AI model predicts spatial gene expression from histopathology slides in breast cancer, identifying subgroups with distinct survival outcomes and improving treatment predictions.
- Open access
- 17 cites
Moving in Cell, Genome Research, Bioinformatics, eLife, GigaScience, Nature, Nature Immunology.
Rebuilt
AI model predicts spatial gene expression from histopathology slides in breast cancer, identifying subgroups with distinct survival outcomes and improving treatment predictions.
Universal cell embedding (UCE) creates a single biological latent space that accurately represents 36 million cells across eight species, over 1,000 cell types, and diverse tissues.
Distinct pediatric intestinal cell states at diagnosis predict Crohn's disease severity and response to anti-TNF therapy, with treatment shifting toward a more refractory adult-like profile.
scBaseCount is the largest public repository of single-cell gene expression data, containing over 502 million cells from 27 organisms and 75 tissues, standardized through AI automation.
Tahoe-100M maps 100 million single-cell transcriptomes across 50 cancer cell lines under 1,100 drug conditions, revealing detailed drug-induced molecular phenotypes.
A novel immunomodulatory CAF population expressing CHL1 promotes Treg recruitment and tumor immune suppression in lung cancer.
FIERCE accurately reconstructs cellular differentiation trajectories without prior assumptions, outperforming existing methods in complex systems.
iPSC-derived neurons predominantly retain fetal-like chromatin architecture, with 96 high-coverage contact maps revealing strong compartmentalization and developmental gene enrichment.
Moderated experimental designs can effectively balance batch-effect reduction and cell loss, with cell loss doubling for each additional hashtag used in single-cell experiments.
Newest first · last 60 days
Macrophage-induced senescent cancer-associated fibroblasts drive 78% chemoresistance in colorectal cancer via SASP factors IL6 and CXCL12.
Persistent expansion of alloreactive T cell clones and phenotypic plasticity drive severe graft-versus-host disease in human transplant recipients.
A spatial atlas of muscle-invasive bladder cancer uncovers lineage-specific vulnerabilities and immune structures across 22 tumors.
CAR T cell therapy targeting CD19 can induce long-term remission of intestinal inflammation in ulcerative colitis, revealing a new therapeutic avenue for IBD.
Deep learning genetic perturbation models outperform uninformative baselines when evaluated with calibrated metrics across multiple datasets.
Multimodal lineage tracing combined with advanced computational methods enables detailed reconstruction of cellular development with high accuracy and resolution.
Multi-sample spatial transcriptomics reveals distinct regional fibrotic programs and cellular niches in human IPF, with 10X Xenium validation confirming key findings.
Senescent-like neutrophils expressing BHLHE40 promote immunosuppressive and angiogenic niches in colorectal cancer liver metastasis, correlating with poor prognosis.
Single-cell genomics reveals diverse pathogenic immune cell subsets in rheumatoid arthritis, paving the way for targeted therapies.
3D multi-omics tumour atlases reveal complex cellular interactions and evolution, offering new insights into tumor biology and potential clinical applications.
BatchVaria reveals that subtype-protected ComBat maintains nearly unchanged subtype contributions after correction, unlike naive ComBat which reduces them by 4.23%.
Fibro-inflammatory remodeling in bone marrow is a defining feature of clonal hematopoiesis of indeterminate potential, with a significant expansion of fibroblasts and extracellular matrix changes.
ArchMap enables accessible, no-code reference-based analysis of single-cell datasets, democratizing the use of comprehensive atlases for researchers worldwide.
CytoVI enables comprehensive analysis of antibody-based single-cell data, accurately modeling 350 proteins and revealing disease-related immune cell states.
SpaCEy, an explainable graph neural network, predicts clinical outcomes from spatial proteomics data with 100% accuracy in identifying key tissue patterns.
Journals this month
Moving areas, week to 3 Oct 2026