This research provides a comprehensive proteomic and functional analysis of the membrane-associated periodic skeleton (MPS), a lattice structure in neurons made of actin and spectrin. By utilizing mass spectrometry, the scientists identified hundreds of candidate proteins that interact with this skeleton, spanning categories like motor proteins, ion channels, and signaling molecules. Through super-resolution imaging (STORM), they confirmed that approximately 20 of these proteins exhibit the periodic distribution characteristic of the MPS. The study further demonstrates that non-muscle myosin II interacts with the MPS to regulate axonal diameter through radial contraction. Additionally, the findings suggest that this molecular framework is essential for maintaining structural integrity and facilitating interactions between different neuronal processes. Overall, the work establishes the MPS as a sophisticated signaling and structural platform vital for proper brain development and function.

References:

• Zhou, R., Han, B., Nowak, R. et al. Proteomic and functional analyses of the periodic membrane skeleton in neurons. Nat Commun 13, 3196 (2022). doi.org

This research details the creation of a high-resolution cell atlas to track how the mouse brain changes during the aging process. By utilizing advanced single-cell transcriptomics and spatial imaging, scientists mapped the molecular architecture of the frontal cortex and striatum across various life stages. The findings indicate that non-neuronal cells, such as glia and immune cells, experience much more significant alterations in state and gene expression than neurons do. Specifically, the subcortical white matter was identified as a primary site for age-related inflammation and cellular activation. Furthermore, the study compares these natural aging signatures to changes triggered by systemic inflammatory challenges, highlighting both shared and distinct biological pathways. Ultimately, this resource provides a comprehensive framework for understanding the spatial and molecular triggers of age-related cognitive decline.

References:

• Allen W E, Blosser T R, Sullivan Z A, et al. Molecular and spatial signatures of mouse brain aging at single-cell resolution[J]. Cell, 2023, 186(1): 194-208. e18.

The paper introduces RoboChem-Flex, an innovative, low-cost self-driving laboratory designed to make autonomous chemical research accessible to resource-limited institutions. By utilizing 3D-printed components and open-source software, the researchers reduced the system's entry cost to approximately $5,000, significantly lower than traditional high-end robotic platforms. The system features a flexible Python-based framework that manages hardware control and utilizes Bayesian optimization to navigate complex chemical spaces independently. Its versatility is demonstrated through six diverse case studies, ranging from photocatalysis and biocatalysis to complex cross-coupling reactions. This platform supports both fully autonomous closed-loop operations and human-in-the-loop configurations to maximize experimental efficiency. Ultimately, the source highlights a significant step toward democratizing automation in synthetic organic chemistry by lowering financial and technical barriers.

References:

• Pilon S, Savino E, Bayley O M, et al. A flexible and affordable self-driving laboratory for automated reaction optimization[J]. Nature Synthesis, 2026: 1-13.

The paper introduces CellAtria, an innovative agentic AI framework designed to automate the complex process of retrieving and analyzing single-cell RNA sequencing (scRNA-seq) data. By utilizing a large language model to orchestrate a suite of pre-validated tools, the system allows researchers to transform scientific literature or raw data into standardized results through a conversational interface. This approach avoids the common pitfalls of AI-generated code, such as hallucinations and errors, by relying on a stable, multi-actor architecture and a specialized companion pipeline called CellExpress. Ultimately, CellAtria democratizes bioinformatics by enabling scientists without extensive programming expertise to perform high-quality, reproducible data analysis in a fraction of the time required for manual workflows.

References:

• Nouri N, Artzi R, Savova V. An agentic AI framework for ingestion and standardization of single-cell RNA-seq data analysis[J]. npj Artificial Intelligence, 2026, 2(1): 8.

This research utilizes electron microscopy connectomics to establish the first comprehensive logic of how mitochondria are organized within neural circuits. By analyzing datasets from both fruit flies and mice, the authors discovered that mitochondrial size and shape are so specific to individual cell types that they can function as biological fingerprints to identify neurons. The study further reveals that mitochondria are positioned with micrometer precision along neurites, showing a strong preference for clustering near presynaptic sites and within regions of high neural activity. In the fly mushroom body, mitochondrial enrichment varies across specific axonal compartments, suggesting that organelle distribution is directly linked to synaptic plasticity and learning. These findings demonstrate that mitochondria are not randomly scattered but are systematically embedded to support the specific energetic and functional demands of brain connectivity.

References:

• Sager G, Pfeiffer P, Wu H, et al. Spatial and morphological organization of mitochondria in neurons across a connectome[J]. Science, 2025: eads6674.

Researchers have developed MitoCatch, a novel system designed to deliver healthy mitochondria to specific, disease-affected cell types to treat mitochondrial dysfunction. While mitochondrial transplantation was previously possible, it lacked the precision and efficiency necessary to target specific cells; this new technology overcomes those hurdles by using protein binders to link donor organelles to target cell surfaces. The system employs three distinct strategies—monospecific binders on either the cell or the mitochondrion, and bispecific binders—to ensure versatile and effective delivery. Once internalized, these transplanted mitochondria move within the cytosol, associate with other organelles like the endoplasmic reticulum, and successfully undergo fission and fusion with the host’s native network. Experiments in human neurons, cardiac cells, and retinal tissue demonstrate that this method can rescue degenerating cells, including those from patients with optic nerve atrophy. Ultimately, MitoCatch represents a significant advancement in organelle therapy, offering a potential strategy for treating various currently untreatable neurodegenerative and metabolic disorders.

References:

• Ayupov T, Moreno-Juan V, Curtoni S, et al. Cell-type-targeted mitochondrial transplantation rescues cell degeneration[J]. Nature, 2026: 1-11.

This research introduces a universal method for mapping the proteins on the inner surface of blood vessels without requiring genetic modification. By using a chemical perfusion technique, the researchers successfully labeled and identified the specific protein landscapes within the brain, kidney, and intestine of vertebrates. The study specifically highlights how the luminal surface of the brain vasculature changes as an organism progresses from birth through old age. Through this proteomic analysis, the authors discovered that the arginine transporter SLC7A1 and the enzyme HYAL2 are critical regulators of the blood-brain barrier's integrity. These findings establish a new toolkit for studying vascular biology and age-related diseases across various species. Overall, the work provides a comprehensive resource for understanding how molecular shifts at the blood-tissue interface impact health and development.

References:

• Zhu Z, Jiang Z, Wang Y, et al. Luminal surface proteome of the brain vasculature uncovers blood-brain barrier regulators[J]. Science, 2026, 392(6794): eaea2100.

This study introduces a spatial gene expression atlas of the healthy human liver by analyzing samples from live donors. Unlike previous references derived from deceased or diseased tissues, this research uses spatial transcriptomics to define the natural zonation of liver functions along the porto–central axis. The authors discovered that humans exhibit a pericentral shift in key metabolic activities, such as gluconeogenesis and the urea cycle, which differs significantly from mouse models and other mammals. Additionally, the study identifies specific transcriptional changes in early-stage steatosis, particularly regarding mitochondrial protein decline during lipid accumulation. This atlas serves as a high-resolution resource for understanding human hepatic biology and the cellular landscape of the liver in both health and disease.

References:

• Yakubovsky O, Bahar Halpern K, Shir S, et al. A spatial atlas of the healthy human liver from live donors[J]. Nature, 2026: 1-10.

The paper introduces PerturbFate, a highly scalable and cost-effective single-cell platform designed to map the complex regulatory programs that drive disease phenotypes. By integrating combinatorial indexing with CRISPR interference, the method simultaneously captures chromatin accessibility, nascent transcription, and mature mRNA from individual cells. The authors demonstrate its utility by profiling over 300,000 melanoma cells, specifically examining how diverse genetic alterations lead to a shared state of vemurafenib resistance. Their findings highlight a convergent dedifferentiated cell state governed by cooperative transcription factors, such as AP-1 and TEAD, which are activated across various perturbations. Furthermore, the study elucidates how different modules of the Mediator complex uniquely influence cell plasticity and therapeutic response. Ultimately, PerturbFate provides a multimodal framework for uncovering the core molecular nodes that dictate cell-state transitions in cancer.

References:

• Xu Z, Lu Z, Ugurbil A, et al. Mapping convergent regulators of melanoma drug resistance by PerturbFate[J]. Nature, 2026: 1-11.

This research identifies a specific neural pathway that controls maternal aggression in mice, specifically the circuit connecting the posterior amygdala (PA) to the ventromedial hypothalamus (VMHvl). The study demonstrates that PA cells expressing estrogen receptor alpha are essential for driving protective attacks against intruders, showing significantly higher activity and synaptic strength during the lactation period compared to the virgin state. Scientists found that the hormone oxytocin acts as a key modulator, boosting the output of this circuit to increase aggression when pups are present. Furthermore, the intrinsic excitability of specialized neurons in the hypothalamus increases after birth, making the mother more responsive to social threats. When the offspring are removed, a decline in oxytocin levels causes a corresponding drop in aggressive behavior, highlighting a plastic neural mechanism that adapts to the mother's immediate needs. Overall, the sources reveal how hormonal shifts and structural brain changes collectively implement the rise and fall of maternal protective instincts.

References:

• Yamaguchi T, Yan R, Khan M, et al. The neural mechanisms supporting the rise and fall of maternal aggression[J]. Nature, 2026: 1-11.