Explore the evidence
Search across discoveries, vulnerabilities and models · Compare exact models and release rhythm · Read the antibiotic evidence dossier. Open each discovery’s result, AI contribution and limits; the list summary describes your full filtered selection.
Latest and most relevant
The newest entries and the highest editorial scores of the past twelve months. Browse all discoveries.
Explore by field
The number beside each field is its discovery count.
By the numbers
Full-catalog aggregates remain unchanged by list filters. Counts measure catalog entries, which may describe overlapping results; see the counting policy. Click a row or year to filter the list below; click again to clear the filter.
All discoveries
Each entry links to the original announcement. Figures are reported by the sources.
All entries by year
Plain links to the page of each of the 1141 entries, newest first.
By field: biology · chemistry · climate · computer science · humanities · materials · mathematics · medicine · multidisciplinary · other · physics.
2026 (316) · page for 2026
- Machine-learned potentials resolve a disorder trade-off in lithium cathode transport
- Foundation model and generative AI yield 3CL protease inhibitors confirmed at sub-micromolar IC50
- WenHai-E2E: end-to-end machine learning for global eddy-resolving ocean forecasts
- Human-in-the-loop AI copilot improves topology-optimized structure design
- AI-enhanced electrocardiography for detection and prediction of dilated cardiomyopathy
- Antibody design across observed and forecasted antigenic evolution
- Unrealistic emissions scenarios help AI climate models prepare for unfamiliar futures
- GPT-6 Astra constructs mutually orthogonal Latin squares that improve lower bounds at 33 orders
- t-SNE-selected LMC radio source receives a spectroscopic AGN classification at z = 2.80
- GenDIReCT recovers the M87* ring from archival telescope data
- Wearable-based health age models for cardiovascular-kidney-metabolic risk profiling and coaching
- Machine learning reveals incomplete concordance between DNASE1L3 genotype and cfDNA phenotype
- Cross-state representation learning enables prospective discovery of dynamic protein allosteric sites
- TriSCOPE links cell-type-specific Alzheimer’s expression programs to inherited risk loci
- PREVIA: LLM ensemble predicts PTSD risk from emergency department trauma narratives
- Generative design of de novo AAV vectors with unique serological profiles
- Deep learning ECG model separates Takotsubo syndrome from anterior STEMI, with a randomized reader study
- Machine learning guides discovery of selective submicromolar CK2 inhibitors
- PCFNet-guided spectroscopy confirms four protoclusters at redshift 3.3–3.4
- Signs of hypertension detected by a smartwatch
- AlphaFold-Multimer predicts 2,895 interactions among human mitochondrial proteins
- Deep-learning denoising reveals new short-period seismic observations of known ultra-low-velocity zones
- DeltaCata guides methyltransferase engineering with improved catalytic efficiency in vitro
- First 3D structure of a prokaryotic vault particle, with Claude
- Agentic cheminformatics: new macrocycles from literature analysis
- Writing seen in PHerc. 306B for the first time in 2000 years
- LightGBM feature selection identifies a four-protein panel associated with incident chronic kidney disease
- Symbolic regression derives aircraft aerodynamic scaling across sharp-cone and blunt-body geometries
- Machine learning guided design of strong transcriptional activation domains
- Machine learning guides yeast alcohol dehydrogenase specificity for trans-2-nonenal
- LHCb reports first single-experiment evidence for B⁰ → μ⁺μ⁻ decay
- Evo 2 uses in-context learning to design de novo regulatory sequences with only tens of examples
- Collaborating LLM agents propose and test a calibration rule for diamond quantum spins
- GPT-6 Astra resolves the spectral isometry conjecture for arbitrary unital C*-algebras
- A low-rank subspace of LLM representations carries most of their brain alignment
- Machine-learning mining uncovers hidden MXenes, including 11 new rare-earth MXenes
- D2ML guided prospective ROS1 potency and selectivity optimization
- Claude's early progress in drug development
- AI agent maps proton-gated water loss in perrhenate-catalysed deoxydehydration
- OpenAI presents three results in harmonic analysis, complexity, and quantum theory
- Machine-learning-guided photoredox design experimentally enables three transformations
- AI-designed brain-penetrant HER2 inhibitor IAM1363 shows durable responses in Phase 1/1b
- Depth-aware machine learning for 3D reconstruction of fibrous porous electrodes from FIB-SEM tomography
- Chest X-rays and AI for tuberculosis screening
- Healthy neural priors improve BCI decoding in patients with motor-cortex gliomas
- CochleaNet: deep learning analysis of cochlear connectomics and gene therapy
- ForTy 2020: global map of forest types at 10 meters
- Machine learning identifies SIK1 as an epilepsy driver validated in rat models
- A language-model approach to species distribution reveals alternative stable states
- The Virtual Biotech: a multi-agent AI framework for drug development decisions
- APOLLAI, a multimodal AI system, versus EMS personnel for prehospital detection of stroke-related deficits
- AI-assisted ultrasound for dating pregnancy
- GPT-Rosalind agents connect genetic perturbations and neuronal signals
- Pinal designs functional proteins from natural-language descriptions
- Claude designs binders against five scorpion neurotoxins
- Bolero predicts cell-state-specific gene regulation from DNA across species
- ProVADA steers a protein generator to adapt proteins to non-native biological contexts
- AI agents in the lab: aptamers and proteins designed in a closed loop
- Claude designs more efficient quantum circuits
- Searching for antimicrobial peptides with Google Antigravity
- LLM-based virtual screening identifies two NIR-II fluorophores tested in vitro and in vivo
- IRIS model reconstructs signaling histories of single cells
- Deep-learning design of zinc metalloproteases
- Machine-learning phase picks constrain the 2025 Drake Passage earthquake depth and activated fault geometry
- Dense-array seismicity analysis infers a buried fault beneath the Three Gorges Reservoir area
- GPT-6 Pro proves a quadratic-size threshold suffices for random permutations to contain every length-n pattern
- SPARCS screen identifies a Golgi-pH regulator of STING signaling
- ATLAS graph neural network enables first LHC observation of single-top s-channel production
- Earth AI for public health: geospatial models across five use cases
- GPT-6 Pro proves an optimal anticoncentration bound for polynomial Littlewood–Offord inequalities
- GPT-6 Astra assists a gauge-invariant lattice chiral-fermion measure construction
- Claude discovers algorithms that refute the 3SUM and APSP time-complexity hypotheses
- Recurrent neural network reveals near-critical seeding of topological defects in Kibble–Zurek simulations
- AI-assisted proof establishes a dimension-independent KLS isoperimetric bound
- GPT-6 Astra proves a linear coloring bound for graphs excluding clique minors
- Deep-learning prioritization of metagenomic products yielded cytotoxic compounds
- SynthID Bio: watermarking for proteins and synthetic structures
- WDNet-guided DESI follow-up confirms 75 new white dwarfs and 7 cataclysmic variables
- Quine ranks compounds that shift pancreatic cancer cells toward a basal state
- PCA mapping and machine learning prioritize antiplasmodial chemotypes
- Generative AI designs and experimentally validates an ultrathin broadband ENZ absorber
- Catalysts without iridium or ruthenium for acidic oxygen evolution
- Claude identifies a new enzyme system: array-associated reverse transcriptases
- ML force-field screening identifies 12 superionic conductor candidates
- Machine learning attributes post-2015 emissions to intensification of Europe’s recent heat extremes
- RetroChimera: predicting small-molecule synthesis
- GPT-4o-assisted literature screening nominates low-lithium LiMoOBr₄ conductor
- GPT-assisted proof reduces Zieve’s Hurwitz realizability conjecture to the three-branch-point case
- LHCb observes the doubly charmed Ωcc⁺ baryon with an 8.7σ global significance
- Claude optimizes more than 30 open models for biology
- GPT-5.6 constructs a counterexample to Talagrand’s operator cotype problem
- Paper2Agent integrates perturbation datasets to prioritize GPR137 at a psoriasis risk locus
- GPT-6 Astra finds a counting proof of the Erdős–Sós tree-embedding conjecture
- Deep learning-guided synthesis yields fluid ferroelectric molecules
- AI-guided natural-product-inspired molecules inhibit two multidrug-resistance transporters
- AI-guided autonomous experiments map phase domains in bismuth–titanium oxide films
- AI-discovered proof resolves the Komlós vector-balancing conjecture
- AlphaGenome Atlas: predictive map of human DNA variation
- OpenAI proposes a solution to the Navier–Stokes problem
- Codex agents perform measurements and calibrations on a six-qubit chip
- AI-discovered construction improves the lower bound for the complex Grothendieck constant
- Claude formalizes Fermat's theorem in Lean
- YOLO detector tracks endangered European hamster activity in camera traps
- WeatherNext 3: a global AI model for weather forecasting
- Deep-learning seismic picks reveal upper-mantle upwelling conduits beneath Southwestern US volcanic fields
- ERTransformer-generated TCR candidates activate T cells in defined receptor contexts
- Global methane point sources mapped from EMIT imaging spectroscopy with a vision transformer
- External validation reveals distinct plasma proteomic signatures across biopsy-proven primary glomerular diseases
- ESM-FEP combines protein-language-model fitness screening and free-energy calculations to engineer herbicide-detoxifying maize dioxygenase
- GPT-5.6 helps prove an exact mapping between frustrated and field-driven spin chains
- Physics-informed retrieval-augmented language model guides synthesis of five previously unreported inorganic crystals
- Knowledge-guided graph learning designed a printable nickel superalloy
- Deep-learning-designed enhancers drive target-tissue reporter activity in mouse embryos
- Deep learning maps six high-potential deep-Earth scattering zones
- AI-guided experiments establish feasible 500–950 W printing configurations for GRCop-42
- MACE-guided screening yields three corundum high-entropy oxides and a cation-ordered oxide
- Claude designs protein binders for 15 targets
- LLM-guided iterative optimization yields a previously unreported Rb3BiI6 perovskite-inspired thin film
- FluEmbed predicts H3N2 antigenic drift from sequence alone
- MP4 text-to-protein design yields ATP-active proteins, including one with a novel fold
- RAI-939: an autonomous-lab alloy tested with a torch
- Data-driven equations reveal rainfall-linked North Atlantic decadal variability
- Deep learning controls electrical stimulation patterns in human visual cortex
- WeatherNext Cyclones: forecasting cyclones one day earlier
- Machine-learning-guided CsrA homolog screens reveal divergent bacterial regulatory activities
- Machine learning identifies the first enzyme for phenazine–thiol conjugation
- XunZi prioritizes CHK2 as a Parkinson’s disease mechanism and therapeutic target
- Boosted decision tree enables first observation of two simultaneously longitudinally polarized Z bosons
- ML-guided synthesis conditions produce EMT zeolite
- REAP-guided engineering improves P450 BM3 activity
- Machine learning and directed evolution improved an L-Dopa-producing monooxygenase
- OpenAI presents ten results in mathematics and theoretical computer science
- Machine learning prioritizes a protective Arabidopsis leaf fungus that suppresses oomycete infection
- Machine-learning triage identifies a dormant wandering black hole in a galaxy’s outskirts
- A quantum computer that learns from its own errors
- Machine-learning-derived seven-protein panel distinguishes remission from active ANCA-associated vasculitis
- Active learning selects an oxygen electrode for protonic ceramic cells
- Machine learning enriched MthUPO variants for beta-damascone hydroxylation
- Machine-learning-selected plasma markers distinguish Lewy body and frontotemporal dementias
- AI-guided proof search solves five open problems in Banach space theory
- Deep-learning splice prediction links a PARP2 variant to altered protein isoforms and influenza susceptibility
- EnzGFM-Agent experimentally prioritizes activity-enhancing secreted alkaline phosphatase variants
- Deep-learning ComPACT catalogue adds 116 previously unreported galaxy clusters
- Deep learning identified minimal mutations for cross-reactive anti-PD-L1 antibodies
- UniKP prioritizes a sulfotransferase for sulfotyrosine production
- Transformer-based SDSS search identifies two previously unreported optical–UV tidal disruption events
- Bayesian optimization finds mild solvent-free mechanochemical conditions for cyclopropanation and epoxidation
- Multiscale machine-learning microscopy identifies robust paleomagnetic carriers in meteorite WIS 91600
- FastNet: physically realistic AI forecasts from the Met Office
- CHAI image-only histology biomarker stratifies metastatic hormone-sensitive prostate cancer outcomes in independent trial data
- Diffusion-guided peptide design yields antimicrobial candidate Arcinin
- Quake Neural Operator builds microseismic catalogues without phase picking
- AI-prioritized CRISPR screen identifies ALOX5 and OXTR as regulators of keratinocyte IL17RA in psoriasis
- Euclid and machine-learning selection identify 31 spectroscopically confirmed quasars at redshift 6.6–7.8
- Active learning improved whole-cell antibiotic screening and validation
- AI-guided mining identifies and engineers an alkaline xylanase
- Claude-assisted analytic proof establishes a+b=1 for mean-field jamming exponents
- Machine learning guided a PET-binding module into a reusable whole-cell catalyst
- BirdNET embeddings distinguish individual North Atlantic right whales
- MSFA-Net and Balmer-line fitting identify 3,583 new blue horizontal-branch stars
- Deep-learning lens search leads to first confirmed galaxy-scale lensed superluminous supernova
- A Herculaneum scroll read end to end without opening
- Deep learning finds an ECG biomarker for sudden cardiac death
- A1: deep learning identifies an antibiotic candidate against gonorrhea
- Few-shot protein-language-model design yields high-output bicyclogermacrene synthase variants
- Observation-trained machine learning halves estimated Arctic winter sea-ice surface cooling
- MutexaGPT-guided cold-adapted amylase variants
- Deep learning identifies antimicrobial peptides hidden in prion-related proteins
- Genetic reanalysis assisted by o3 contributes to 18 diagnoses
- Deep learning guides higher-activity SpuFz1 nuclease variants
- An AI system improves a Chan–Lam coupling reaction
- Machine learning pre-screening finds kagome superconductors YRu3B2 and LuRu3B2
- Routine-laboratory LightGBM screening identifies monoclonal immunoglobulin positivity across independent hospitals
- Fung-AI prioritizes peptides with measured antifungal activity
- Machine learning identifies a low-tin lead-free relaxor composition window
- Machine-learning screening identifies bacterial-selective membrane-active compounds
- AI-designed bacterial anti-defense proteins enable phage infection and plasmid retention
- Machine-learning screening guides thermochemical measurements of salt hydrates
- Machine-learning seismicity maps a 250-km eastern edge of the subducted Yakutat microplate
- ElectrolyteGPT can generate new formulations for battery development
- Family-specific machine learning prioritizes enzyme variants for prospective testing
- Satellite model maps collapse and surviving Acropora corals in Belize
- Deep learning finds multimetallic single-atom catalysts across material families
- Deep learning reveals more than 500 previously unrecognized earthquakes beneath David Glacier
- ML ranking identifies active sesquiterpene-synthase variants
- Spatial-DC maps a candidate fibroblast–cancer signaling interaction in pancreatic tumors
- AI-driven formal proof search resolves Erdős problems and OEIS conjectures
- Machine-learning-assisted design predicts Netsene, a 3D carbon nodal-surface semimetal
- Machine-learning mapping finds broader preservation of pre-glacial landscapes across Arctic Canada and Greenland
- OpenAI presents a disproof of the unit-distance conjecture
- AI generates compounds with experimentally validated pancreatic cancer cell selectivity
- Co-Scientist: repurposed drugs against liver fibrosis
- Co-Scientist and aging: metabolism and the integrated stress response
- Co-Scientist and mechanisms of metabolic steatohepatitis (MASH)
- Co-Scientist: genetic factors to rejuvenate cells
- WeatherNext and Hurricane Melissa: earlier warning in Jamaica
- Empirical Research Assistance writes expert-level scientific code across six domains
- A graph neural network nominates OSM–OSMR for melanoma immunotherapy, with ex-vivo explant evidence
- Spectroscopic follow-up revises Gaia machine-learning white-dwarf classifications and reveals a large magnetic and carbon-rich population
- Unsupervised learning separates rockfall and glacier-sliding precursors before the Blatten collapse
- Machine learning guides a near-infrared Mo/Sb halide phosphor
- Machine learning prioritizes compounds active against heterogeneous glioblastoma stem-cell models
- BioGDR predicts cell-line sensitivity to an ALDH1B1 inhibitor
- MatterSim and tantalum phosphide as a good heat conductor
- Machine learning designs bulk metallic glasses with peak hardness near 2,450 HV
- Machine-learning metabolomics identifies widespread promiscuous activities in Escherichia coli enzymes
- MISPOP machine learning screening identifies the oncolytic peptide Dermaseptin-S9
- Machine-learning earthquake catalog reveals scale-dependent fault controls at Gonghe geothermal site
- AI-assisted analysis reveals retrospective seismicity before the Dingri earthquake
- PoLARIS: a self-driving microfluidic lab finds brighter lead-free perovskite nanoplatelets
- Fragment-guided AI protein redesign yields a Kemp eliminase and fluorophore binder
- Unsupervised seismicity categories distinguish documented preparatory phases before three large earthquakes
- New Zarankiewicz numbers and bounds from OpenEvolve search
- A machine-learning-assisted TESS search confirms a previously unknown hot Jupiter
- AI-guided design enables E. coli ribosomal proteins without isoleucine
- CellOracle-guided pancreatic islet differentiation experiments reveal STAT1 and ZEB1 lineage roles
- PEPIC identifies a BHET-hydrolyzing enzyme candidate
- REDMOD: pancreatic cancer signals in CT scans taken before diagnosis
- Qiushi Discovery Engine proposes and validates an optical bilinear interaction
- AggreProt-guided mutations improve LinB solubility and yield
- Simulated training data locates archaeological tar-kiln-like sites in Louisiana, field check finds WWII features
- Gauss completes the Lean formalization of the dimension-8 sphere packing proof
- Machine learning maps modeled fault-system CO₂ flux within active tectonic regions worldwide
- GPT-5.5 and a proof about Ramsey numbers
- Joint molecular modeling identifies seven low-micromolar kinase hits beyond its training chemistry
- Computer-vision and metabolomics fusion links paired serum–urine flux mismatch to early diabetic kidney disease
- LLM-guided experiments identify three phases unique to the ternary Co–Al–Ge system
- MetGenX annotates two previously uncharacterized metabolites in mouse liver
- ML-guided discovery of 5ANI passivator yields high-efficiency, stable perovskite solar cells
- An LLM agent in an air-free robotic lab maps lithium halide spinel conductors
- LLM-guided alloy design yields self-limiting oxidation behavior at 900 °C
- Machine-learning selection of a six-protein panel for atherosclerosis severity and patient stratification
- CALPHAD-based transfer learning designs high-strength Al–Mg–Zn alloys
- Stability-based machine learning selects IL-8 and CCL3 as a two-protein serum signature for early silicosis
- Anderson's 2014 commutative algebra problem resolved by Rethlas and Archon
- Weakly supervised deep learning maps seamount morphology patterns associated with tectonic setting
- DefensePredictor uncovers dozens of previously unknown bacterial anti-phage systems
- Self-supervised imaging search identifies 16 spectroscopically confirmed quasars at z≈5.94–6.45
- Erdős Problem #650 solved with ChatGPT strategy and Aristotle Lean proof
- AI-designed protein family binds and senses small molecules, including cortisol
- Deep active learning identifies lithium-metal battery electrolytes with longer symmetric-cell lifetime
- Catalysis AI Agent derives design rules for copper single-atom alloys for CO2 electroreduction
- Machine-learning-guided genome mining links an Amycolatopsis BGC to dipyrimicins
- Transformer-guided genetic search finds six new champion linear error-correcting codes over F₈
- Machine learning identifies covalent organic frameworks for hydrogen peroxide photosynthesis
- Theoretical physics with Claude: the Sudakov shoulder in the C parameter
- RAVEN pipeline validates 118 new TESS planets and vets over 2,000 candidates
- Claude Code produces the first primary Lund jet-plane density measurement in electron–positron collisions
- ParticleNet enables first hadron-collider W-boson mass measurement from hadronic decays
- Adaptive AI experiment planning finds multi-promoter catalysts in under 50 experiments
- ORI engineers active lysozymes, heat-tolerant chitinases and dual-activity enzymes
- Horizyn-1 experimentally identifies enzymes for orphan and nonnatural reactions
- Graph embedding identifies LC-61 with anti-Leishmania activity
- ChatHEA assistant guides high-entropy alloy catalysts for fuel cells
- Machine learning selects W-doped nickel–iron oxyhydroxide for electrolysis
- AlphaEvolve raises nine Ramsey lower bounds
- AI-assisted work proves a Latin-square imbalance bound, with optimal examples through order 52
- M3-CAD designs antimicrobial QLX-3DV peptides
- Machine learning prioritizes microbiome metabolites that stimulate interleukin-8 secretion
- Bayesian optimization identifies enzyme nanohybrids with high activity recovery
- Semi-supervised image classifiers lead to confirmation of a new lensed quasar
- Gemini derives an exact analytic solution for the cosmic-string gravitational-radiation power integral
- Time-delay neural networks expose three fault-development phases across laboratory pressure regimes
- Generative design and screening produced experimentally tested antimicrobial QACs
- GPT-5.2 Pro extends a scattering-amplitude result to gravitons
- Prospective multicentre study combines EV protein biomarkers and machine learning for preclinical oesophageal cancer detection
- HMD-AMP mining uncovers evolutionarily remote antimicrobial peptides with validated antibacterial activity
- Machine learning uses historical warming patterns to constrain future climate projections
- Biochemical-constrained generative AI yields experimentally active antimicrobial peptides against E. coli
- Claude's Cycles: a Hamiltonian cycle decomposition found by Claude
- Random-forest target selection supported nine millisecond pulsar discoveries
- First observation of single-top-quark production with a W and Z boson at CMS
- Machine learning guides a low-precious-metal RWGS catalyst
- Machine-learning analysis separates genetic from environmental drivers of stimulated cytokine responses in healthy adults
- Deep-learning tracking of mantle simulations finds continental-margin subduction zones persist longer
- ETopt-guided screening identifies thermophilic beta-agarases
- LFAST: an LLM agent drives robotic synthesis of the new COF-2000 framework
- Language model discovers general construction rules for two previously unknown families of quantum experiments
- Machine learning screening identifies GW3965 activity against invasive group A Streptococcus
- Pichia-CLM improves recombinant protein expression in Komagataella phaffii
- De novo designed proteins modulate GPCR signalling in the dopamine D1 receptor
- GPT-5.2 proposes a formula for nonzero gluon amplitudes
- Deep learning identifies 216 previously unrecognized Green Pea galaxies in DESI Legacy Survey imaging
- Gemini Deep Think on research problems in mathematics, physics, and computer science
- AI-derived histology biomarker stratifies first-line chemotherapy outcomes in advanced pancreatic cancer
- Perovskite-R1 suggests precursor additives tested in solar-cell devices
- EnzySFC discovers nitrile-hydratase activity in nitrilases and evolves amide-selective variants
- UMAP analysis of DESI spectra surfaces newly recognized inhomogeneous white dwarfs
- Fel's conjecture proved and verified in Lean
- GPT-5 reduces the cost of cell-free protein synthesis
- Reinforcement-learning-guided experiments produce a Zr-based bulk metallic glass with 1.7 GPa yield strength and 10.2% elongation
- First-principles neural network finds crystallization of fractional quantum Hall liquids
- ML-guided FeCoCrMnCu hydroxide for oxygen evolution
- Machine-learning and expert co-screening identifies two new silver-catalyzed cycloadditions
- DinoTracker: AI classifies dinosaur footprints
- AnomalyMatch finds over 1,300 unusual objects in the Hubble archive
- Machine-learning screening and experiments identify two new superconductors, Be₂HfNb₂ and Be₂Hf₂Nb
- Machine-learning seismic catalogue reveals a long-period swarm before Sierra Negra eruption onset
- Co-FunSearch finds adversarial instances raising the Best-Fit bin packing bound
- Text mining and machine learning guide Eu-activated phosphor wavelength prediction
- Iterative discovery of potent polymeric antibiotics via multi-stage and multi-task learning against antimicrobial resistance
- Claude in biology labs: Biomni, MozzareLLM, and others
- DeepMet anticipates and helps identify dozens of previously unrecognized mammalian metabolites
- AI and developmental profiling reveal conserved rules for microRNA strand selection
- Self-iterative learning finds RNA-targeting prokaryotic Argonautes and an RNA-modification-sensing activity
- Deep-learning histopathology identifies tertiary lymphoid structure features linked to stage II colorectal cancer relapse
- Erdős Problem #728 solved with GPT-5.2 Pro and Aristotle
- Tranylcypromine shows in-vitro activity against Leishmania major
- DrugCLIP screens about 10,000 human proteins against 500 million compounds
- Random-forest analysis of continuous seismic waveforms identifies five previously unreported slow-slip events in western Shikoku
- CleaveNet designs peptide substrates with MMP13 cleavage profiles
- AI-generated virtual spatial proteomics reveals lung-cancer tumor–immune niches associated with immunotherapy response
- A seven-gene machine-learning score shows an Alzheimer-associated signal in an independent plasma cohort
- Uni-Mol screening guides experimental validation of β-sheet tetrapeptide assemblies
- SMRTnet predicts small molecules that bind RNA
- Knowledge-graph AI prioritizes TPP1 as a Parkinson’s disease target linked to α-synuclein homeostasis
2025 (256) · page for 2025
- PROBY screening uncovers target-selective fluorescent probes for tau, BCL-2 and TDP-43
- A double-Lambda hypernucleus of boron-13 found with deep learning
- Machine-learning metabolomics prioritizes succinate in sepsis-associated encephalopathy
- Machine-learning prioritization and CRISPR knockout identify OsWOX13 as a rice heat-tolerance regulator
- CycleRFdiffusion-designed TP4 is a reproducible cyclic peptide binder to TREM2
- Diffusion-guided design tests bimetallic ammonia-decomposition catalysts
- AstroREG maps functional enhancer–gene links in human astrocytes
- AI-guided experiments uncover a new polymer polymorph and improve organic transistor performance
- NOAA puts three AI-based global models into operation
- Active learning selected antimicrobial polymer P7
- GPT-5 and molecular cloning: 79-fold greater efficiency
- CellPaint-POSH screening links AURKAIP1 knockout with reduced mitochondrial abundance and mTORC1 signaling
- Graph neural networks identify oxides for hydrogen production
- Machine-learning screening identified experimentally active pore-forming peptides
- ML-guided CRISPRi combinations improve isoprenol production
- GPT-5.2 Pro and an open problem about learning curves
- tFold designs binding antibodies against four defined antigens
- PICS-assisted survey reported 12 new Large Magellanic Cloud pulsars
- A self-driving laboratory discovers lens-and-grating patterns that steer spontaneous emission
- CACLENS predicts and validates a zearalenone-degrading enzyme
- Heat-resistant crops: a hybrid photosynthesis enzyme
- The levers of political persuasion with conversational AI
- RFdiffusion2 scaffolds enzyme active sites at atomic level
- Computational design of zinc metallohydrolases with RFdiffusion2
- Riff-Diff: computational enzyme design by catalytic motif scaffolding
- AI-guided subtilisin variants improve aqueous amide formation
- A 16-feature AI model distinguished Sjögren disease from clinically similar conditions using routine tests
- TARSA prioritizes peptides active against breast cancer cells
- The structure of the LDL protein apoB100
- Healthier bees: AlphaFold predicts vitellogenin's structure
- Deep learning forecasts aftershock rates in seconds, comparable to ETAS
- AI-guided experiments uncover new zeolitic imidazolate framework crystals
- popEVE: diagnosing rare developmental disorders
- GPT-5 helps solve a 40-year-old problem about Nesterov's accelerated gradient
- Data-centric machine learning prioritizes compounds for Trypanosoma cruzi assays
- RFD2-MI designs sequence-specific cysteine proteases
- EqGPT infers a previously unreported equation for steepening surface gravity waves
- Early experiments in accelerating science with GPT-5
- Machine learning-guided discovery of thermophilic carbonic anhydrases
- ML prioritization identifies two antimicrobial peptide candidates
- Evo genomic language model designs functional toxin-antitoxin and anti-CRISPR systems
- Bayesian optimization experimentally identifies a high-AHE Fe–Co–Ni–Ta–Ir amorphous film
- LightGBM guides discovery of argyrodite lithium-ion conductors
- Machine-learning metabolomics identifies DHEAS and citrate signatures in anti-MDA5 dermatomyositis
- Multicenter targeted metabolomics and machine learning identify a six-metabolite rheumatoid-arthritis diagnostic panel
- Kosmos: an AI scientist for autonomous discovery
- Mind captioning: text descriptions of viewed videos decoded from fMRI
- AlphaEvolve on 67 math problems with Terence Tao
- AI screening identifies poly-PDO for recyclable barrier packaging
- Machine learning predicts timing and fault-state evolution in meter-scale laboratory earthquakes
- Machine-learning microscopy isolates micronucleated cells and reveals chromosome-abnormality formation
- A six-metabolite machine-learning panel and isovalerylcarnitine findings for gastric cancer
- Deep-learning virtual screening finds antibacterial hits in over one billion compounds
- Explainable neural-network analysis nominated GPC6 as a Parkinson disease dopaminergic-neuron marker
- DeepSomatic: a neural network for somatic variants in tumors
- C2S-Scale: a model-generated cancer hypothesis confirmed in cells
- Deep-learning feature-guided design yields antimicrobial peptides D1 and D2
- AI-guided reactor design improves a specific continuous-flow cycloaddition
- Machine-learning metabolomics distinguishes SARS-CoV-2, influenza A and RSV from respiratory swabs
- Active learning discovers three high-entropy thermoelectrics with zT above 2
- Multi-centre validation tests AI-empowered OncoSeek across 15,122 participants and four assay platforms
- Neural-network satellite soil-moisture data expose global hotspots and pathways of rainfall coupling
- Barbarians at the Gate: AI-driven research for systems (ADRS)
- CellTransformer maps previously uncatalogued regions of the mouse brain
- ProT-VAE designs synthetic enzymes with measured activity and high-temperature stability
- A seven-metabolite AI model detected renal-cell carcinoma across six validation cohorts
- Enterololin: AI helps clarify an antibiotic and its mechanism
- HuDiff humanizes antibodies and nanobodies while retaining measured binding
- Prospective multicenter validation of TriVerity links a 29-mRNA host response to infection type and near-term severity
- Deep learning of salivary-gland biopsies identifies acinar CD8 T-cell infiltration associated with Sjögren disease
- Active learning identifies high-performing electrolyte solvents for anode-free lithium metal cells
- CRESt discovers a multielement formate-oxidation catalyst
- ML-guided optimization improves photochemical cysteine arylation
- Spectroscopic confirmation of neural-network lens candidates from DESI imaging
- DiffSyn guides synthesis of a high-silica UFI zeolite
- New families of unstable singularities in fluid equations
- First bacteriophage genomes designed by a genome language model
- Delphi-2M forecasts the risk of more than 1,000 diseases
- KPPepGen generates experimentally active pathogen-specific antimicrobial peptides
- Gauss formalizes the Strong Prime Number Theorem in Lean
- AI-assisted transfer-matrix analysis yields an exact solution for the 1D frustrated q-state Potts model
- Machine-learning analysis attributes the apparent pre-Norcia earthquake b-value rise to changing activity across two fault structures
- Machine-learning screening identifies natural RORγt inhibitors with preclinical psoriasis activity
- Machine learning selects an N-TIMP2 variant with altered MMP specificity
- Deep Loop Shaping reduces LIGO control noise
- Machine learning reveals broad and previously uncharacterized fungal polyketide-synthase activity
- AI-enhanced seismic catalog exposes a ring-fault system at Campi Flegrei
- Machine-learning-guided mining identifies 91 previously unknown PET hydrolases, including acid-active enzymes
- Machine learning guided discovery of enhanced anomalous Hall resistivity in Fe–Ir–Pt films
- Machine learning and tear-fluid proteomics identify a six-protein ALS signature
- A chemical foundation model guides high-conductivity electrolyte formulation
- BindCraft designs protein binders in one shot
- Physics-informed generative learning yields experimentally verified single-phase B2 intermetallic alloys
- New combinations of additives for LNMO batteries found with machine learning
- ESM-guided metagenome mining identifies seven new Cas12a subtypes with distinct activities
- GPT-4b micro: Yamanaka factor variants with over 50-fold higher expression
- FlowER predicts ten-step pyrazole pathways consistent with known product selectivity
- Machine-learning analysis identifies a plasma protein signature associated with ALS years before symptoms
- AlphaCD maps experimentally measured catalytic properties across cytidine deaminases and supports a high-fidelity editor
- Machine learning identifies serum protein signatures for progression from asymptomatic atherosclerosis to ischemic stroke
- Antibiotics designed with generative AI against gonorrhea and MRSA
- Decoding inner speech with a brain-computer interface
- Machine learning finds plasma protein signals up to 14 years before Parkinson diagnosis
- Machine-learning earthquake catalog reveals transform fault in the Zabargad Fracture Zone
- Machine-learning alert flags supernova 2023zkd as a black hole and star binary merger
- Machine learning uncovers new small-molecule chloride transporters from DrugBank
- PepMLM designs experimentally binding target-conditioned peptide ligands, including peptides used for targeted protein degradation
- HydrAMP-guided design of short tryptophan-rich antimicrobial peptides
- Archaeasins: deep learning finds antibiotic candidates in the archaeal proteome
- Symbolic-regression turbulence analysis separates transferable flow physics from dataset-specific closure terms
- Machine-learning analysis motivates proved criteria for Galois groups
- BEAUT uncovers new gut-microbial bile-acid enzymes and the novel bile acid 3-acetoDCA
- Machine-learning screening identifies zeolite structure-directing agents
- Blood DNA methylation machine learning signatures classified brain insulin resistance across three cohorts
- Machine learning guides screening of redox perovskites for solar hydrogen
- Physics-tailored machine learning reveals unexpected non-reciprocal forces in dusty plasmas
- AlphaEarth Foundations: a 10-meter map of Earth
- Virtual Lab: AI agents design nanobodies against SARS-CoV-2
- Active learning repurposes BMS-262084 as a potent TMPRSS2 inhibitor that blocks coronavirus entry
- LHCb makes the first high-significance observation of the rare Σ⁺ → pμ⁺μ⁻ decay
- Generative model identifies high-conductivity liquid electrolyte formulations
- An autonomous platform speeds research on polymer mixtures
- AutoBot: robotic optimization of halide perovskites
- Aeneas: restoration, dating, and provenance of Latin inscriptions
- SONAR identifies 3′UTR sequence features that raise T-cell receptor protein expression and CD8-cell killing
- EarthLink identifies an eight-month predictor set for Atlantic Niño hindcasts
- Gemini Deep Think reaches gold-medal level at the 2025 IMO
- EHR deep learning and sequencing identify two previously undiagnosed Noonan syndrome cases
- LLM-guided search yields three synthesized lithium-ion cathodes with higher initial capacity than NMC811
- Machine learning helps select nickel(I)–phenyl complexes that insert CO₂ at room temperature
- Deep learning reveals migrating foreshocks during laboratory fault nucleation
- AI-guided CORONA imagery search finds four archaeologically verified tells in Abu Ghraib
- EchoNext: detecting structural heart disease from ECGs with AI
- LHCb observes CP violation for the first time in baryon decays
- Machine learning guides magnetocaloric Laves-phase synthesis
- Deep learning mines venom proteomes for antimicrobial peptides active in vitro and in a mouse infection model
- Progressive learning screened single-atom oxide catalysts for acidic oxygen evolution
- Prospective silent-trial validation of a fine-tuned pathology foundation model for EGFR mutation screening in lung adenocarcinoma
- PhaseNet catalog reveals a 2023 injection-linked earthquake cluster in the Raton Basin
- Machine-learning-selected α-MoO3 and ReS2 deliver high measured responsivity in extreme-ultraviolet detectors
- VenusMine identifies PET hydrolase KbPETase
- Offline reinforcement learning identifies an AgBiI4 film-processing condition
- Centaur: a foundation model that predicts human behaviour across psychology experiments
- Machine-learning screening selects compounds active against Candida
- A closed-loop platform engineers enzymes with minimal human input
- Multi-omics machine learning prioritizes a plasma lipid signature for active tuberculosis differentiation
- BroadAMP-GPT generated and experimentally tested broad-spectrum AMP_S13
- AlphaGenome: predicting the effects of DNA variants
- Machine learning identifies disordered twist-angle sequences with strongly reduced thermal conductivity in multilayer graphene
- Active learning identifies 28 new DFT-validated pressure-induced phase transitions
- Skala: a deep-learning exchange-correlation functional
- Instantaneous voice-synthesis neuroprosthesis for a man with ALS
- FerroAI guides discovery of a ferroelectric composition with high dielectric response
- Generative machine learning nominates a chiral ligand validated in asymmetric C–H activation
- Generative machine learning guides experimental synthesis of LiZn₂Pt and NiPt₂Ga Heusler phases
- Topological learning and simulations predict lithium superionic-conductor candidates
- Dead Sea Scrolls redated with AI and radiocarbon
- Rentosertib: AI-designed drug reports Phase IIa results in pulmonary fibrosis
- Machine-learning-guided Co–Mn diatomic oxygen-reduction catalyst
- Machine-learning tremor catalog reveals Japan Trench spatial and slow-slip relationships
- Proteomic profiling and machine learning identified an externally validated diagnostic panel and severity markers for infective endocarditis
- Autonomous BTSbot-nearby follow-up captures an infant supernova’s flash spectrum and constrains progenitor mass loss
- Active learning discovers eight experimental binders for the LRRK2 WDR domain
- Machine-learning proteomics prioritizes four plasma markers associated with ankylosing-spondylitis activity
- Symbolic-regression active learning identifies 12 acid-stable oxide candidates by HSE06 calculations
- Samudra: an AI emulator of the global ocean for climate simulations
- Robin: ripasudil as a candidate for dry macular degeneration
- Machine-learning proteomics identifies a replicated plasma signature associated with Alzheimer’s disease
- Microsoft Discovery: a PFAS-free refrigerant in about 200 hours
- AlphaEvolve: an agent for designing algorithms
- Deep learning identifies MYT1L as a regulator that restricts glioblastoma plasticity
- Contrastive-learning biomarker discovered from phase 2 data retrospectively generalizes to phase 3 immunotherapy outcomes
- Machine learning reconstructs deep-time crustal thickness and links thickening episodes with porphyry deposits
- AI-designed enhancers control gene expression in healthy mammalian cells
- Deep learning maps previously unknown disease signals in ECGs across two health systems
- Experimental feedback guides ESM-2 to brighter CreiLOV variants
- Neural-network-ranked survey found candidate lenses and spectroscopically followed seven
- A dense nodal array and machine learning reveal a blind fault zone beneath the San Fernando Valley
- Machine-learning-guided screening validates 14 previously unreported drug–GPCR interactions
- LIMITS-guided peptide design yields selective anti-tuberculosis host-defense peptides
- Regional deep-learning phase picking reveals 24 additional earthquakes in a Gorda Ridge swarm
- Machine learning nominates synergistic drug combinations in pancreatic cancer cells
- CNN-guided spectroscopy confirms 28 new extremely metal-poor galaxies
- AlphaFold structure analysis helps reveal PHGDH transcriptional activity linked to Alzheimer’s amyloid pathology
- Serine hydrolases designed from scratch
- MOFGen: agentic AI generates MOF linkers and five AI-designed MOFs are synthesized
- Active learning identifies salt-rich media for flaviolin production
- Machine-learning-guided synthesis identifies a TbIr3 structural phase
- Aardvark Weather: end-to-end forecasting without a physical model
- Urania designs new gravitational-wave detectors
- Seven organic-solar-cell molecules designed with generative models
- MICrONS: a map of one cubic millimeter of mouse visual cortex
- Deep transfer learning detects methane super-emitters in oil and gas fields from Sentinel-2
- AMIE: conversational diagnostic AI rated above primary care physicians in simulated consultations
- OxiGraphX predicted vacancy formation energies in high-entropy perovskites
- ESM-Ezy mining identifies multicopper oxidases and asparaginases with experimentally improved catalytic properties
- Deep-learning-guided mining of NRPS scaffolds yields active naturally inspired antimicrobial peptides
- Leveraging Data Mining, Active Learning, and Domain Adaptation for Efficient Discovery of Advanced Oxygen Evolution Electrocatalysts
- AI-assisted macaque claustrum atlas links cell types with projection zones
- Deep-learning seismic catalog reveals a potentially reactivated blind fault in Taiwan’s 2025 Dapu earthquake sequence
- Deep learning maps Angkorian-period reservoirs beyond Greater Angkor
- Machine-learning models guide design of two vitrimers
- CataPro-assisted enzyme mining identifies a high-activity Sphingobium enzyme for vanillin production
- Euclid: 500 candidate strong gravitational lenses
- Machine-learning search of archived spectra reveals vinyl–NHC coupling products in palladium catalysis
- The AI Scientist-v2: a fully generated paper passes a workshop review
- TeleProt-guided engineering yields a nuclease variant with improved neutral-pH biofilm degradation
- Machine learning and peptide-array experiments reveal new substrates for protein-modifying enzymes
- AI-guided promoter editing raises maize ZmVTE4 expression and kernel alpha-tocopherol
- WenHai: a deep neural network forecasts the eddying ocean
- AI structural modeling helps map nearly 300 metabolite-binding interactions in E. coli
- Canopy traits across tropical forests mapped with deep learning and satellite data
- LogiRx predicts and tests an escitalopram pathway in cardiac hypertrophy
- Machine learning identifies a Ce-doped green-emitting phosphor
- ECMWF puts AI forecasts (AIFS) into operation
- Machine learning screens fluoroether electrolytes for lithium-metal batteries
- CorrDiff: kilometer-scale weather forecasts with generative AI
- Mal-ID maps disease-associated immune-receptor patterns across infections and autoimmune conditions
- Prospective Swedish cohort validates multimodal AI prostate-biopsy biomarker for recurrence and adverse pathology after prostatectomy
- Deep clustering surfaces tremor-like episodes before the 2023 Türkiye earthquake
- BioEmu-1: ensembles of protein structures with deep learning
- FragFold identifies inhibitory fragments within bacterial proteins
- AI co-scientist: a multi-agent system for generating hypotheses
- Evo 2: the largest AI model for biology
- Machine-learning-assisted taxonomy identifies Limnognathia desmeti
- Polybot: an autonomous lab finds processing recipes for conductive polymer films
- Machine-learning network analysis identifies a conserved cell-mechanics gene module
- CNN selection leads to spectroscopic confirmation of a Euclid strong lens
- Adaptive learning selected an additive for a magnesium-air battery
- OptPDE identifies new candidate integrable PDE families with nontrivial conserved quantities
- A radio survey reveals megaparsec-scale diffuse emission around a cosmic-web node
- Machine learning screening identifies an efficient molecular photocatalytic CO2-reduction system
- ProtGPS reveals sequence signals that direct proteins to subcellular compartments
- AI-built Baihetan reservoir catalog reveals fault-controlled and depth-dependent seismicity
- Latent diffusion designs antimicrobial peptides active against fungi and drug-resistant bacteria
- Transformer-assisted survey identifies 87 previously unreported low-surface-brightness galaxies in Abell 194
- CHANCE uses tumor genomes to prioritize non-cancer drugs with activity in patient-derived cancer cells
- Random forest prioritizes investigational compounds with antiplasmodial activity
- Deep learning helps prioritize potent MAGL inhibitor candidates
- Two molecules against KRAS from a hybrid AI and quantum process
- EVOLVEpro improves proteins up to 100-fold with few experiments
- Machine learning-guided cell-free engineering creates specialized amide synthetases for pharmaceutical molecules
- MaSIF designs binders for protein–drug complexes
- Org-Mol prioritizes two experimental immersion-coolant candidates
- EvoGradient evolves antimicrobial peptides active against resistant pathogens
- MatterGen: materials design with a generative model
- Computer-designed proteins neutralize snake toxins
- Active learning identifies a protoglobin variant with 99% yield and cis-selective cyclopropanation
- Graph neural network enables first observation of Λc⁺ → ne⁺νe decay
- Active learning guides experimental design of lead-free solder alloys
- GET and interaction assays link a leukemia-risk PAX5 variant to stronger NR2C2 binding
- Machine-learning selection and Gaia astrometry confirm a 62-Jupiter-mass brown dwarf orbiting G 15-6
- PRAIM: AI finds more breast cancers in German screening
- A protein-language-model workflow discovers high-activity kynureninase enzymes for tumor models
- GlaViTU estimates decadal glacier-area decline from satellite imagery
- Deep learning enables multi-site bone marrow adiposity genetics in UK Biobank
2024 (217) · page for 2024
- OmniFold enables an unbinned 24-dimensional Z+jets cross-section measurement by ATLAS
- Machine-learning screening identifies two drug-repurposing candidates against tuberculosis
- Protein language model guides creatinase stability optimization
- Machine-learning-guided synthesis confirms superconductivity in Mo20Re6Si4
- Deep learning identifies externally replicated fast and slow cognitive-progressor groups in early Alzheimer dementia
- Active learning identified a KNN ceramic composition with high piezoelectric response
- Deep-learning CT response score stratifies outcomes in lung-cancer immunotherapy cohorts
- Bayesian optimization finds high-performance hole-transport molecules for perovskite solar cells
- PICNIC machine learning predictions revealed previously poorly characterized human proteins in cellular condensates
- XLuminA computationally discovers a super-resolution microscopy blueprint combining two optical principles
- PlantRNA-FM identifies translation-associated RNA motifs and structure motifs in plants
- Machine-learning classification dates a tectonic transition and probable slab detachment in central Tibet
- Swin-Transformer discovery of planetary nebulae with spectroscopic confirmation in VPHAS+
- GenCast: ensemble weather forecasts with a diffusion model
- Spatial glycan imaging and machine learning identify a serum N-glycan panel for HCC in cirrhosis
- Representation learning uncovers the extragalactic X-ray transient XRT 200515 in Chandra archives
- ApexGO optimizes antimicrobial peptide activity
- Neural network guides a Ru–Ti–Mn oxygen-evolution electrode
- Machine-learning protein signatures predicted Crohn disease years before diagnosis
- Machine learning screened metal oxides for acetylene semihydrogenation
- GPT-4 agents guide chemical synthesis development and purification
- ClickGen designs PARP1 inhibitors with measured nanomolar activity
- Machine learning selects silver-electrode CO₂ electrolysis conditions
- AlphaQubit: a neural decoder for quantum error correction
- SCimilarity links fibrosis-associated macrophage states across diseases and validates a matching in-vitro cell state
- Deep learning resolves subtle seismic-wave changes across the 2016 Norcia earthquake cycle
- OCx24: experimental catalysts for evaluating prediction models
- CHOOSER uses protein foundation models to discover functional Casλ homologs
- Generative agents built from interview transcripts simulate 1,000 people
- Evo: a foundation model for DNA, from molecules to genomes
- Prospective deep-learning screen identifies nanomolar IRAK1 inhibitor scaffolds
- Machine learning helps identify experimentally synthesized cobalt superalloys
- Machine-learning-guided optimization finds low-loading conditions for platinum-catalyzed amide reduction
- Seven-day flood forecasts with reliability similar to earlier five-day forecasts
- GRACE designs two carbonic-anhydrase-like enzymes with measured catalytic activity
- Mobile autonomous robots for exploratory synthetic chemistry
- Similarity-informed machine learning guides experimental high-entropy alloy catalyst discovery
- XRD-guided learning identifies Li₆AsSe₅I as a room-temperature superionic conductor
- HiDDEN finds malignant-like cell signals in early multiple-myeloma precursor samples
- PatternBoost: constructions in mathematics with a little help from AI
- Transcriptome-based machine learning reveals an MCL1 dependency in alveolar soft part sarcoma
- AI reconstruction extends the 1911 French heatwave signal into September
- Machine-guided synthetic regulatory DNA drives cell-type-specific expression across cell lines and vertebrates
- SMDET-selected ultracool-dwarf candidates include two objects with spectroscopically confirmed T-dwarf types
- CRISPR perturbation and Bayesian network inference map immune-disease regulatory cascades in primary CD4+ T cells
- A gradient-boosted model detected untreated infection before antimicrobial administration
- Self-supervised learning identifies two previously unreported seismic sequences during the Mayotte submarine eruption
- Robotic learning identifies nine polyoxometalate-based frameworks
- Machine learning-guided amidase engineering improves enantioselectivity
- Machine learning and Mendelian randomization prioritize CA2 as a diabetic nephropathy biomarker
- Iterative Gaussian process screening discovers selective aurein-derived antifungal peptides
- Noise-aware active learning selects a high-latent-heat Ti–Ni–Hf shape-memory alloy
- FlyWire: the complete map of neurons in the Drosophila brain
- Minerva identifies nickel-catalyzed Suzuki reaction conditions
- Unsupervised learning reveals mixed-frequency seismic precursors before Axial Seamount’s 2015 eruption
- AI-assisted cerebellar atlases reveal primate-specific cellular organization
- Machine-learning-predicted approved drugs show in-vitro Leishmania activity
- TxGNN: repurposing existing drugs for rare diseases
- Nazca: AI helps discover 303 new figurative geoglyphs
- Solvent mixtures for flow batteries found through an AI–experiment cycle
- Machine learning simulations reveal nonequilibrium short-range order in metallic alloys
- GPU phase folding and deep learning identify five statistically validated ultra-short-period sub-Earth planets in Kepler data
- COSTAR predicts drug activity tested in glioblastoma samples
- Deep learning identifies previously unreported protein interactions in human bacterial pathogens
- Descriptor-guided Co–Co and Co–Ir oxygen electrocatalysts
- Active learning found unconventional self-assembling pentapeptides
- Conversations with an AI chatbot durably reduce conspiracy beliefs
- Deep learning expands the Utah FORGE DAS microearthquake catalog and resolves stimulation structure
- Chemical language models produce experimentally confirmed dual ligands for three target pairs
- MILTON-augmented disease cohorts surface putative new gene–disease associations in UK Biobank data
- Machine learning identifies alkaline laccases from Lepista nuda
- Human-guided active learning improves nickel-cobalt methanation tests
- Machine learning identifies a high-risk diabetic cardiomyopathy phenotype linked to future heart failure
- AlphaProteo: AI-designed protein binders
- AI-accelerated virtual screening identifies small-molecule binders to KLHDC2 and Nav1.7
- Deep learning maps fine-scale plant distributions and fire-associated change
- CHIEF: detecting cancer and predicting survival from slides
- Generative machine learning identifies and validates new dual- and tri-phase eutectic compositionally complex alloys
- Co-training identifies a high-efficiency Cr-activated NIR phosphor
- Neural operators reveal nonreciprocal interactions behind microfluidic nonlinear waves
- deepAMP designs broad-spectrum antimicrobial peptides active against resistant bacteria
- Learned call analysis identifies receiver-specific marmoset vocal patterns
- Machine-learning Treg framework identifies SMARCC1 as a previously unrecognized regulator
- Closed-loop learning links donor–acceptor molecules’ photostability to higher triplet states
- Machine learning identifies regional low-magnitude seismic unrest before the Ridgecrest and Anchorage earthquakes
- Machine learning detects volcanic fingerprints and suggests a northern extratropical signal for the unidentified 1809 eruption
- MitoReID identifies epicatechin as a cyclooxygenase-2 inhibitor
- Machine-learning screening identified three compounds with measured ULK1 inhibition
- A speech neuroprosthesis reaches 97.5% accuracy after calibration
- Machine learning identifies predicted zero- and extreme-thermal-expansion monolayers
- Generative antibody language model yields a pseudovirus-neutralizing XBB spike binder
- PICS-assisted survey reported seven new Large Magellanic Cloud pulsars
- Machine-learning-guided proteomics identifies APOC3, RAB39A and KNG1 for plaque-rupture classification in acute coronary syndrome
- Unsupervised learning guides experimental superalloy design
- ML-guided experiments synthesize single-phase Ba–Sr cerate perovskite solid solutions
- MODIFY-guided library screening identifies a cytochrome c variant active in both carbon–boron and carbon–silicon bond formation
- Machine learning uncovers four new extracellular contractile injection-system toxins
- Transfer learning guides experimental discovery of oxygen-evolution perovskites
- AlphaProof and AlphaGeometry 2: silver-medal level at the Math Olympiad
- AI-guided HuJ3 mutation improves HIV-1 gp120 binding and neutralization in assays
- Structure-informed model selects antibody variants active against SARS-CoV-2 escape variants
- Machine-learning-assisted screening yields an experimentally active FtsZ inhibitor candidate
- NeuralGCM: a hybrid physics-and-learning model for weather and climate
- Virchow: pan-cancer detection using a pathology foundation model
- Machine learning identifies plasma protein signatures that predict onset of common and rare diseases
- AI-guided polyVERSE discovery yields a high-temperature dielectric polymer with 8.3 J cm⁻³ energy density
- AI helps resolve an unknown regulatory mechanism in bacterial acetyl-CoA synthetase
- AI-classified XPCS links local relaxation episodes with bulk stress changes in a colloidal glass
- Multi-omics machine learning identifies lysophosphatidic acid in PRRSV infection and links it to immune suppression
- Machine learning identifies sediment-derived granites in the Hadean zircon record, supporting early subduction
- Machine-Learning-Guided Peptide Drug Discovery: Development of GLP-1 Receptor Agonists with Improved Drug Properties
- Multimodal AI distinguishes ten dementia etiologies across nine clinical datasets
- FRASE-bot screening identifies a small-molecule ligand that selectively kills CIB1-dependent cancer cells
- Machine-learning selection and external mass-spectrometry testing identify an ARDS 28-day prognosis protein panel
- Bayesian optimization improves printed BiSbTe thermoelectrics
- Simulation-trained inference maps accreted stellar mass in nearby galaxies
- Machine learning helps identify 61 new cataclysmic variables in SDSS-V spectroscopy
- ESM3 and esmGFP: a fluorescent protein designed by a language model
- Machine-learning design yields a high-strength printable aluminum alloy
- Deep-learning design produced soluble analogues of membrane-protein folds with preserved binding functions
- A machine-learning plasma panel classifies some REM sleep behavior disorder participants years before motor Parkinson diagnosis
- Machine learning prioritizes 1-methylnicotinamide as a serum signal across HBV cirrhosis-to-HCC progression
- Automated methylome analysis identified candidate blood epigenetic markers for first-episode schizophrenia
- PI-CAI: AI matches or beats radiologists in prostate cancer detection on MRI
- Bayesian search identifies organic metallophotocatalyst formulations
- African elephants address one another with calls resembling names
- PhaseNet-TF expands Tonga earthquake catalog and resolves the intermediate-depth double seismic zone
- AMPSphere: antimicrobial peptides from the global microbiome
- Machine-learning-guided lysine dioxygenase variant improves turnover
- Aurora: a foundation model for the atmosphere
- AlphaFold2-guided mining identifies context-independent cytidine deaminases for efficient base editing
- Deep learning on routine pathology slides prescreens FGFR-altered bladder cancers in a global study
- LightGBM links a three-protein blood panel to biopsy-staged MASLD fibrosis
- Machine-learning analysis linked two sepsis-associated neutrophil clusters to five mouse-validated marker genes
- ML-guided mechanosynthesis yields a lead-free photocatalyst
- Active learning identifies a higher-activity streptavidin variant for gold-catalyzed hydroamination
- A self-driving lab finds a polymer shell with record energy-absorbing efficiency
- PICS-assisted survey reported seven new Small Magellanic Cloud pulsars
- Machine learning guided multi-metric optimization of photocatalytic CO2 reduction
- Machine-learning-designed peptides show dual GCGR and GLP-1R activity
- Vision transformer detects methane point-source plumes in Sentinel-2 imagery at global scale
- Machine learning links mycobacterial protein disorder to targeted degradation potential
- Ensemble machine learning identified CDK8 inhibitor hits
- AlphaFold 3: structure and interactions of life’s molecules
- ChemCrow: a language-model agent with chemistry tools
- Sperm whales: a possible ‘phonetic alphabet’ in their calls
- Machine-learning metabolomics prioritizes serum metabolites for SLE diagnosis and activity assessment
- HydrogelFinder: A Foundation Model for Efficient Self-Assembling Peptide Discovery Guided by Non-Peptidal Small Molecules
- LAMBDAZERO generates an experimentally active soluble epoxide hydrolase inhibitor scaffold
- GPT-3.5 reasoning helps connect MRE11 to social-context sleep regulation in Drosophila
- Closed-loop, distributed discovery of organic laser emitters
- Reinforcement learning and hit expansion identify SARS-CoV-2 Mpro inhibitor series
- XGBoost-guided splice switching identifies NEDD4L exon 13 as a TNBC cell vulnerability
- Protein-surface-designed molecules activate PPARs
- OpenCRISPR-1: a genome editor designed by language models
- Machine-learning multi-omics links gut metabolites to GPCR targets and tau changes in Alzheimer’s patient-derived neurons
- Iterative machine learning identifies potent inhibitors of alpha-synuclein secondary nucleation
- Graph machine learning identifies polymer gas-separation membranes
- Structure-based virtual screening identifies mammalian enzymes catalyzing the second step of carnitine biosynthesis
- Machine learning reveals tropical aerosol-driven cloud expansion despite stronger rainfall
- Antibodies designed from scratch with RFdiffusion
- AI identifies semapimod against dormant bacteria
- AI-assisted seismic catalog links the 2018 Baige landslide-dammed lakes to increased local earthquakes
- Machine learning guides new six-colorings of the Euclidean plane
- Progeni predicts candidate melanoma and colorectal-cancer targets validated in cell models
- DeepEvo yields heat-tolerant G3PDH variants
- Tree-based machine learning nominates Parabacteroides merdae and N-acetyl-D-mannosamine as ulcerative-colitis alleviators in a mouse model
- Spectroscopy validates machine-selected populations of obscured active galaxies
- LAISS-assisted review reclassifies anomalous ZTF transients
- SEEDS: weather ensembles generated with a diffusion model
- Deep-learning-enhanced Sparta earthquake catalog resolves a shallow multifault rupture
- SyntheMol: generated antibiotic candidates against Acinetobacter baumannii
- Machine learning links circulating tyrosine and distinct lipid-metabolite features to diabetic kidney disease and retinopathy detection
- Unsupervised clustering identifies three blood immune endophenotypes associated with distinct multiple-sclerosis trajectories
- Deep-learning-assisted DIP-MS identifies POLR2E as a constitutive prefoldin-like-complex subunit
- Three spectroscopically confirmed z≈6 quasars found through contrastive learning
- Feature-fusion model prioritizes antibacterial Ugi products
- Machine learning screening reveals ABBV-744 as a modulator of P2Y purinergic receptors
- Machine learning selected nucleoside derivatives for hydrogel testing
- Reliable flood forecasts even where gauges are missing
- A biophysics-informed protein language model designs functional GFP variants
- AI-assisted Hubble asteroid sample extends the shallow size-distribution slope to H≈20
- AI-ranked SARS-CoV-2 protein interactions guide discovery of an NSP10–NSP16 inhibitor
- MolE prioritizes existing drugs with measured Staphylococcus aureus growth effects
- AI- and DFT-guided synthesis identifies a previously unreported calcium-ruthenate phase
- Machine learning guides engineering of an alkaloid-biosynthesis enzyme
- An ML-generated MinE homolog restores bacterial cell division patterning in vivo
- CoSaNN designs and tests an induced-fit carbohydrate kinase
- Machine learning and genetic search design antimicrobial peptides
- Gaussian-process-assisted engineering of a ketoreductase for an ipatasertib precursor
- ModelAngelo: automated atomic model building in cryo-EM maps
- Deep learning designs selective antimicrobial peptides against Cutibacterium acnes
- Machine learning identifies a projected shift from DIC- to alkalinity-dominated ocean CO₂-flux variability
- Machine learning identifies plasma metabolite panels for gastric-cancer diagnosis and prognosis
- Preventing tearing-mode instabilities in a tokamak with AI
- EHR machine learning and genetic colocalization link osteoporosis to female Alzheimer’s risk near MS4A6A
- Selective-bias screening nominates LabMol-394 antimalarial candidates
- SLIDE uncovers a keratinocyte-centered signal in systemic sclerosis
- SLIDE links islet T-cell clone expansion to immune gene programs
- Interpretable machine learning finds a perpendicular-spin signature consistent with a spinon Fermi surface
- CAMIL regression predicts continuous homologous-recombination deficiency and tumor immune biomarkers from histology
- Sequential model-based optimization discovers two previously unreported chemical reactions
- Machine-learning X-ray analysis provides bulk evidence for a Bragg-glass phase in PdₓErTe₃
- Vesuvius Challenge: the first Herculaneum scroll can be read
- Mechanistic machine learning links PI3K to Src-inhibitor effects in cardiac fibroblasts
- Machine-learning models identify carbon-source-dependent metabolic reactions that control Escherichia coli growth
- Machine learning prioritizes ruthenium complexes with increased MRSA antibacterial hit rates
- COMPOSER deep learning sepsis model linked to 17% lower in-hospital sepsis mortality
- Deep learning reveals a new seismicity distribution along the Kabaw Fault in central Myanmar
- AlphaGeometry: Olympiad geometry without human demonstrations
- AlphaFold-Multimer-guided experiments corroborate six novel human protein interfaces
- Active learning selected oxidative methane coupling catalysts
- From 32 million candidates to a new battery electrolyte
- Satellite map of industrial activity at sea
- Machine learning selects CSF and serum protein panels associated with early Alzheimer disease
- Active learning identifies experimentally improved ammonia catalysts
- Machine learning-guided synthesis of two-dimensional AgBi iodide perovskites
2023 (122) · page for 2023
- Machine learning screens cinnamic-acid derivatives for Leishmania enzyme inhibition
- Machine-learning-guided closed-loop synthesis experimentally realizes unreported dye molecules
- Machine learning discovers D-amino-acid mixtures that boost anti-biofilm antibiotic cocktails
- A new class of antibiotic candidates against MRSA
- Coscientist: autonomous chemical research with language models
- Designed proteins bind peptides with the highest affinity reported
- A machine-learning-selected P4 serum panel detects hepatocellular carcinoma and signals cirrhosis-to-cancer progression
- Machine-learning denoising exposes strong size scaling in deep-earthquake rupture parameters
- FunSearch: new mathematical results with a language model
- Bayesian optimization identifies active non-noble oxygen-evolution oxides
- MatterGen: stable materials generated from desired properties
- AI-assisted hippocampal atlas maps Alzheimer-associated cellular changes around amyloid plaques
- 3D-HiChAT predicts embryonic enhancer hubs validated by CRISPR interference
- A recurrent neural network uncovers a non-canonical human GRHL1 DNA-binding site
- Machine learning identifies bacterial strains and combinations that reduce pathogen colonization in Arabidopsis
- GNoME: 2.2 million new predicted crystals
- A-Lab: an autonomous laboratory for synthesizing inorganic materials
- Deep learning detects slow slip events in Cascadia GNSS data
- PANDA: pancreatic lesion detection on non-contrast CT with deep learning
- MAIP enriches high-throughput screening for antimalarial hits
- Chroma: a generative model for protein design
- GraphCast: global 10-day weather forecasts
- Multi-omics models guide experimental validation of flowering-time genes
- Machine-learning screening reveals enhanced formate selectivity of Cu–Ga for CO₂ electroreduction
- Deep-learning-guided cell-free screening identifies antimicrobial peptides
- AlphaZero finds improved finite extremal-graph bounds
- Active learning identifies a high-performing four-metal perovskite oxygen-evolution catalyst
- MetNet-3: short-range neural forecasts in Google products
- Machine learning exposed a structure in terminal-singularity classification
- Machine-learning models identify cerebrospinal-fluid protein signatures for multiple-sclerosis activity and disability progression
- CAIMAN validates AI prescreening of typical and atypical large-bowel biopsies across UK and Portuguese cohorts
- Machine-learning image search helps identify Pavo, an isolated star-forming dwarf galaxy
- Integrated radiogenomic machine learning predicts neoadjuvant chemotherapy response in metastatic ovarian cancer
- Physics-guided machine learning links Southern Ocean regime shifts to a northward-shifting Antarctic Circumpolar Current
- Neural active learning guides stronger bone-scaffold architectures
- Meta AI: near-real-time image decoding from brain activity
- Machine-learning-guided screening discovers potent innate immune signaling modulators
- Vesuvius Challenge: first letters read from an unopened Herculaneum scroll
- Bayesian optimization identifies a methane-combustion catalyst formula
- EVEscape: predicting viral variants before they emerge
- BTSbot workflow automatically confirms and classifies SN 2023tyk
- CNN-assisted search confirms quadruply lensed quasar PS J2107−1611
- Extrapolative machine learning identifies a multielement RWGS catalyst
- AlphaMissense: cataloging missense genetic mutations
- RETFound: a foundation model for disease detection from retinal images
- AI-assisted protein-family analysis identifies a tested toxin-antitoxin superfamily
- PhotoCat guides experimental discovery of photocatalytic reactions
- Machine learning helps recover dimension from quantum periods for a class of Fano varieties
- AI-assisted seismic mapping reveals segmented fault activity at the Discovery transform fault
- Deep learning detects an emerging anthropogenic warming fingerprint in daily precipitation
- A neuroprosthesis translates attempted speech at 62 words per minute
- Woman with paralysis communicates through a brain-controlled avatar
- Machine learning assists a proved dimension bound for affine Deligne–Lusztig varieties
- Deep-learning knee imaging phenotypes reveal osteoarthritis genetic and fracture associations
- Active-learning search experimentally realizes a high-zT Cu₁₋ₓAgₓGaTe₂ thermoelectric
- Neanderthal and Denisovan peptides reconstructed as antibiotic candidates
- Avoiding aircraft contrails with AI forecasts
- Machine learning guides engineering of fluorescent calcium indicators
- MASAI: AI-assisted mammography screening in a randomized trial
- Physics-informed machine learning guides experimental Curie-temperature validation in Ce–Zr–Fe alloys
- CMS makes the first hadron-collider observation of boosted Z→cc̄ decay
- Machine learning guides proofs of braid and flat-braid invariants
- Machine-learning analysis of plasma metabolomic fingerprints yields a 21-metabolite gastric-cancer panel validated across multicentre cohorts
- RFdiffusion: designing new proteins with a diffusion model
- AI-assisted macaque cortex atlas reveals regional cell-type organization
- Pangu-Weather: medium-range forecasts outperform the operational ECMWF model
- NowcastNet: extreme-precipitation forecasts up to three hours ahead
- Deep learning detects elevated tissue-derived cfDNA across cancer and cirrhosis cohorts
- Machine learning guides Fe–Mn–Ni catalysts for NOx reduction
- Machine-learning-guided experiments identify a high-power-factor PbSe:SnSb thermoelectric family
- AlphaFold-assisted structural clustering discovers new deaminase functions and compact base-editing enzymes
- CogSim helps predict fusion ignition at the NIF
- Sequence-conditioned CogMol designs SARS-CoV-2 spike and main-protease inhibitors
- Organ-specific learned networks prioritize tested seed-lipid regulators
- SIMMER-guided experiments identify gut bacterial strains that deplete methotrexate
- ATLAS graph neural network enables first observation of four-top-quark production
- AlphaDev: reinforcement learning discovers faster sorting routines
- Machine learning guides direct synthesis of higher-silica faujasite
- Abaucin: a targeted antibiotic against Acinetobacter baumannii
- Active learning maps wind-tunnel terrain equivalence
- ExoMiner and multiplicity validate 69 new exoplanets
- Nonlocal quantum interference without entanglement of intrinsic properties
- First draft of the human pangenome reference
- EmbryoNet links statins to FGF-like zebrafish developmental effects
- BacterAI experimentally maps amino-acid growth requirements of two oral Streptococcus species
- New zirconium-indium-nickel superconductor identified with machine learning
- LinearDesign: more stable mRNA sequences designed by an algorithm
- An MRI decoder turns brain activity into text
- Machine-learning classification shows rigidity-corrected moment release is similar across earthquake depths
- Single-cell imaging and machine learning reveal a Cdk2–COP9 interaction that induces a senescence-like immune state in Drosophila cells
- Longitudinal EHR machine learning identifies patients who may need autoimmune rheumatic-disease evaluation years earlier
- Learned promoter models identify human-specific transcription motifs
- Language models guide improvements to seven antibodies
- Machine-learning screening yields novel potent acetylcholinesterase and butyrylcholinesterase inhibitors
- Learned community models prioritize a tested bacterial colonization inhibitor
- Machine learning finds ribbon-knot certificates and rules out candidate pairs
- PRIMO: a sharper image of the M87 black hole
- ExoGRU-guided experiments support RBM24 involvement in small-RNA secretion
- Machine-learning analysis identifies kinematic evidence for a protoplanet in the HD 142666 disk
- CLEAN experimentally identifies SAM hydrolase activity in previously uncharacterized proteins
- Machine-learning-guided calculations predict four rare-earth-free magnetic Fe–Co–Si compounds
- Acoustic learning maps spider-monkey associations with forest disturbance
- Reinforcement-learning-guided CdS shell-growth route using AlphaFlow
- Adaptive learning finds higher-activity Fe–N–C oxygen-reduction catalysts
- LuxSit: de novo luciferases active in the laboratory
- Learned morphology guides targeted isolation of gut bacteria
- Virtual screening identifies antiplasmodial beta-haematin inhibitor chemotypes
- AI-generated catalyst design led to experimentally tested polymerization catalysts
- CellOracle-guided zebrafish perturbations identify axial-mesoderm effects of noto and lhx1a loss
- Unsupervised machine learning reveals previously missed hydraulic-fracturing seismicity in Live Oak County
- Real-time wildfire perimeters from satellite imagery
- Machine learning guided XylM variant improved biosensor-indicated productivity
- Deep learning searches for technosignatures around 820 nearby stars
- Elastic-net proteomic risk modeling and multi-cohort validation identify proteins associated with incident diabetes
- Support-vector regression selects more selective tetrahydropyran catalysts
- CNN-augmented focal mechanisms reveal stress-linked faulting changes around major southern California earthquakes
- Random-forest metabolite profiling reduces false-positive newborn screens for four inherited disorders
- Sybil: lung-cancer risk prediction up to six years from a low-dose CT
- A liver-cancer target and molecule identified with Pharma.AI
- Autonomous X-ray mapping reveals previously unseen block-copolymer morphologies, including a nanoscale ladder
- Machine-learning pipeline identifies antimicrobial hexapeptides
- Machine learning guides a proposed UCd3 intermetallic phase assignment
2022 (91) · page for 2022
- Integrated proteomics and metabolomics with LASSO nominate a four-analyte early IgA-nephropathy plasma panel
- Random Forest selection guided discovery of nine TRAPUM millisecond pulsars
- Machine-learning-guided electrochemical screening isolates five new reaction products
- Unsupervised seismic clustering reveals distinct drainage signals at Gornergletscher
- CNN tracking reveals irradiation-driven nanovoid size fluctuations and shrinkage in copper
- Machine-learning-guided PBMC proteomics identifies SLE diagnosis and activity biomarker panels
- Machine learning selects Eu²⁺-activated green and blue-green phosphors
- Machine-learning-guided search discovers and synthesizes the rare-earth-free magnet Fe₃CoB₂
- Generative deep learning discovers the RIPK1 inhibitor RI-962 with an unreported scaffold
- Machine-learning screening guides ionic-liquid polymer electrolytes for lithium-metal batteries
- Autonomous microscopy discovers an edge plasmon in layered MnPS₃
- Machine-learning-assisted neutron scattering shows hydrostatic pressure tunes interactions in spin ice
- ESM Metagenomic Atlas: over 600 million protein structures
- Semi-supervised machine learning identifies two strong gravitational lenses in the Deep Lens Survey
- Deep-learning control links single-cell expression to growth responses
- Ligand-based discovery of coronavirus main protease inhibitors using MACAW molecular embeddings
- Prospective machine-learning screen identifies six previously unreported OATP1B1 inhibitors
- Mass spectrometry and machine learning identify bioactive peptide NHPD-50
- FLEET-guided observing campaign classifies superluminous supernovae
- Generative learning selects experimentally active EGFR inhibitor analogs
- Machine-learning screening finds previously unreported antibacterial compounds active against Burkholderia and ESKAPE pathogens
- LIGHTHOUSE identifies riboflavin 5′-monophosphate as a PPAT inhibitor
- Bayesian language-model design improved measured scFv binding
- Machine learning and DFT discover high-entropy Invar alloys
- AlphaTensor: new algorithms for matrix multiplication
- Machine learning and synthesis clarify metastable phases of carbon
- BIONIC links the natural product NP329 to yeast cell-wall biogenesis and β-1,6-glucan synthesis
- Deep-learning screening identifies covalent SARS-CoV-2 main-protease inhibitor scaffolds
- Crystal-graph learning and DFT identify 19,512 additional thermodynamically stable materials
- Robot-guided optimization identifies six non-aqueous electrolytes with improved fast-charge performance
- AlphaFold helps interpret activation of PINK1, a Parkinson’s-linked protein
- ProteinMPNN: protein sequences designed in about one second
- MIMaL predicts and experimentally validates five yeast gene functions from protein–metabolite profiles
- Deep learning detects pancreatic cancer on CT in a nationwide study
- Machine learning screens perovskite oxygen-reduction electrodes
- AI detects Parkinson's disease from nocturnal breathing signals
- Deep learning helps experimentally map a ribozyme neutral network
- Machine-learning-guided peptide design yielded non-hemolytic anticancer candidates
- Machine learning generates active daunomycin-binding synthetic nucleic-acid polymers outside selected sequence families
- ATLAS boosted decision trees enable first observation of triple-W boson production
- Deep learning and spectroscopy confirm four faint quasars at redshift about 5
- Machine learning isolates aerosol-driven cloud-cover increase after the Holuhraun eruption
- AlphaFold Database: over 200 million structures
- ProtGPT2: a language model generating novel protein sequences
- Random forest prioritizes dual-stage antimalarial compounds
- TREWS: early sepsis alerts evaluated in five hospitals
- Interpretable machine learning discovers ultrapermeable polymer membranes for gas separation
- Bayesian search and synthesis reveal a high-anisotropy [Fe/Co/Fe/Ni]₁₃ multilayer
- Semi-supervised learning selects a lithium thioborate for vacancy engineering
- Machine learning selected fractionally doped perovskites for synthesis
- Active learning improves a cell-free carbon-fixation cycle
- Machine-learning selection identifies a six-metabolite signature for one-year cardiovascular events after PCI in type 2 diabetes
- BirdNET: citizen science through bird-song identification
- Machine-learning-guided Fe₂P-type alloys reach cryogenic magnetocaloric temperatures without rare earths
- Machine learning and genetic search predict stronger refractory alloys
- Model of the human nuclear pore with AlphaFold2 and electron microscopy
- X-ray machine-learning analysis reveals opposing Cd/Re displacements and Re 5d involvement in Cd₂Re₂O₇ structural order
- EQTransformer-based catalog reveals a 50-km-deep outer-rise fault at the Southern Mariana trench
- ColabFold: free protein folding with AlphaFold2 or RoseTTAFold
- Machine learning suggests a more general theory of microlensing
- Learned cluster expansion guides an acidic oxygen-evolution catalyst
- Machine learning model predicts mental health crises from health records
- Machine-learned design of CuPd nanocubes for electrocatalytic nitrate reduction to ammonia
- Knowledge-graph-guided discovery of antibiotic-resistance associations
- FAST-PETase: an AI-engineered enzyme degrades PET waste
- Swarm learning predicts colorectal cancer mutations from histopathology slides
- Random-forest selection links a food compound to COX-1 inhibition
- Murmurations: unexpected oscillations in elliptic-curve data
- Machine-guided microscopy characterizes an atypical ataxia presentation
- Random-forest alert screening leads to spectroscopic confirmation of four changing-look AGNs
- Learned phase models identify synergistic photoanode mixtures
- Active-learning experiments reveal distinct on-field and off-field polarization-switching behavior in ferroelectrics
- Deep learning assesses heart allograft rejection from biopsy whole-slide images
- LSTM-designed peptides show antibacterial activity and mixed efficacy in a mouse sepsis model
- Machine-learning selection uncovers plant enzymes for alkaloid pathway branches
- Unsupervised clustering links deep-earthquake scaling to hydration differences among northwest Pacific slabs
- Ithaca: restoring and dating ancient Greek inscriptions
- Unsupervised learning guides high-zT ScNiSb-based half-Heusler thermoelectrics
- Learned activity models improve viral diagnostic assay designs
- Machine-learning-guided screening identifies Bcl-2-family inhibitors IS20 and IS21
- FourCastNet: high-resolution global forecasts with neural networks
- Tokamak plasma control with reinforcement learning
- Knowledge-graph learning prioritizes experimentally examined Alzheimer genes
- Learned subgroup rules guide higher-component oxidation catalysts
- HydrAMP generates antimicrobial peptide analogues validated in bacterial assays
- Machine learning identifies new CYP2C9-inhibiting drugs and drug metabolites
- AMPlify prioritizes bullfrog-genome antimicrobial peptides active against priority pathogens
- Machine learning screening selects an energetic material for laboratory characterization
- Cell2location: mapping fine-grained cell types in spatial transcriptomics
- Machine learning guides discovery of hard Fe–Nb–B metallic-glass films
- Machine-learned multiscale simulations reveal lipid-dependent RAS organization at membranes
2021 (86) · page for 2021
- Bayesian optimization guides complex nanoparticle heterostructures
- Machine learning reveals two previously undetected phases in Rydberg-atom quantum-simulator data
- Autonomous robotic system maps Bi2O3 synthesis phase diagram with active learning
- Retrosynthetic software finds experimentally validated supply routes for umifenovir
- DINGO: gravitational-wave parameters in seconds
- DM21: a neural network as a density-functional theory functional
- Learned soundscape features predict tropical-forest species occurrence
- AI-guided pure mathematics: knots and representation theory
- A neural network “hallucinates” new protein structures
- Unsupervised machine learning identifies palladium catalysts that form dinuclear complexes
- ExoMiner: 301 new planets validated in Kepler data
- Machine-learning-guided evolution identifies higher-activity sortase A variants
- Machine-learning search surfaces extreme T-subdwarf candidate CWISE J0523
- Neural-network-guided autonomous experiments reveal an unreported TosMIC reaction
- MutCompute guides thermostable Bst DNA polymerase variants
- Deep learning extends the mapped extent of a North Anatolian Fault slow-slip event
- High-resolution machine-learning catalog outlines a potential unmapped fault in the Weiyuan shale-gas region
- EVE: evolutionary model predicts disease-variant pathogenicity without labels
- Machine learning prioritizes agonists for two human olfactory receptors
- Machine learning guides automated synthesis of 19F MRI copolymer agents
- Real-time discovery and characterization of AT2020xnd, an AT2018cow-like fast transient
- Machine learning guides fatty acyl-ACP reductase engineering
- Enformer: gene expression predicted from DNA sequence
- ECNet-guided selection of higher-order TEM-1 variants
- Short-term rainfall prediction with a generative model
- Cross-species machine learning identifies transcription factors that negatively regulate plant nitrogen-use efficiency
- Unsupervised machine learning guides discovery of a lithium solid electrolyte
- ML-guided PtFeCu nanocatalyst shows enhanced oxygen-reduction activity and durability
- Federated learning model predicts oxygen needs in COVID-19 patients
- Climate TRACE: first independent database of global emissions
- Data-science-guided experiments discover a Co–Ta–Sn rutile alloy oxide with tunable transparency and acid-stable electrocatalysis
- COMPOSER sepsis model flags unfamiliar cases instead of guessing
- Machine learning identifies moth-receptor agonists that attract larvae
- IceNet: seasonal forecasts of Arctic sea ice
- Machine learning and experiments yield tolerant polymer dielectrics
- THESEUS derives resource-efficient designs for high-dimensional photonic GHZ and heralded Bell states
- Hybrid machine learning guides experimental discovery of multicomponent metallic glasses
- Machine learning-assisted directed evolution with in silico combinatorial screening
- Learned models reveal nonmonotonic bacterial patterns in boreal lakes
- Deep learning designs abiotic miniproteins that improve antisense-oligonucleotide nuclear delivery
- INS018_073: AI identifies a preclinical candidate for renal fibrosis
- AlphaFold Protein Structure Database: first 350,000 models
- ProGen generates artificial proteins with enzyme activity
- RoseTTAFold: accessible prediction of protein structures
- A neuroprosthesis restores words for a man with paralysis
- Machine learning reveals a nitrogen-assimilation strategy in Bacteroides
- Boolean-network learning prioritizes an IBD organoid barrier target
- Single-station CNN analysis adds 273 visually confirmed earthquakes to the L’Aquila near-fault catalog
- iTARGEX analysis identifies ELG1 as a regulator of yeast transcription buffering and four protein-turnover regulators
- Machine-intelligence design yields selective synthetic COX-1 inhibitors from a marine natural-product template
- Deep learning reads HIV rapid tests photographed in rural South Africa
- Machine-learning unmixing links acid rain and subsurface flowpaths to watershed CO₂ drawdown
- Generative AI and on-chip synthesis produce new LXR agonists
- Machine learning identifies five synthesized Chevrel-phase tellurides with unusual channel-site occupancy
- Cross-tissue machine learning prioritizes a 13-gene rheumatoid-arthritis activity score
- Satellite-image clustering guides discovery of four late Miocene fossil sites in Mozambique’s Urema Rift
- Explainable neural-network analysis isolates aerosol-sensitive regional climate signals in CESM1
- H01: reconstruction of one cubic millimeter of human cerebral cortex
- COVID Moonshot uses synthesis-directed de novo design for SARS-CoV-2 main protease inhibitors
- Med-BERT pretrained on structured health records improves disease prediction
- AI-generated cell-penetrating peptide selected with molecular-dynamics simulations
- Random-forest selection and spectroscopy identify 20 new high-redshift quasars
- Neural network ensemble for prenatal detection of complex congenital heart disease
- EAGLE: AI electrocardiograms to detect reduced heart function
- Combinatorial constructions and counterexamples with neural networks
- Machine learning guides refinement of DNA aptamers binding NGAL
- Two scribes in the Great Isaiah Scroll: new evidence from handwriting
- Bayesian optimization and graph learning identify two synthesized ultra-incompressible hard materials
- An integrated Arabidopsis regulatory network predicts 13 previously unlinked reactive-oxygen-stress regulators
- Machine learning guided Cu catalyst additive discovery for CO₂ reduction
- Deep learning finds comet C/2020 T2 in a single ZTF image
- Autonomous ML search and experiments find Fe–Co–Ir/Pt alloys above the Slater–Pauling magnetization limit
- QSAR screening identifies antimalarial sesquiterpene lactones
- AI-assisted virtual screening identifies a dual-stage antimalarial candidate
- Earth-abundant Ni3Mo alloy outperforms platinum for ethane dehydrogenation
- IBM designs two new antimicrobial peptides with generative models
- trRosetta releases over 6,300 predicted protein-family structures
- Biological activity-profile modeling identifies SARS-CoV-2 antiviral leads
- Machine learning and ab initio screening predict 60 stable two-dimensional ferroelectric bimetal phosphates
- Mirai: future breast cancer risk from mammograms
- AI-photoswitchable CRISPRi screening identifies new regulators of TFEB localization during starvation
- Sequential X-rays to predict COVID-19 deterioration
- Language models identify viral regions less prone to immune escape
- 1,210 gravitational-lens candidates in the DESI Legacy Surveys
- Bayesian models prioritize compounds for Schistosoma mansoni assays
- Learned resistance models expose tested enzyme-substrate activities
2020 (53) · page for 2020
- Autonomous machine-learning experiments identify a dual-anion aqueous sodium electrolyte with a wider stability window
- Fusing CT images and EHR data with deep learning improves pulmonary embolism detection
- AI-derived symbolic formula improves dark-matter halo overdensity estimates in a cosmological simulation
- AlphaFold wins CASP14 for protein folding
- CAMEO discovers GST467, a phase-change material with double the optical contrast
- A model analyzes coughs to detect COVID-19 infections
- The QUBRICS machine-learning survey spectroscopically identifies 224 new bright high-redshift quasars
- FermiNet: quantum physics of molecules from first principles
- Over 1.8 billion trees mapped in Africa’s drylands
- Chematica: computational synthesis planning for natural products
- Akita: predicting 3D genome folding from DNA sequence
- OPEX-guided experiments reveal bacterial cross-stress responses
- Active deep learning identifies emission-line object candidates in LAMOST spectra
- Machine-learning validation and independent cross-checks classify 50 Kepler candidates as planets
- Machine learning derives new idealized ocean-eddy momentum and heat-flux closures
- Random-forest screening identifies experimentally active anti-CRISPR families
- Machine learning identifies seismic tremor precursors to Whakaari eruptions
- An attention-based neural network generates functional secretion signal peptides
- A mobile robotic chemist
- DEL-trained models identify experimentally active small-molecule hits
- CNN distinguishes pancreatic cancer on contrast CT with cross-racial validation
- AcRanker-guided discovery of AcrIIA20 and AcrIIA21 anti-CRISPR proteins
- AI and clinical data to identify COVID-19 from CT scans
- An AI system predicts progression of macular degeneration
- Cu–Al electrocatalysts for CO2 found with active learning
- Self-driving laboratory optimizes organic hole-transport thin films
- HoB2 exhibits a large magnetocaloric response near 15 K
- Self-reported symptom app predicts likely COVID-19 infection in real time
- Machine-learning-guided synthesis realizes four single-phase refractory carbide compositions
- LCLS magnet tuning with machine learning
- AI augmentation lifts radiologists in telling COVID-19 from other pneumonia
- Adaptive cloud laboratory reveals optical chirality in perovskite nanocrystals
- PENet detects pulmonary embolism on volumetric CT angiography
- Graph networks to understand glass dynamics
- Machine-learning-guided synthesis yields a new blue-emitting perovskite-like single crystal
- MetNet: neural rainfall forecasting up to 8 hours ahead
- EchoNet-Dynamic: video-based AI for left ventricular ejection fraction
- COVNet deep learning separates COVID-19 from other pneumonia on chest CT
- Learned genomic models predict Baltic microbial niche placement
- DeepScribe: reading cuneiform signs on Persepolis tablets
- KCML links olfactory-receptor perturbations to multicellular organization and colorectal-cancer signaling
- Predicted structures of SARS-CoV-2 proteins
- Combining AI algorithms with radiologists improves mammography specificity
- Halicin: deep learning identifies an antibiotic
- Closed-loop optimization of fast-charging battery protocols
- Deep learning screens LabMol compounds for antiplasmodial activity
- Materials informatics and high-throughput synthesis identify durable oxide-ion conductors
- DSP-1181: an AI-created drug candidate enters Phase I
- Deep-learning CADe colonoscopy raises adenoma detection in a double-blind trial
- Interpretable multi-omics machine learning reveals unexpected drug mechanisms in Huntington disease models
- deepDTnet repurposes topotecan as an RORγt inhibitor for multiple-sclerosis models
- Deep neural networks diagnose brain tumours in minutes from stimulated Raman histology
- An AI breast-cancer screening system evaluated in two countries