Biological systems exhibit oscillatory dynamics across spatial and temporal scales. From intracellular signaling to organ-level physiology, these rhythms support coordinated regulation and collective behavior. The resulting signals are nonstationary and reflect the superposition of multiple physiological processes, posing challenges for conventional analytical approaches. By providing time-frequency localization, wavelet analysis offers a powerful framework for characterizing transient oscillations and their scale-dependent organization. Recently, its integration with functional connectivity enabled quantification of time- and frequency-resolved interactions among biological units and the construction of frequency-specific functional networks. In turn, multilayer representations provide a foundation for examining network organization across frequency bands and their interrelations. Motivated by these developments, this review provides an integrated overview of the principles of wavelet analysis and its applications in biomedical research. Using examples from cardiovascular physiology, neuroscience, and pancreatic islet calcium dynamics, we illustrate how wavelet-based methods extend from the time-frequency characterization of individual signals to the analysis of functional interactions and collective network dynamics. Overall, wavelet analysis emerges as a versatile framework for linking nonstationary multiscale dynamics with physiological coordination and network organization. Advances in experimental resolution, computational methodology, and data-driven analysis are likely to extend its value, particularly when wavelet-derived features are integrated with functional connectivity, multilayer network models, and machine learning approaches. Such combinations may transform complex physiological recordings into interpretable representations of coordinated function across scales. Wavelet-based analysis thus has the potential to become a key methodological tool for studying complex biological systems, from intracellular processes to organs, in health and disease.
Excitability is a fundamental dynamical paradigm underlying both local and collective activity across a broad range of living systems, from neurons and cardiomyocytes to pancreatic β-cells, cancer cells, and the emerging field of network physiology. This review summarizes the current state of research on coherence-incoherence patterns in coupled excitable systems, covering theoretical advances and experimental evidence for their roles in physiological and pathological processes. Particular emphasis is placed on how excitable dynamics modifies the mechanisms of pattern formation relative to coupled oscillator networks, highlighting the importance of inhibitory/repulsive interactions and the constructive role of noise through phenomena such as coherence resonance, in generating pattern classes characteristic of excitable media, including bumps, patched patterns, and noise-facilitated chimera states. We further discuss how existing theoretical concepts can be extended to biological systems characterized by heterogeneous local dynamics, complex coupling architectures, and metastable behavior. In addition, we survey state-of-the-art electrophysiological and optical imaging techniques for observing coherence-incoherence patterns and assess current evidence linking them to sleep, cognition, spatial navigation, epilepsy, cardiac arrhythmia, pancreatic islet dynamics, and cancer progression. Finally, we outline major open challenges, including characterization of chaos, long-term dynamics and finite-size effects, experimental validation, control of pattern emergence/termination and switching, and the development of biologically realistic theoretical frameworks.
First discovered by L.R. Taylor in 1961, Taylor's power law (TPL) describes the relationship between the mean (M) and variance (V) of population abundances as a power function: V=aMb. For more than six decades, TPL has been documented across a remarkable range of fields, including entomology, ecology, epidemiology, physics, number theory, economics, computer science, and microbiome science. This “universality” has inspired two parallel lines of inquiry: one from mathematicians and statisticians, who discovered a convergence theorem (analogous to the Central Limit Theorem) explaining TPL's ubiquity via Tweedie distributions; and another from biologists, physicists, and social scientists, who have sought system-specific generative mechanisms. This review integrates these dual prongs across a "punctuated landscape" defined by three chronological periods (1960s–1980s, 1990s–2000s, 2010s–2020s) and eight thematic areas: population aggregation and ecological mechanisms; skewed statistical distributions; generating mechanisms; sample versus population TPL; stability, synchrony, and early warning signals for tipping points; TPL on complex networks; TPL in macrobial ecology; and TPL in microbiome ecology. We argue that TPL is not merely an empirical curiosity but a fundamental statistical regularity—anti-Gaussian, cross-scale invariant, and linked to phase transitions—that provides a quantitative lens for studying heterogeneity and stability in complex systems. We explicitly distinguish TPL from other power laws (scale-free networks, 1/f noise, allometric scaling, diversity-area relationship) and discuss their relationships. We also identify three future directions: integrating abstract and concrete prongs, developing TPL-based heterogeneity analysis and stability analysis, and exploring evolutionary perspectives on TPL.
Our perception of input from one sensory modality is malleable by inputs from other sensory modalities, also when they are non-coincident in time. This function has been critical for survival throughout evolution. How it arises may reveal fundamental principles of brain operation and inform the design of more intuitive human-machine interfaces and Virtual Reality (VR). Here we outline how the brain can acquire what we refer to as cross-modal invariants, brain representations that are applicable across any context to integrate information from multiple modalities. We begin by reviewing evidence that early unimodal brain circuitry processing of tactile inputs in the CNS, before the cortex, is organized around kinematic invariants. Published data is compatible with kinematic invariants in retinal processing, too. We show how such invariants can form natural components of biologically adapted versions of the previously introduced plenhaptic and plenoptic functions, plus a proposed matching plenauditory function. Behavioral observations, we argue, indicate that cross-modal invariants in addition rely on the capability of recording forward information in working memory and therefore this function should be dominated by the cortex. Whereas the unimodal invariants appear to reside in pre-cortical structures and have a ground truth in physics, we explain why cross-modal invariants have not. Therefore cross-modal invariants will need to be individually learned. This likely renders them intelligible only from the point of view of the intrinsic functional organization of the circuitry operations that emerge from that learning. Hence, cross-modal invariants can be expected to be more individually variable and furthermore ubiquitous in the representation space of the cortical circuitry, which can explain why cross-modal integration in the brain has proven so elusive to map out and understand.
The extracellular space (ECS) of the brain and its associated interstitial fluid (ISF)-treated as a single functional entity-make up an ever-dynamic but nevertheless functionally restricted medium within the brain. This system is critical for the transport of substances, diffusion of extracellular signals and preservation of neural homeostasis, while also being closely involved in neuronal activity regulation, metabolic waste drainage, disease progression and brain disease development. While many relevant studies have been documented in recent years, the microscopic geometric heterogeneity of the ECS, spatiotemporal dynamics in the ISF, and coupled organization of neurons-glia-vasculature across multiple scales are currently lacking from a broad and operable quantitative analysis framework for interstudy evaluation. This limitation also restricts the ability to systematise understanding of the global role of brain extracellular microenvironments. The extracellular space-interstitial fluid (ECS-ISF) functional unit is summarized and operationally defined as a local, parameterized tissue domain based on ECS geometry, diffusion-dominated transport, context-dependent convective/advection terms, and dynamic regulation by cells and vasculature as part of a state-dependent reconstructible and coordinated system. The geometric and physical properties of the brain ECS, as well as mechanisms regulating ISF generation, transport and exchange in this functional unit that governs tissue organization and contributes to metabolic regulation and brain homeostasis under normal versus pathological conditions, are summarized. In addition, recent quantitative research techniques for investigating modifications of the ECS-ISF system at various scales are summarized, including high-resolution imaging approaches, tracking methods, diffusion-convection transport models and multiscale computational simulation analyses, as well as collaborative multiphysics approaches. Their strengths and weaknesses are evaluated with respect to spatio-temporal resolution, parameter interpretability and applied scale. This review proposes the signal propagation efficiency index (SPEI) as a dimensionless reporting vector for signal-carrier propagation around explicitly defined ECS-ISF functional units (E-I units) by combining structural variables, dynamic indices and functional phenotypes, while highlighting that it remains a conceptual and testable framework rather than a validated biomarker. This perspective further outlines future research directions driven by the functional unit- and quantitative framework to advance fundamental principles of complex brain organization and the pathology of prevalent brain disorders.
Sepsis is an infection-induced syndrome characterized by systemic immune dysregulation and has traditionally been treated primarily with antibiotics. Although antibiotics remain essential for pathogen control, they rarely reverse immune dysfunction, barrier disruption, or microbial ecological imbalance in sepsis, suggesting that a purely anti-infective paradigm may be insufficient. In such phenotypes, microbiome-centered approaches may complement conventional anti-infective strategies by supporting microbial ecological restoration, host immune recalibration and disease tolerance. There is increasing evidence that the microbiome plays an important role in sepsis pathogenesis and immune modulation, both as a byproduct of immune perturbation and as a context-dependent mediator of immune maladaptation. Such insights underpin therapeutic strategies that integrate immune reprogramming with ecological restoration beyond infection control. Recent progress in single-cell omics, spatial transcriptomics and metabolomics is starting to uncover the complex interactions between microbial communities and the host immune system, providing new conceptual and translational pathways. Precise microbiome modulation combined with targeted immune recalibration may help shape future sepsis therapy, centered on restoring immune-microbial homeostasis rather than simply suppressing inflammation.
Chimeric antigen receptor T (CAR-T) cell therapy has demonstrated remarkable success in hematologic malignancies but faces persistent challenges in solid tumors and broader clinical translation, including limited target specificity, heterogeneous tumor antigen expression, unpredictable functional outcomes, and complex manufacturing processes. Recent advances in artificial intelligence (AI) are beginning to transform how these challenges are addressed by enabling data-driven design, prediction, and optimization across the entire CAR-T development pipeline. In this review, we examine how modern AI approaches such as machine learning, deep learning, and generative models are reshaping key stages of CAR-T engineering. We first discuss AI-assisted antigen discovery strategies that integrate multi-omics and clinical datasets to identify tumor-specific targets. We then examine AI-enabled engineering of antigen-recognition modules, including computational design and optimization of antibody- and TCR-derived binding domains. Next, we highlight emerging efforts to program CAR architectures and synthetic signaling circuits using AI models. We further review AI-assisted prediction of CAR-T functional performance, therapeutic efficacy, and clinical outcomes. Finally, we discuss the growing role of AI in manufacturing, quality control, and process optimization, including image-based cellular phenotyping and digital monitoring of production pipelines. Together, these advances suggest a shift from empirical CAR-T engineering toward programmable, predictive, and increasingly autonomous design frameworks that may accelerate the development of safer and more effective cellular immunotherapies.
Nucleic acid modification constitutes a pivotal regulatory mechanism in cancer, influencing the entire process of tumor development, diagnosis, treatment, and prognosis. This review delineates the role of diverse chemical modifications—including methylation, demethylation, N6-methyladenosine, and 5-methylcytosine—in governing genomic stability and cellular dysfunction across DNA, coding RNA, and non-coding RNA levels. The research paradigm of nucleic acid modification in cancer is transitioning from static modification maps to dynamic, interconnected modification networks. Moreover, the crosstalk between nucleic acid modifications and nucleic acid processing further intensifies epigenetic remodeling and oncogenic risk. Elucidating the fundamental mechanisms underlying these modifications will provide critical insights into their overarching significance in cancer initiation and maintenance, while offering promising diagnostic and therapeutic targets for precise targeted treatment of cancer at the nucleic acid level.
Low-frequency, low intensity ultrasound (LIUS) has emerged as a promising physical modality capable of inducing selective apoptosis of cancer cells, while sparing healthy epithelial cells and fibroblasts. Hitherto, the mechanism underlying this selectivity has been unclear, but we now propose and develop a theoretical framework linking the distinct mechanical behaviours of cancer versus healthy cells to their differential responses to LIUS. We point out that cancer cells exhibit inhomogeneous ventral stress-fiber networks, which can produce irregular focal adhesion geometry and inward membrane curvature near focal adhesions under low-intensity ultrasound (LIUS). These curvature irregularities can favor loose packing of Piezo1 channels, thereby preserving their activity. In contrast, healthy epithelial cells and fibroblasts display more homogeneous cytoskeletal organization, which can result in more regular curvature profiles adjacent to focal adhesions. This leads to curvature-driven cholesterol redistribution, resulting in altered spatial organization of Piezo1 clusters and reduced coordinated channel activity and allowing cells to remain in their active, proliferative state when exposed to LIUS. Based on theoretical modeling and previous experimental findings, we propose that differences in cytoskeletal organization and membrane curvature can contribute to distinct Piezo1 activation patterns between healthy and cancerous cells. Our analysis identifies curvature-mediated Piezo1 redistribution as a potential physical basis for LIUS selectivity and provides a mechanistic foundation for designing ultrasound-based therapies to exploit the intrinsic cytoskeletal vulnerabilities of cancer cells.
