Spatial transcriptomics (ST) maps gene expression within intact tissues, but its cost and technical complexity limit broad adoption. A growing body of deep-learning methods now seeks to infer ST directly from routine histology images, especially hematoxylin and eosin (H&E)-stained slides, with the aim of turning archived pathology material into virtual molecular data. This review examines more than 40 ST prediction models, comparing their data requirements, modelling strategies, evaluation practices, and translational potential. Recent advances in model architecture, histology foundation models, and higher-resolution in situ ST platforms have improved prediction performance and expanded the range of potential applications. However, most current methods still show limited robustness and generalisability, with performance constrained primarily by the quantity, quality, and diversity of available ST training data. Comparisons between models also remain difficult due to inconsistent pre-processing pipelines, training datasets and evaluation strategies. The field is increasingly recognising that progress depends not only on more sophisticated models, but also on standardised benchmarks, improved data harmonisation and clearer evaluation frameworks. Despite these limitations, ST prediction is already showing utility in research settings, including biomarker discovery, tissue domain inference, molecular super-resolution and large-scale analysis of archived histology cohorts. Emerging applications also include patient stratification, virtual molecular profiling and multi-modal pathology systems that integrate histology, transcriptomics and language models. As datasets continue to expand and models become more reliable, ST prediction has the potential to become an important component of next-generation digital pathology workflows.

Understanding how molecular mechanisms occur and shape brain function and dysfunction remains a central challenge in neuroscience. Although bulk omics methods have contributed significantly to the field, they fail to address the cellular and spatial heterogeneity of the brain. Single-cell and spatial multi-omics approaches emerged to address these limitations by enabling integrated, high-resolution profiling of molecular layers while preserving cellular and tissue context. However, despite their impact on basic neuroscience, the clinical translation of these methods remains limited by cost, technical complexity, and analytical challenges. In this review, we summarize recent advances in single-cell and spatial multi-omics applied to brain research, critically evaluating their technological capabilities, translational potential, and current limitations. We further highlight emerging directions, including spatiotemporal integration, morphomics, improved reproducibility, and the expansion of multi-omics research to biologically and environmentally diverse populations.

Once dismissed as mere cellular waste, extracellular vesicles (EVs) have undergone a conceptual redefinition, emerging as programmable therapeutic scaffolds with broad biomedical applications. Modern EVs design has progressed past the conventional framework of localized cargo delivery to isolated recipient cells; instead, the focus has shifted toward systemic, multi-cellular niche remodeling aimed at restoring tissue-level homeostasis. This review provides a comprehensive analysis of the engineering strategies to overcome the biological bottlenecks of naive EVs, specifically rapid systemic clearance and inefficient cytosolic delivery. We detail current strategies for active loading and for bypassing endolysosomal entrapment to facilitate in-situ translation of therapeutic mRNA. Furthermore, we discuss how the synergy between engineered EVs and responsive biomaterial scaffolds provides the spatiotemporal control necessary for localized reprogramming of diseased microenvironments. Finally, by examining application paradigms across oncology, regenerative medicine, and neurodegeneration alongside existing regulatory classification frameworks, this review provides a roadmap for transitioning intelligent vesicle platforms from benchtop discovery to clinical-grade compliance.

Aims: Spatially resolved omics technologies enable investigation of cellular interactions within their local microenvironments (neighbourhoods) directly in situ. Although several computational methods have been developed for neighbourhood analysis, significant limitations remain in how neighbourhoods are defined and interrogated. Here, we present Kandinsky, a toolkit that provides a flexible and versatile framework for defining and analysing cell neighbourhoods.

Methods: We developed Kandinsky to improve flexibility in neighbourhood analysis and maximise compatibility with a wide range of spatial omics data. We therefore implemented multiple approaches for identifying cell- or spot-based neighbourhoods that serve as input for four analytical modules: differential gene or protein expression analysis, neighbourhood clustering, co-localisation/dispersion, and spatial hot and cold areas. In addition to its core functionality, Kandinsky enables the execution of external tools within the same analytical framework.

Results: We applied Kandinsky to real and simulated spatial datasets to benchmark its performance against existing methods and demonstrate its ability to uncover biologically meaningful spatial interactions. Kandinsky achieved competitive performance in terms of accuracy, memory usage, and runtime. In real datasets, it suggested transcriptional changes associated with acinar-to-beta cell reprogramming in the healthy pancreas; recapitulated stromal, immune, and tumour-associated clusters in pancreatic cancer; revealed the spatial co-localisation of myoepithelial cells with specific breast cancer subpopulations; and confirmed the association between regions of high CD74 expression and immune cell infiltration.

Conclusion: Kandinsky is a flexible and versatile toolkit for neighbourhood analysis that facilitates the exploration and interpretation of complex spatial omics data.

Metabolic stress driven by mitochondrial dysfunction underlies a wide range of human diseases, yet the same defect can be detrimental to some cells while sparing their neighbors. We argue that this paradox reflects lineage mosaicism. Tissues are built from diverse clonal lineages whose differences remain hidden until mitochondrial dysfunction unmasks them. Rather than failing uniformly, cells diverge, engaging distinct stress programs shaped by developmental history and local context. By applying lineage-resolved approaches to mitochondrial dysfunction, we can move beyond average cellular behavior to understand when, where, and why individual cells adapt, persist, or fail.

Regular aerobic exercise is associated with increased survival for patients suffering from solid tumor cancers. In the last decade, pre-clinical exercise oncology studies have begun to explore the mechanisms governing the protective effects of exercise, leading to the translation of exercise-based regimens into the clinic. However, many patients with intractable solid tumors or those diagnosed at a late stage may be physically unable to partake in exercise-based regimens, or lack access to them. In this perspective piece, authors argue that the value of pre-clinical exercise oncology work is not limited to direct translation, but should be considered as an additional means of discovery for novel anti-tumor mechanisms. When exercise-based pre-clinical work is considered as a discovery engine, mechanisms identified at the intersection of exercise physiology and tumor biology can be autonomously evaluated for their clinical potential, independent of the exercise intervention.

Extracellular vesicles (EVs) are membrane-enclosed particles released by cells carrying proteins, lipids, metabolites and nucleic acids that can alter the behavior of recipient cells. In mammalian systems, EVs have been studied extensively as important mediators of intercellular and interorgan communication in development, tissue homeostasis, immunity, regeneration, metabolism, cancer and neurobiology. In parallel, Drosophila has emerged as a powerful in vivo model for EV research owing to its genetic tractability and the availability of well-established tools for studying interorgan communication. Work in Drosophila has shown that EVs participate in synaptic cargo transfer, developmental and reproductive signaling, neuronal homeostasis and systemic immune responses. Importantly, most of the pathways that regulate endosomal sorting, multivesicular body dynamics, membrane budding and vesicle secretion are conserved between flies and mammals. This review summarizes current understanding of EV nomenclature, biogenesis, cargo selection and biological function, with emphasis on points of convergence and divergence between mammalian and Drosophila systems. It further discusses the strengths and limitations of Drosophila as a model for mammalian EV biology and highlights how comparative approaches can sharpen mechanistic insight and translational EV studies.

Predicting developmental trajectories and disease progression is a central goal of virtual biology. While conventional biological network models struggle with noise and computational complexity, black-box deep learning obscures mechanistic interpretability. In this Perspective, we introduce “gene programs” (GPs) as the essential intermediate layer to close this gap. Extracted from massive single-cell datasets, GPs provide relatively robust and interpretable modules that capture coherent biological functions. Supported by recent multi-modal and large-scale perturbation advances, GPs may reflect underlying cross-layer biophysical structures. We propose a framework where next-generation virtual cells encode GPs as latent variables in neural networks. By incorporating spatial multicellular contexts and multimodal GP extraction, this GP-centric framework will reduce data noise and allow direct perturbation-to-phenotype mapping, enabling high-fidelity simulations of complex tissue dynamics. Anchoring virtual models on GPs paves the way for highly personalized “Patient Digital Twins”, empowering predictive in silico clinical simulations and advancing mechanism-driven precision medicine.

Single-cell RNA sequencing (scRNA-seq) has transformed the profiling of cellular heterogeneity; however, tissue dissociation for scRNA-seq eliminates three critical classes of information that often define cell state: spatial organization, morphology, and protein localization. Spatial transcriptomics partially restores this context, yet many platforms are limited by cost, throughput, or experimental complexity. Single-cell transcriptomics analysis and multimodal profiling (STAMP) tackle these limitations by immobilizing cells or nuclei on a slide, enabling high-content imaging prior to molecular readout and consequently preserving each cell’s visual characteristics alongside high-plex transcript and/or protein measurements. This perspective focuses on the downstream analytical workflow for STAMP datasets using Python, illustrating how standard single-cell methods can be applied to image-derived cell-by-feature matrices enriched with per-cell covariates. The pooled MIX sub-STAMP serves as a working example to illustrate practical steps for quality control, normalization, dimensionality reduction, clustering or label transfer, and program-level interpretation. Morpho-transcriptomic coupling is highlighted as an exploratory strategy that links inferred gene programs to image-derived morphology. Collectively, these analyses establish STAMP as a practical bridge between single-cell transcriptomics and quantitative cell phenotype, enabling interpretable state inference that can be validated by imaging-derived measurements.

Spatial omics is a broad term referring to technologies that allow for biomolecules to be observed within their native tissue context. These technologies have been used by biomedical researchers to gain a better understanding of cellular interactions, tumor microenvironment dynamics, and immune cell infiltration. While the basic outputs, such as spatial coordinates, segmentation masks, and transcript/protein matrices, are provided by the instrument software, the true biological insights come from several downstream, specialized analysis steps. Since spatial omics remains a relatively new field, no unified analysis pipeline has yet been established to encompass all platforms. Most workflows are adapted from single-cell RNA sequencing analysis frameworks, while incorporating additional steps that are specific to spatial data, especially for imaging-based technologies. At the same time, the diversity of platforms, data modalities, and output formats has introduced substantial challenges for data representation, interoperability, and cross-platform integration, highlighting the need for flexible, spatially aware, and user-friendly data structures made specifically for imaging-based data, not merely adapted from other methods. This review summarizes the general analytical steps following spatial omics data acquisition, commonly used data infrastructures and tools, existing gaps, and future directions in the field.

Neurons rely on precise nutrient-sensing mechanisms to sustain proteostasis and stress resilience across a lifetime. Among these, mechanistic target of rapamycin complex 1 (mTORC1) functions as a central metabolic hub, integrating amino acid availability, growth factor signals, and energetic status to coordinate protein synthesis, autophagy, and neuronal survival. Neuronal mTORC1 regulation is highly specialised, reflecting unique metabolic demands, axonal compartmentalisation, and dependence on long-term homeostatic control that is not shared by non-neuronal cell types. Beyond canonical PI3K–Akt and AMP-activated protein kinase (AMPK) signaling, emerging evidence highlights metabolic intermediates, most notably leucine-derived acetyl-coenzyme A (AcCoA), as critical upstream regulators that couple nutrient flux to mTORC1 activity via EP300-mediated Raptor acetylation. Chronic dysregulation of these pathways drives persistent mTORC1 hyperactivation, progressive autophagy impairment, and accumulation of proteotoxic species, collectively contributing to neurodegeneration. In Alzheimer’s disease, aberrant mTORC1 activity is linked to tau hyperphosphorylation and amyloid-β accumulation; in Parkinson’s disease, to α-synuclein aggregation and mitophagy failure; in Huntington’s disease, to impaired clearance of mutant huntingtin; and in amyotrophic lateral sclerosis (ALS), to dysregulated proteostasis in motor neurons. This mini review synthesizes current understanding of neuronal mTORC1 regulation, with an emphasis on the AcCoA–acetylation axis as an emerging metabolic control mechanism, its disease-specific implications across major neurodegenerative conditions, and the therapeutic opportunities these insights reveal upstream of mTORC1.

Secreted proteins mediate intercellular and inter-organ communication and are essential for coordinating physiological processes across tissues. Advances in proteomics and proximity labeling have greatly expanded the catalog of circulating secreted factors; however, for many of these molecules, their cognate receptors and mechanisms of action remain unknown. This lack of receptor annotation represents a major bottleneck in understanding systemic signaling networks and translating secretome discoveries into biological insights. In this review, we summarize and evaluate the strengths and limitations of current strategies for deorphanizing secreted proteins, including 1) biochemical approaches such as affinity purification–mass spectrometry and crosslinking-based receptor capture, 2) genetic screening strategies in both in vivo and in vitro systems, including RNA interference and Clustered Regularly-Interspaced Short Palindromic Repeats (CRISPR)-based perturbation and activation platforms, and 3) computational frameworks based on AI-driven protein structure modeling. Finally, we outline future directions aimed at accelerating ligand–receptor identification, including multiplexed screening platforms, approaches to improve sensitivity for low-affinity interactions, synthetic biology tools that convert transient binding events into stable readouts, and integration with single-cell and spatial transcriptomic technologies. Together, these advances provide a roadmap for transforming classical ligand deorphanization into a scalable, context-aware framework for decoding inter-organ communication.

Lymphatic endothelial cells (LECs) and the lymphatic vasculature have evolved from being viewed as passive conduits for fluid drainage and metastatic dissemination to active, dynamic regulators of inflammation and tumor immunity. In solid tumors, both tumor-associated and lymph node (LN)–resident LECs engage in complex interactions with their environment to orchestrate immune processes, including antigen transport and presentation to T cells, leukocyte recruitment and trafficking via chemokine gradients, and local immune modulation through the expression of co-inhibitory ligands such as programmed death-ligand 1 (PD-L1). These multifaceted roles enable LECs to either amplify effector responses or induce tolerance, profoundly influencing the efficacy of cancer immunotherapies depending on their activation state, tissue context, and molecular programming. This minireview synthesizes and discusses recent advances in tumor lymphangiogenesis, the role of LECs and their intensive crosstalk with the immune compartments, in the coordination of anti-tumor immune responses, with particular focus on LEC-autophagy as a lipid metabolic checkpoint controlling lymph node T cell egress, and its far-reaching implications for optimizing immunotherapy outcomes in solid tumors.

Caspase-2 is a key genome-surveillance protease that functions to kill or arrest cells with abnormal chromosome content, via PIDDosome-dependent and -independent mechanisms. However, this surveillance function presents a biological paradox in the liver, where physiological polyploidy, is essential for organogenesis, genomic buffering and adaptive stress responses to maintain liver homeostasis. While short-term caspase-2 loss promotes adaptive polyploidy, our recent work demonstrates that prolonged caspase-2 deficiency drives pathogenic hyperploidy, fuelling chronic inflammation, and increased age-associated hepatocellular carcinoma in mice. These findings underscore the need for tight ploidy control in the maintenance of liver homeostasis. They also highlight potential risks for therapeutic targeting of caspase-2 in fatty liver disease and suggest that pathways governing lipid metabolism also influence long-term tumour surveillance mechanisms. This perspective discusses caspase-2 as a guardian of hepatic genome integrity, and the dual, context-dependent roles of polyploidy in liver physiology and metabolic liver disease.

Background: Cystic fibrosis (CF) is a progressive genetic disease characterized by defective ion transport, mucus accumulation, chronic infection, and inflammation that drive airway damage and ultimately end-stage lung failure. Previous studies show that high levels of proteolytic enzymes in the sputum of CF patients correlate with declining lung function, but the related effects on distal lung extracellular matrix (ECM) and immune responses are unclear.

Methods: To address this gap, induced pluripotent stem cell (iPSC) lines from healthy donors and CF patients were differentiated into macrophages, and stimulated with lipopolysaccharide (LPS) to compare their inflammatory responses. Bulk RNA sequencing, functional assays, and secreted protein profiling revealed key differences between healthy and CF-derived macrophages, providing insight into how these cells may contribute to inflammatory responses in CF patients. Further, human lung ECM from distal CF lung tissue was isolated, used to generate ECM biomaterials, and combined with iPSC-derived macrophages from healthy and CF donors in vitro. Macrophage phenotype was evaluated through cytokine profiling and RNA sequencing.

Results: CF macrophage inflammation was dysregulated, with elevated baseline IL-8, IL-18, and MCP-1 expression, and a blunted inflammatory response to CF ECM compared to healthy macrophages. By using CF ECM and healthy macrophages, we characterized how healthy cells may be altered in a persistent CF milieu after anticipated CFTR modulator therapy.

Conclusion: These findings reveal altered innate immune behavior in CF and demonstrate the utility of iPSC-derived macrophages for modeling extrinsic immune-ECM interactions in disease.