The front page of AI-driven biology
Thursday, October 15, 2026Thu, Oct 15
Updated 5:32 PM PT
Machine learning classifiers trained on early embryonic stages revealed that tumors widely reactivate gene expression programs from preimplantation development, with opposing effects on survival depending on the stage mimicked.
An artificial intelligence system quantified human brown adipose tissue activity across clinical scans, revealing strong associations between active brown fat, improved metabolic profiles, and reduced rates of psychiatric illness.
Agilent Technologies agreed to pay $240 million in cash upfront, plus up to $30 million in milestone payments, to acquire the digital pathology software developer Proscia.
Researchers trained machine-learning surrogates on wide parameter ranges to evaluate which patient factors drive simulated pharmacokinetic and pharmacodynamic drug trials across multiple virtual cohorts.
Investigators trained a weakly supervised model, named HistoProfile, to predict intratumoral cell densities for CD8, FOXP3, PD-1, and PD-L1 directly from standard hematoxylin and eosin (H&E) pathology images across multiple cancer types.
A machine learning tool called unduliner uses long-read sequencing data to connect specific somatic mutations with local changes in DNA methylation patterns.
Researchers adapted the DINOv3 vision transformer to classify five common bacterial pathogens from phase-contrast time-lapse imaging of cells grown in microfluidic traps without fluorescent dyes.
A new cell-tracking framework uses human-in-the-loop feedback to propagate manual corrections across an entire live-microscopy dataset rather than editing one track at a time.
Investigators paired high-content microscopy with a machine learning pseudotime model tracking dengue virus proteins over time to screen 960 repurposed compounds in infected cells.
The Advanced Research Projects Agency for Health introduced Surpass, a program using predictive computational modeling, shared trial infrastructure, common controls, and real-time analysis to modernize clinical trials.
The Medicines and Healthcare products Regulatory Agency will continually monitor artificial intelligence-based medical devices throughout their lifecycle, replacing the current one-time approval model.
Researchers introduced FD-SCoPE, an auditable language-model system designed to answer clinical questions using trial summary tables, including requests that require inferring unlisted drug targets or harmonized endpoints.
Researchers built a machine learning model that analyzes glucose metabolism in the brain to predict Alzheimer's-type tau pathology across aging and diverse dementia disorders.
CPathOGen uses conditional latent diffusion to generate paired synthetic histology images with independent controls for cell layout, nuclear shape, and tissue stain colors.
The Department of Health and Human Services has unveiled SURPASS, a five-year program run by the Advanced Research Projects Agency for Health to speed up clinical trials using artificial intelligence and computational modeling.
An evaluation of 14 frozen pathology foundation models found they fail to match a conventional convolutional neural network on neurodegenerative disease tissue, despite performing well on cancer biopsies.
Researchers have built CROWN, a foundation model designed to analyze cell samples across diverse organs, diseases, and clinical settings.
Researchers using an artificial-intelligence tool called AI-MARRVEL identified variants in the BRSK1 gene as the cause of an undiagnosed neurodevelopmental disorder across 10 patients from seven unrelated families.
In the Phase 3 Daybreak study in wet age-related macular degeneration, Kodiak Sciences showed that tarcocimab tedromer matched Regeneron's VEGF inhibitor aflibercept on the primary vision-gain endpoint at one year.
Researchers built SpaCEy, an explainable graph neural network that analyzes spatial proteomics data to forecast disease progression and overall survival without requiring prior cell-type labels.
Researchers developed a machine-learning classifier that analyzes label-free optical microscopy images of surgical tissue margins to predict where glioblastoma tumors will recur.
Researchers fine-tuned large language models on a small clinical corpus to create Dementia Language Models that generate realistic narratives reflecting cognitive impairment.
Archival duodenal biopsies classified as having no pathologic abnormality show significant clinical and gene expression variations depending on whether a child has inflammatory disease elsewhere in the digestive tract.
A mechanistic mathematical model tracking both immune cell infiltration and circulating tumor DNA (ctDNA) sheds light on pseudoprogression, a phenomenon where immunotherapy causes tumors to temporarily swell before shrinking.
Researchers built the Neoantigen Elimination Model (NEMo), a two-compartment machine learning tool that predicts which mutant tumor targets trigger immune attack by tracking which cancer subclones disappear during immunotherapy.
The US Advanced Research Projects Agency for Health has unveiled the SPECTRA program, an effort combining computational tools and biological data to personalize autism care.
Scientists developed MMAD-Risk, a multivariate survival model designed to predict long-term risk across 271 diseases simultaneously from blood plasma proteomic profiles.
OrbiMed and Qiming Venture Partners USA co-led the $25 million round, joined by Innovation Endeavors, Benchstrength, Fusion Fund, and Lumira Ventures.
Researchers released POME, a self-supervised model that creates low-dimensional numerical representations for messy patient records characterized by mixed data types and missing values.
A hierarchical transformer architecture called HTT combines report-level language embeddings with continuous-time visit intervals to analyze sequential cancer pathology records.