Type 2 deiodinase (DIO2), the enzyme converting T4 to T3, is a conserved marker of cancer-associated fibroblasts (CAF) across thyroid cancer and six additional solid tumor types, and is enriched in myofibroblastic CAF. Elevated DIO2 expression correlated with worse overall survival in pancreatic and bladder cancers, while thyroid cancer showed an inverse association linked to loss of epithelial DIO2 during dedifferentiation. In a TPO-CreERT2/BrafV600E/Trp53-deficient mouse model, Dio2 loss in the tumor microenvironment significantly reduced tumor growth, and D2 activity was suppressed by cefuroxime and lenvatinib.
A single-nucleus transcriptome-wide association study across diverse populations links genetic risk for human brain disorders to specific cell types. The approach reveals previously hidden gene-trait associations and conserved, cell-type-specific mechanisms. It changes how genetic risk for brain disorders can be assigned to the cell types involved.
Integrating public IBD single-cell datasets with 10X Xenium spatial transcriptomics, researchers built a map of nearly 4.3 million cells and identified an interaction circuit between FCN1+IL1B+ macrophages and inflammation-associated fibroblasts (IAFs). IAFs were enriched in the N1 and N14 pathological niches and spatially co-localized with these macrophages. Bidirectional CRISPR screens pinpointed GLIS3 as the core transcription factor driving this immune-stromal loop, providing a basis for anti-fibrotic targets in IBD.
Fluorescent protein ticker tapes (FPTT) record multiple transcriptional histories in single living cells by combining multispectral fluorescent proteins with self-assembling protein fibers. FPTT logged dose-dependent, reversible endogenous cFos transcription in hippocampal neurons at 3-hour resolution over 8 days, and variants were engineered for NF-κB, STAT3, mTOR, NFAT and cAMP signaling, enabling simultaneous analysis of four pathways during T cell activation. The tool revealed NFAT/STAT3 cross-talk and cell cycle-dependent mTOR oscillations.
A single-cell atlas of bone marrow T cells from patients with bone marrow-resident malignancies reveals tumor-reactive populations in a state of latent competence. A 15-gene signature identifies this subset, which is expanded by immunotherapies and predicts clinical responses. The work pinpoints a definable T cell state that clinicians could use to gauge who will respond to immunotherapy.
CIRI3 detects and quantifies circular RNAs in terabyte-scale RNA-seq datasets, running an order of magnitude faster than existing tools with increased accuracy through dynamic multithreaded task partitioning and a blocking search strategy for junction reads. Applied to 2,535 cancer-related samples, it identified differentially spliced circRNAs and yielded the CIRIonco database with a pretraining model and biomarker network.
UV light induces an extended array of noncanonical mutations in single-stranded DNA, including AT>AM, GT>GV, AC>AA, AT>TT, and TA>TT substitutions that resist photo-reversion and likely originate from atypical photoproducts. The work coupled temperature-induced telomeric end resection in yeast with serial UV irradiation and whole-genome sequencing. These ssDNA-specific lesions offer an additional damage-mediated route to oncogenic melanoma mutations such as BRAF V600E, which involves a GT>GA substitution.
The serotonergic projectome in the mouse brain is organized by functional relatedness of target regions rather than their physical proximity. Whole-brain axon tracing shows dorsal and median raphe serotonin neurons fall into five projectomic groups, each preferentially innervating the hippocampus, basal ganglia, cortex, medial interbrain, or brainstem/lateral thalamus. Spatial transcriptomics links each group to distinct transcriptomic clusters, and projection-specific serotonin depletion yields dissociable behavioral phenotypes.
Why it matters: It maps the full serotonin projection architecture and ties it to transcriptomics and behavior, reframing how serotonin modulation is organized.
A single-cell spatial transcriptomic atlas of metastatic breast cancer profiles 520,850 cells from 126 tumor samples of 44 patients, covering 171,766 cancer cells and 349,084 tumor microenvironment cells across primary and metastatic lesions. Multiple time points per patient revealed treatment-induced transcriptional changes driving therapy resistance, immune subpopulations predictive of hormone receptor loss, and transcriptomic features of subclones selected during progression that hold future metastatic potential.
Why it matters: It provides a large longitudinal spatial single-cell resource for studying metastatic breast cancer progression and therapy resistance.
Whole-brain mapping of serotonergic axon collateralization in the mouse reveals five projectomic groups, each preferentially targeting functionally related brain regions. Each group corresponds to specific spatial transcriptomic and behavioral profiles. The work replaces a picture of diffuse serotonin innervation with an organized architecture of five parallel projection systems.
Cell-type signatures of Alzheimer's disease are shared across Latin, white and African American population groups. Single-nucleus RNA sequencing and ATAC-seq of post-mortem cortical and subcortical brain tissue identified disease-associated microglial, astrocytic and neuronal clusters in all three groups, plus continuous gene-expression factors in astrocytes and oligodendrocytes. Molecular programs defined six subgroups of individuals with cognitive impairment that span all three populations and are not captured by neuropathology.
Why it matters: Representative multi-population sampling shows which Alzheimer's cell-type signatures are shared and reveals molecularly defined disease subgroups.
A multi-omics study led by Fan Jia's team at Zhongshan Hospital, published in Science, used Stereo-seq, single-cell and chromatin accessibility data to identify two key events in liver cancer metastatic colonization: disseminated tumor cells with high Phgdh expression establish an immune-desert niche in a quiescent state, and Cx3cr1-positive macrophages later drive metastatic reactivation. PHGDH-positive tumor cell features were cross-validated in public single-cell datasets from lung adenocarcinoma and ovarian cancer.
Light-sheet live imaging of late-stage preimplantation human embryos, enabled by mRNA electroporation of nuclear DNA labels, shows chromosome segregation errors arising de novo immediately before implantation, including multipolar spindles, lagging chromosomes, misalignment and mitotic slippage. Most lagging chromosomes are passively inherited rather than reincorporated, and most labeled cells remain externally positioned, consistent with placental fate. The findings question clinical uses of preimplantation genetic testing for aneuploidy.
Thalamic input increases production of cortical excitatory neurons during human cortical development, according to fused cortical and thalamic organoids analyzed by single-nuclei RNA-sequencing and cellular imaging. Neurexin-1 (NRXN1) mediates physical contact between thalamic axons and cortical outer radial glia; genetic knockout of thalamic NRXN1 reduced these contacts and attenuated production of upper-layer excitatory neurons. The work identifies a thalamic mechanism controlling human cortical progenitors.
A spatial transcriptomics method called 3D DNase-Enhanced Expression Profiling captures gene expression hundreds of microns deep in cleared intact tissue, and applied to neonatal mouse skin it mapped hundreds of developing hair follicles in 3D. Ordering follicles by molecularly inferred developmental age turned one snapshot into a four-dimensional map of organogenesis, revealing stem cell compartment stratification, new cell subtypes, and hair canal formation. In Foxn1-deficient nude mice, organ-wide delays and reduced coordination appeared before overt structural defects.
A heat-wide association study of 916,904 acute-care visits among 372,140 adults across four Chicago health systems (2011 to 2023) identifies diagnoses linked to extreme heat beyond the usual restricted categories. After quasi-Poisson screening and case-crossover modeling, 33 clinically reportable diagnoses showed elevated cumulative odds over lags 0 to 3, including fluid-balance and renal disorders, dermatologic conditions, multiple sclerosis, and injuries. The full-diagnostic-spectrum approach broadens the clinical picture of heat-associated morbidity.
Cell-type-specific eQTLs, mapped with a variance component model applied to population-scale single-cell RNA-seq of peripheral blood mononuclear cells from the OneK1K cohort, enrich for complex trait heritability, unlike cell-type-shared eQTLs. Trans eQTLs were mostly cell-type-specific (60%) while cis eQTLs were mostly shared (30% specific). Findings, replicated in a second dataset across ancestries, explain partly why known eQTLs are depleted in gene regulatory effects on complex traits.
Malva is a computational platform that lets researchers search raw sequence space across single-cell and spatial transcriptomics data without a reference genome, at species-agnostic scale. Its expanding index already covers about 74 million cells from thousands of experiments in health and disease, supporting searches for any sequence, mutation, splice junction or pathogen, and reference-free cell type identification. It turns single-cell atlases from static gene count tables into dynamic, sequence-resolved resources.
The Human Protein Atlas now includes a multiplex tissue resource built on large-scale multiplex immunohistochemistry, combining optimized marker panels with candidate proteins to place expression in cellular and subcellular context. Version 25 covers 1106 proteins across eight antibody panels and nine tissue settings, including testis, kidney, endocrine pancreas and salivary gland. The resource refines spatial protein annotation where single-marker staining falls short.
Lightsheet microscopy on intact tumors pinpointed junctions between precancerous and invasive regions, and a multimodal serial-section workflow combining volumetric reconstruction with spatial transcriptomics analyzed 319 spatial assays from 51 cases. In breast cancer, loss of MGP and PLAT marked and promoted invasive transition; in prostate cancer, GDF15, ALDH1A3, ANPEP and FASN rose with invasion and their knockdown suppressed PC-3 proliferation and migration. The approach exposes spatially defined drivers of invasion and candidate targets in the two most prevalent cancers.
