Single-cell transcriptomics, epigenomics, and other 'omics applied at single-cell resolution can significantly advance hypotheses and understanding of glial biology. Omics technologies are revealing a large and growing number of new glial cell subtypes, defined by their gene expression profile. These subtypes have significant implications for understanding glial cell function, cell-cell communications, and glia-specific changes between homeostasis and conditions such as neurological disease. For many, the training in how to analyze, interpret, and understand these large datasets has been through reading and understanding literature from other fields like biostatistics. Here, we provide a primer for glial biologists on experimental design and analysis of single-cell RNA-seq datasets. Our goal is to further the understanding of why decisions might be made about datasets and to enhance biologists' ability to interpret and critique their work and the work of others. We review the steps involved in single-cell analysis with a focus on decision points and particular notes for glia. The goal of this primer is to ensure that single-cell 'omics experiments continue to advance glial biology
Can AI autonomously design mechanisms for computer systems on par with the creativity and reasoning of human experts? We present Glia, an AI architecture for networked systems design that uses large language models (LLMs) in a human-inspired multi-agent workflow. Each agent specializes in reasoning, experimentation, and analysis, collaborating through an evaluation framework that grounds abstract reasoning in empirical feedback. Unlike prior ML-for-systems methods that optimize black-box policies, Glia generates interpretable designs and exposes its reasoning. When applied to a distributed GPU cluster for LLM inference, it produces new algorithms for request routing, scheduling, and auto-scaling that perform at human-expert levels in significantly less time, while yielding novel insights into workload behavior. Our results suggest that combining reasoning LLMs with structured experimentation, an AI can produce creative and understandable designs for complex systems problems.
Traumatic Brain Injury (TBI) results from an impact or concussion to the head with the injury being specifically characterized through pathological degradation at various biological length scales. Following injury, various mechanical modeling techniques have been proposed in the literature that seek to quantify neuronal-scale to tissue-scale metrics of brain damage. Broadly, the two categories of degradation encompass physiological deterioration of neurons and upregulation of chemical entities such as neurotransmitters which causes initiation of downstream pathophysiological effects. Despite the many contributing pathways, in this work, we delineate and model a potential glia-initiated injury pathway that leads to secondary injury. The goal of this work is to demonstrate a continuum framework which models the multiphysics of mechano-chemical interactions underlying TBI. Using a coupled PDE (partial differential equation) formulation and FEM (finite element method) discretization, the framework highlights evolution of field variables which spatio-temporally resolve mechanical metrics and chemical species across neuronal clusters. The modeling domain encompasses microglia, neurons an
The accumulation of potassium in the extracellular space surrounding nerve cells is a fundamental aspect of biophysics that has garnered significant attention in recent research. This phenomenon holds implications for various neurological conditions, including spreading depression, migraine, certain types of epilepsy, and potentially, learning processes. A quantitative analysis is essential for understanding the dynamics of potassium clearance following a series of action potentials. This clearance process involves multiple structures along the nerve, including glia, the extracellular space, axons, and the perivascular space, necessitating a spatially distributed systems approach akin to the cable equations of nerve physiology. In this study, we propose a multi-domain model for the optic nerve to investigate potassium accumulation and clearance dynamics. The model accounts for the convection, diffusion, and electrical migration of fluid and ions, revealing the significant roles of glia and the perivascular space in potassium buffering. Specifically, our findings suggest that potassium clearance primarily occurs through convective flow within the syncytia of glia, driven by osmotic
Automatic 3D reconstruction of glia morphology is a powerful tool necessary for investigating the role of microglia in neurological disorders in the central nervous system. Current glia skeleton reconstruction techniques fail to capture an accurate tracing of the processes over time, useful for the study of the microglia motility and morphology in the brain during healthy and diseased states. We propose Hieroglyph, a fully automatic temporal 3D skeleton reconstruction algorithm for glia imaged via 3D multiphoton microscopy. Hieroglyph yielded a 21% performance increase compared to state of the art automatic skeleton reconstruction methods and outperforms the state of the art in different measures of consistency on datasets of 3D images of microglia. The results from this method provide a 3D graph and digital reconstruction of glia useful for a myriad of morphological analyses that could impact studies in brain immunology and disease.
Olfactory ensheathing glia (OEG) cells are known to foster axonal regeneration of central nervous system (CNS) neurons. Several lines of reversibly immortalized human OEG (ihOEG) have been previously established that enabled to develop models for their validation in vitro and in vivo. In this work, a constitutively GFP-expressing ihOEG cell line was obtained, and named Ts14-GFP. Ts14-GFP neuroregenerative ability was similar to that found for the parental line Ts14 and it can be assayed using in vivo transplantation experimental paradigms, after spinal cord or optic nerve damage. Additionally, we have engineered a low-regenerative ihOEG line, hTL2, using lentiviral transduction of the large T antigen from SV40 virus, denominated from now on Ts12. Ts12 can be used as a low regeneration control in these experiments.
Network representations of the nervous system have been useful for the understanding of brain phenomena such as perception, motor coordination, and memory. Although brains are composed of both neurons and glial cells, neuron-glial networks have been little studied so far. Given the emergent role of glial cells in information transmission in the brain, we developed a mathematical representation for neuron-glial networks ($Υ$-graph). We also defined the concepts of isomorphisms, unnested form (multidigraph) and matrix equation for $Υ$-graphs. Although we found several network motives where the isomorphism between unnested forms does not guarantees the isomorphism between their respective $Υ$-graphs, we found that if the matrix equations satisfy some conditions, the unnested forms isomorphism guarantees the isomorphism between $Υ$-graphs. Finally, we introduced a novel approach to modeling the network shape. Our work presents a mathematical framework for working with neuron-glia networks.
In these companion papers, we study how the interrelated dynamics of sodium and potassium affect the excitability of neurons, the occurrence of seizures, and the stability of persistent states of activity. We seek to study these dynamics with respect to the following compartments: neurons, glia, and extracellular space. We are particularly interested in the slower time-scale dynamics that determine overall excitability, and set the stage for transient episodes of persistent oscillations, working memory, or seizures. In this second of two companion papers, we present an ionic current network model composed of populations of Hodgkin-Huxley type excitatory and inhibitory neurons embedded within extracellular space and glia, in order to investigate the role of micro-environmental ionic dynamics on the stability of persistent activity. We show that these networks reproduce seizure-like activity if glial cells fail to maintain the proper micro-environmental conditions surrounding neurons, and produce several experimentally testable predictions. Our work suggests that the stability of persistent states to perturbation is set by glial activity, and that how the response to such perturbatio
Learning and memory are acquired through long-lasting changes in synapses. In the simplest models, such synaptic potentiation typically leads to runaway excitation, but in reality there must exist processes that robustly preserve overall stability of the neural system dynamics. How is this accomplished? Various approaches to this basic question have been considered. Here we propose a particularly compelling and natural mechanism for preserving stability of learning neural systems. This mechanism is based on the global processes by which metabolic resources are distributed to the neurons by glial cells. Specifically, we introduce and study a model comprised of two interacting networks: a model neural network interconnected by synapses which undergo spike-timing dependent plasticity (STDP); and a model glial network interconnected by gap junctions which diffusively transport metabolic resources among the glia and, ultimately, to neural synapses where they are consumed. Our main result is that the biophysical constraints imposed by diffusive transport of metabolic resources through the glial network can prevent runaway growth of synaptic strength, both during ongoing activity and duri
Despite the molecular evidence that close to linear steady state I-V relationship in mammalian astrocytes reflects a total current resulting from more than one differently regulated K+ conductances, detailed ODE models of membrane voltage Vm incorporating multiple conductances are lacking. Repeated results of deregulated expressions of major K+ channels in glia, Kir4.1, in models of disease, as well as their altered rectification when assembling heteromeric Kir4.1/Kir5.1 channels have motivated us to attempt a detailed model adding the weaker potassium K2P current, in addition to Kir4.1, and study the stability of the resting state Vr. We ask whether with a deregulated Kir conductivity the nominal resting state Vr remains stable, and the cell retains a potassium electrode behavior with Vm following E_K. The minimal 2-dimensional model near Vr showed that certain alterations of Kir4.1 current may result in multistability of Vm if the model incorporates the typically observed K+ currents: Kir, K2P, and non-specific potassium leak. More specifically, a decrease or loss of outward Kir4.1 conductance introduces instability of Vr, near E_K. That happens through a fold bifurcation giving
We build on progress in understanding the role of glia cells in building the initial networks in the foetal brain. This has led us to make three significant postulates. Short term memory results from glia cells forming speculative links directly and solely as a result of neural activity generated by life experiences. These temporary 'glia bridges' create long term memory by stimulating the growth of axons, dendrites and synapses and providing the pathways enabling permanent neural structures to be created. Problem solving, idea creation and memory maintenance results from this fundamental algorithm for how the brain generates new links.
Neural tissues of the central nervous system are among the softest and most fragile in the human body, protected from mechanical perturbation by the skull and the spine. In contrast, the enteric nervous system is embedded in a compliant, contractile tissue and subject to chronic, high-magnitude mechanical stress. Do neurons and glia of the enteric nervous system display specific mechanical properties to withstand these forces? Using nano-indentation combined with immunohistochemistry and second harmonic generation imaging of collagen, we discovered that enteric ganglia in adult mice are an order of magnitude more resistant to deformation than brain tissue. We found that glia-rich regions in ganglia have a similar stiffness to neuron-rich regions and to the surrounding smooth muscle, of ~3 kPa at 3 $μ$m indentation depth and of ~7 kPa at 8 $μ$m depth. Differences in the adhesion strength of the different tissue layers to the glass indenter were scarce. The collagen shell surrounding ganglia and inter-ganglionic fibers may play a key role in strengthening the enteric nervous system to resist the manifold mechanical challenges it faces.
Effective metabolic waste clearance and maintaining ionic homeostasis are essential for the health and normal function of the central nervous system. To understand its mechanism and the role of fluid flow, we develop a multidomain electro-osmotic model of optic-nerve microcirculation that couples hydrostatic and osmotic fluid transport with electro-diffusive solute movement across axons, glia, the extracellular space, and arterial/venous/capillary perivascular spaces. Cerebrospinal fluid enters the optic nerve via the arterial parivascular space, passes both the glial and ECS before exiting through the venous parivascular space. Exchanges across astrocytic endfeet are essential and they occur in two distinct and coupled paths: through AQP4 on glial membranes and gaps between glial endfeet, thus establishing a mechanistic substrate for two modes of glymphatic transport, at rest and during stimulus-evoked perturbations. Parameter sweeps show that lowering AQP4-mediated fluid permeability or PVS permeability elevates pressure, suppresses radial exchange and slows clearance, effects most pronounced for solutes reliant on PVS V export. The model reproduces baseline and stimulus-evoked f
In the field of Blind Image Quality Assessment (BIQA), accurately predicting the perceptual quality of authentically distorted images remains highly challenging due to the diverse and complex distortions present in natural environments. Although existing methods have achieved notable accuracy, their scalability is often constrained by the high cost of subjective annotation and the limited size of available datasets. Recent advances in large-scale pre-trained vision models have introduced powerful semantic and representational capabilities, yet their application to IQA tasks is hindered by substantial computational demands and suboptimal fine-tuning efficiency. To overcome these limitations, we introduce the Global-Local Interaction Adapter (GLIA), a novel framework that effectively harnesses pre-trained Vision Transformers through a dual-stream feature extraction mechanism coupled with interactive global-local fusion. By jointly retaining global semantic information and fine-grained local details, our approach delivers superior prediction accuracy and robustness while requiring significantly fewer trainable parameters. Extensive experiments on multiple benchmarks validate the effec
We present SAM4EM, a novel approach for 3D segmentation of complex neural structures in electron microscopy (EM) data by leveraging the Segment Anything Model (SAM) alongside advanced fine-tuning strategies. Our contributions include the development of a prompt-free adapter for SAM using two stage mask decoding to automatically generate prompt embeddings, a dual-stage fine-tuning method based on Low-Rank Adaptation (LoRA) for enhancing segmentation with limited annotated data, and a 3D memory attention mechanism to ensure segmentation consistency across 3D stacks. We further release a unique benchmark dataset for the segmentation of astrocytic processes and synapses. We evaluated our method on challenging neuroscience segmentation benchmarks, specifically targeting mitochondria, glia, and synapses, with significant accuracy improvements over state-of-the-art (SOTA) methods, including recent SAM-based adapters developed for the medical domain and other vision transformer-based approaches. Experimental results indicate that our approach outperforms existing solutions in the segmentation of complex processes like glia and post-synaptic densities. Our code and models are available at h
Complex systems often involve higher-order interactions which require us to go beyond their description in terms of pairwise networks. Triadic interactions are a fundamental type of higher-order interaction that occurs when one node regulates the interaction between two other nodes. Triadic interactions are found in a large variety of biological systems, from neuron-glia interactions to gene-regulation and ecosystems. However, triadic interactions have so far been mostly neglected. In this article, we propose {the Triadic Perceptron Model (TPM)} that demonstrates that triadic interactions can modulate the mutual information between the dynamical state of two linked nodes. Leveraging this result, we formulate the Triadic Interaction Mining (TRIM) algorithm to extract triadic interactions from node metadata, and we apply this framework to gene expression data, finding new candidates for triadic interactions relevant for Acute Myeloid Leukemia. Our work reveals important aspects of higher-order triadic interactions that are often ignored, yet can transform our understanding of complex systems and be applied to a large variety of systems ranging from biology to climate.
By use of complex network dynamics and graph-based machine learning, we identified critical determinants of lineage-specific plasticity across the single-cell transcriptomics of pediatric high-grade glioma (pHGGs) subtypes: IDHWT glioblastoma and K27M-mutant glioma. Our study identified network interactions regulating glioma morphogenesis via the tumor-immune microenvironment, including neurodevelopmental programs, calcium dynamics, iron metabolism, metabolic reprogramming, and feedback loops between MAPK/ERK and WNT signaling. These relationships highlight the emergence of a hybrid spectrum of cellular states navigating a disrupted neuro-differentiation hierarchy. We identified transition genes such as DKK3, NOTCH2, GATAD1, GFAP, and SEZ6L in IDHWT glioblastoma, and H3F3A, ANXA6, HES6/7, SIRT2, FXYD6, PTPRZ1, MEIS1, CXXC5, and NDUFAB1 in K27M subtypes. We also identified MTRNR2L1, GAPDH, IGF2, FKBP variants, and FXYD7 as transition genes that influence cell fate decision-making across both subsystems. Our findings suggest pHGGs are developmentally trapped in states exhibiting maladaptive behaviors, and hybrid cellular identities. In effect, tumor heterogeneity (metastability) and
Foundation models have transformed computational pathology by providing generalizable representations from large-scale histology datasets. However, existing models are predominantly trained on surgical pathology data, which is enriched for non-nervous tissue and overrepresents neoplastic, inflammatory, metabolic, and other non-neurological diseases. Neuropathology represents a markedly different domain of histopathology, characterized by unique cell types (neurons, glia, etc.), distinct cytoarchitecture, and disease-specific pathological features including neurofibrillary tangles, amyloid plaques, Lewy bodies, and pattern-specific neurodegeneration. This domain mismatch may limit the ability of general-purpose foundation models to capture the morphological patterns critical for interpreting neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, and cerebellar ataxias. To address this gap, we developed NeuroFM, a foundation model trained specifically on whole-slide images of brain tissue spanning diverse neurodegenerative pathologies. NeuroFM demonstrates superior performance compared to general-purpose models across multiple neuropathology-specific downstream
Network systems can exhibit memory effects in which the interactions between different pairs of nodes adapt in time, leading to the emergence of preferred connections, patterns, and sub-networks. To a first approximation, this memory can be modelled through a ``plastic'' Hebbian or homophily mechanism, in which edges get reinforced proportionally to the amount of information flowing through them. However, recent studies on glia-neuron networks have highlighted how memory can evolve due to more complex dynamics, including multi-level network structures and ``meta-plastic'' effects that modulate reinforcement. Inspired by those systems, here we develop a simple and general model for the dynamics of an adaptive network with an additional meta-plastic mechanism that varies the rate of Hebbian strengthening of its edge connections. The meta-plastic term acts on a second network level in which edges are grouped together, simulating local, longer time-scale effects. Specifically, we consider a biased random walk on a cyclic feed-forward network. The random walk chooses its steps according to the weights of the network edges. The weights evolve through a Hebbian mechanism modulated by a me