🧠 Neurobiology & Neurodegeneration
Single-cell multiomics connects 3D genome and transcriptome alterations in Alzheimer’s disease.
Using GAGE-seq, the authors jointly profile 3D chromatin architecture and gene expression in single cells from Alzheimer’s disease and control brains, revealing extensive cell type–specific chromatin reorganization. Integration with spatial transcriptomics and chromatin accessibility shows how genome compartment remodeling and regulatory element rewiring underlie disease-relevant expression changes, supported by a deep learning model (Hicformer) that leverages 3D genome features.
Impact: Establishes higher-order chromatin architecture as a key layer of Alzheimer’s pathology and provides a multi-scale spatial genomics framework for neurodegeneration research.
Zhang Y et al., https://doi.org/10.1126/science.adz1652
Decoding the spatiotemporal development of human meninges.
Single-cell spatiotemporal transcriptomics across 6–23 gestational weeks maps how the three human meningeal layers form asynchronously, with pia mater developing first. The study defines layer-specific fibroblast states, diverse immune niches including meningeal-specific macrophages, and CXCL12–CXCR4-driven recruitment and spatial organization of immune cells in the developing leptomeninges.
Impact: Provides a spatiotemporal atlas of human meningeal development that links barrier formation, immune niches, and CNS maturation.
Li Y et al., https://doi.org/10.1016/j.cell.2026.04.040
Multimodal imaging of gene expression, morphology, and activity of the same neuron.
The authors develop a trimodal platform that combines in vivo calcium imaging, whole-brain projection mapping, and spatially resolved gene expression profiling for the very same neuron. This approach directly links neuronal activity patterns to circuit connectivity and molecular identity, enabling integrative analyses of neuronal function at single-cell resolution.
Impact: Delivers a powerful multimodal framework to connect spatial gene expression with structure and function in individual neurons.
Zhao Y et al., https://doi.org/10.1016/j.cell.2026.05.041
Spatial multi-omics identifies early synaptic pruning and context-specific dopaminergic vulnerability in synucleinopathies.
By integrating spatial transcriptomics, spatial proteomics, and α-synuclein seeding assays in human midbrain, this study uncovers disease-context–dependent relationships between α-syn pathology and dopaminergic neuron loss across PD, incidental Lewy body disease, AD, and AD with Lewy pathology. In early incidental Lewy body disease, they detect complement C1QC upregulation and loss of inhibitory synaptic markers before overt pathology, implicating complement-mediated synaptic remodeling as a precursor to neurodegeneration.
Impact: Reveals early, spatially localized complement-associated synaptic pruning as a candidate initiating event in Parkinson’s disease spectrum disorders.
Rumpf SL et al., https://doi.org/10.1038/s41467-026-74961-6
Focal astrocyte loss reveals nuclear translocation during lesion repopulation.
Using longitudinal in vivo two-photon imaging plus spatial transcriptomics after targeted aquaporin-4 antibody–mediated astrocyte ablation, the authors track how astrocytes repopulate focal cortical lesions. Perilesional astrocytes proliferate, extend polarized processes, form transient multinucleated cells, and gradually translocate nuclei into the depleted territory, accompanied by an injury-associated transcriptional program that resolves as networks are restored.
Impact: Defines the spatiotemporal cellular and molecular choreography of astrocyte network regeneration after focal loss.
Herwerth M et al., https://doi.org/10.1038/s41593-026-02354-5
🧪 Technology & Methods Development
Uncovering spatially resolved functional genomics with CRISPR screen sequencing.
This work introduces SPAC-seq, a high-throughput spatial CRISPR screening platform, together with TARDIS, a statistical toolkit to extract spatial perturbation signatures. Applying these tools in tumors, the authors map how specific gene knockouts, such as Icam1 loss, reshape spatial immune landscapes and macrophage polarization to drive metastasis.
Impact: Establishes a scalable pipeline to directly connect gene perturbations with spatial phenotypes and signaling pathways in situ.
Zhang H et al., https://doi.org/10.1016/j.cell.2026.04.049
SRLST: a unified multimodal representation learning framework for spatial transcriptomics analysis.
SRLST presents a machine learning framework that jointly learns from gene expression, spatial coordinates, and histology images to better delineate spatial domains in ST datasets. By unifying these modalities into a single latent representation, the method improves tissue structure detection and downstream analyses over unimodal approaches.
Impact: Provides a robust computational backbone for multimodal integration and spatial domain discovery in spatial transcriptomics data.
Lan W et al., https://doi.org/10.1093/bioinformatics/btag524
🧬 Cancer Research & Immuno-oncology
Quantitative calibration of a spatial QSP model identifies fibroblast impact on HCC immunotherapy.
Extending a spatial quantitative systems pharmacology (spQSP) model of liver cancer, the authors add a mechanistic fibroblast module and calibrate it using spatial molecular data via an Approximate Bayesian Computation–Sequential Monte Carlo pipeline. The calibrated model recapitulates fibroblast-driven exclusion of lymphocytes, predicts spatial tumor states after combination immunotherapy, and nominates spatial and non-spatial pretreatment biomarkers of response.
Impact: Demonstrates how integrating spatial transcriptomics with mechanistic spQSP modeling can quantify stromal–immune interactions and guide personalized HCC immunotherapy.
Zhang S et al., https://doi.org/10.1073/pnas.2525799123
Single-nucleus multimodal spatial transcriptomics reveals spatial colocalization of neoantigen-expressing tumor cells and cognate T cells.
The Slide-GoTags platform combines droplet-based single-nucleus spatial transcriptomics, targeted neoantigen genotyping, and TCR sequencing on the same frozen tissue section to map T cell–tumor interactions in situ. In mouse and human tumors, the method identifies clonally expanded, neoantigen-specific T cells that colocalize with their cognate neoantigen-expressing tumor cells and reveals distinct immunotherapy-shaped spatial immune niches.
Impact: Offers a powerful spatial multi-omics workflow for pinpointing functional neoantigen–T cell interactions directly in human tumors.
Nagler A et al., https://doi.org/10.1038/s41587-026-03194-1
Crosstalk Between Schwann Cells and CD4+ T Cells Promotes Perineural Invasion in Colorectal Cancer.
Combining single-cell RNA-seq, spatial transcriptomics, and multiplex IHC, this study identifies a HSPA6⁺ CD4⁺ T stress (Tstr) cell subset that colocalizes with Schwann cells in perineural invasion niches in colorectal cancer. Mechanistic work shows bidirectional signaling between Schwann cells and Tstr cells via MAPK and JAK–STAT pathways, TNFα–CCL4–mediated T cell recruitment, and demonstrates that blocking TNFα and PD-1 synergistically impairs perineural invasion and tumor progression.
Impact: Pinpoints a spatially organized Schwann cell–CD4⁺ T cell circuit as a therapeutic target to block perineural invasion in colorectal cancer.
Liu H et al., https://doi.org/10.1158/0008-5472.CAN-25-5781