains. Our findings offer a dynamic framework for antibiotic classification and raise new questions about how bacteria respond to sublethal antibiotic stress.

systems biology↗

bioRxiv2025

Heritability of human lifespan is about 50% when confounding factors are addressed

The heritability of human lifespan is a fundamental question in biology. Current estimates of heritability are low - twin studies show that about 20-25% of the variation in lifespan is explained by genetics, and some large family pedigree studies suggest it is as low as 7%. However, these studies do not distinguish between deaths driven by intrinsic biological processes and deaths caused by extrinsic factors such as accidents or infections. Here we use mathematical modeling and analyses of twin cohorts raised together and apart to show that extrinsic mortality skews heritability estimates by driving down measured lifespan correlations among twin pairs. We also identify a nonlinear effect of the cutoff age--the minimum age of death included in each study -- on estimates of heritability. Correcting for these factors more than doubles previous estimates, revealing that intrinsic heritability of human lifespan is above 50%. Such high heritability is similar to most other complex human traits. We thus challenge the consensus that genetics has only a minor effect on lifespan and show that genes explain the majority of lifespan variation. Since genes are important, understanding the genetics of longevity can reveal aging mechanisms and inform medicine and public health.

genetics↗

bioRxiv2025

An immunologist, ecologist and clinician walk into Plato's cave to discuss infections

If the immune system is an interconnected network, then the evolution of an archetype that is ideal for fighting one pathogen should result in tradeoffs decreasing its ability to fight others. How many archetypes are there in an immune system? We infected diverse mice with Plasmodium chabaudi, and identified five distinct archetypes of responses based on the hosts position in microbial load, immune activity, and host damage space. To better understand the nature of these archetypes, we developed a mathematical model of a generalized host-pathogen system. This model explains the number, and distribution of archetypes across a population of diverse hosts. Mice resilient to P. chabaudi exhibited poor outcomes when challenged with influenza, SARS-CoV-1, or Mycobacterium tuberculosis, and vice versa, supporting our tradeoff hypothesis.

systems biology↗

bioRxiv2025

Pregnancy laboratory test dynamics resemble rejuvenation of some organs and aging of others

Aging and pregnancy both involve changes in many physiological systems. Some of these changes are similar, leading to suggestions that pregnancy may be a model for aging. Recent studies using DNA methylation clocks showed apparent aging during gestation which resolves postpartum. Since aging and pregnancy are complex, it is important to compare them in terms of many physiological parameters and at many time points. Here, we analyzed cross-sectional data on 62 lab tests at weekly resolution in 300,000 pregnancies and 1.4 million nonpregnant females aged 20-80. We trained a regression model to predict age from lab tests. Apparent age dropped by 8 years in early pregnancy, rose by 30 years towards delivery, and recovered postpartum. Certain systems exhibited rejuvenation, with opposite trends in pregnancy and aging, including renal, iron, and most liver tests. Others, such as coagulation, thyroid, muscle, and metabolic systems, showed apparent aging. Some systems displayed mixed trends. Notably, in the systems that showed apparent aging, the physiological mechanisms for the changes differed between pregnancy and aging. Pregnancy complications led to an additional apparent aging of 4-8 years. We conclude that pregnancy has features of rejuvenation for some systems, but its aging-like features involve different mechanisms than true aging. Gestational rejuvenation-like mechanisms may offer clues for slowing aspects of biological aging.

bioinformatics↗

bioRxiv2024

An Atlas of Cellular Archetypes

We sought to discover universal organizing principles behind phenotypic variation within cell types. Pareto optimality describes how trade-offs between optimal solutions account for variation, predicting that the boundary points of a data distribution reflect specialized functions. We hypothesized that Pareto optimality dominates transcriptomic variation across all cell types. We used the Tabula Sapiens atlas of single-cell RNA sequencing across cell types and tissues in the human body to test this hypothesis and discovered that most cell types adhere to this theory. This enabled us to use this principled method to characterize the functions performed by each cell type. These phenotypes are derived from an unbiased approach and do not incorporate ideas from existing biological models or theories, and yet in many cases they recapitulate our understanding of the functions of major cell types. Ultimately, we conclude that multi-objective optimization broadly shapes the observed phenotypic variation within cell types. This finding enables us to write explicit representations of the low-dimensional manifolds on which transcriptomes of single cells reside. This can inform the design of the next generation of virtual cell language models, which aim to statistically learn low-dimensional transcriptomic manifolds.

molecular biology↗

bioRxiv2024

Design of bacterial DNT sensors based on computational models

Detecting explosive compounds such as 2,4,6-trinitrotoluene (TNT) and its volatile byproduct 2,4-dinitrotoluene (DNT) is paramount for public health and environmental safety. In this study, we present the successful application of diverse computational and data analysis models toward developing a bacterial biosensor engineered to detect DNT with high sensitivity and specificity. The Escherichia coli-based biosensor harbors a plasmid-based fusion of a gene promoter acting as the sensing element to a microbial bioluminescence gene cassette as the reporter. By analyzing endogenous and heterologous promoter data under conditions of DNT exposure, a total of 367 novel variants were generated. The biosensors engineered with these modifications demonstrated a remarkable amplification of up to 4-fold change in signal intensity upon exposure to 2,4-dinitrotoluene compared to non-modified biosensors, accompanied by a decrease in the detection threshold. Our analysis suggests that the sequence features with the highest contribution to biosensor performances are DNA folding patterns and nucleotide motifs associated with DNT sensing. These computational insights guided the rational design of the biosensor, leading to significantly improved DNT detection capabilities compared to the previous biosensor strain. Our results demonstrate the effectiveness of integrating computational modeling with synthetic biology techniques to develop advanced biosensors tailored for environmental monitoring applications. A similar approach may be applied to a wide array of ecological, industrial, and medical sensing endeavors.

synthetic biology↗

bioRxiv2024

Unifying regulatory motifs in endocrine circuits

In this study we identify unifying design principles in human endocrine systems. We find that 43 hormone systems, controlling diverse physiological functions, fall into 5 classes of circuits with shared structure - thus only a small number of the possible circuits actually occur. Each class uses a different regulatory logic to perform specific dynamical functions, such as homeostasis, acute input-output response or adjustable set points. The circuits employ interactions on two timescales: hormone secretion on the scale of minutes-hours and growth and shrinkage of endocrine gland mass on the scale of months, which impacts the amount of hormone the glands secrete. This two-timescale principle recurs in several classes of circuits, including the most complex class, which has an intermediate gland, the pituitary. We analyze the pituitary circuit in detail and find tradeoffs between endocrine amplification, buffering of hypersecreting tumors, and rapid response times. These unifying principles of regulation build a foundation for systems endocrinology.

systems biology↗