The front page of AI-driven biology
Thursday, October 15, 2026Thu, Oct 15
Updated 6:32 PM PT
Adicet Bio reported early clinical data for prula-cel, an allogeneic, donor-derived gamma delta CAR-T cell therapy, demonstrating efficacy competitive with autologous benchmarks in patients with lupus.
Mirum Pharmaceuticals reported that subcutaneous brelovitug, an antibody targeting hepatitis surface antigen, achieved composite response rates of 56% for weekly 300 mg doses and 45% for monthly 900 mg doses, against 0% in a 24-week delayed-treatment control arm.
Roche has discontinued development of emugrobart and returned the antibody to its Japanese subsidiary Chugai after a Phase 2 trial interim analysis showed it was unlikely to meet its endpoints.
Merck & Co. acquired worldwide rights to SPR2015, an experimental small molecule targeting the cancer mutation KRAS G12D from Shanghai-based SciBrunch, paying $400 million upfront in an agreement whose potential milestones bring the ceiling to $2.13 billion, reported by some outlets as $2.1 billion.
UPMC Enterprises and Ziff Capital Partners co-led the $48 million launch round for ai3Bio, formed by combining autoimmune assets from Corner Therapeutics with University of Pittsburgh spinout Novasenta.
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.
Merck has stopped supplying Recarbrio, an intravenous antibiotic combination used for severe hospital-acquired pneumonia, complicated urinary tract infections, and resistant intra-abdominal infections in patients with few other options.
The FDA has approved Roche's anti-CD20 antibody Gazyva (obinutuzumab) for idiopathic nephrotic syndrome, a serious kidney disorder characterized by excessive protein loss in urine.
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.
The Biomedical Advanced Research and Development Authority committed an initial $18.5 million to advance SCY-247, an oral and intravenous antifungal candidate, into a Phase 2 trial for invasive fungal infections.
Tolerance Bio licensed ex-Japan rights to MT-2990, a clinical-stage human IgG1 monoclonal antibody targeting IL-33, in an agreement carrying potential milestones of up to $560 million.
Adding the Nodamura virus protein NoV B2 suppresses intracellular immune defenses against self-amplifying RNA, boosting protein production in both stem cells and regular cells, researchers at Queen Mary University of London report in Nature Communications.
Scientists built Generative Attribution Metric Analysis (GAMA), an interpretability framework based on Integrated Gradients for autoregressive biological sequence generators.
A hybrid framework pairing population-state reconstruction with sparse identification of nonlinear dynamics (SINDy) mapped unobserved cell states in Clostridium co-cultures.
Researchers built an active machine learning system called AGENT that navigated formulation space in six rounds of testing to produce solid-state mRNA lipid nanoparticles that survive storage at 37 °C for over two months.
Researchers built an XGBoost classifier trained on 42 genetic and biochemical features to detect effectors of the type VI secretion system, a molecular spear bacteria use to inject toxins into rival microbes or host cells.
Computational biologists built OmegaTherm, a sequence-based language model framework designed to jointly estimate shifts in folding free energy and melting temperature caused by single or multiple mutations.
Computational biologists evaluated whether the viral spike protein S1 subunit from 28 emerging coronaviruses could bind the human entry receptor ACE2 without relying on solved crystal structures.
Researchers introduced Tri-Modality Cross-Attention, a deep learning architecture that fuses 1D chemical text strings, 2D molecular graphs, and 3D conformations to forecast biochemical properties.
Researchers built NexuST, a biological foundation model that interleaves gene-level patterns with tissue-level spatial context during pretraining.
Researchers used deep-learning design methods to transform the cancer driver K-Ras(G12C) into an inert 19-kilodalton self-labeling protein tag called LUCI-tag that rapidly binds commercial covalent inhibitors carrying fluorescent payloads.
Researchers have developed spEnhance, a computational framework that sharpens low-resolution spatial sequencing data by combining tissue images with single-cell reference maps and gene co-expression patterns.
BiomiX 3.0 adds four integration methods to its biological data analysis platform, including PRAMIGO, a new supervised model that uses graph neural networks to connect different molecular measurement layers.
A team created GEMOT, an AI agent framework designed to infer explicit, modular mathematical equations and hidden biological states from sparse experimental data.
Researchers introduced quantitative statistical mechanical modeling, a biophysically informed neural network framework designed to predict dissociation constants directly from amino acid sequences.
Researchers used sequence- and structure-based computational models to design FlexID, an engineered ligase enzyme that labels neighboring proteins without releasing diffuse reactive tags.
Researchers built RNASeek, a 1.6-billion-parameter foundation model trained on transcriptomes across multiple phyla to generate and optimize regulatory RNA sequences.