Topic: Artificial Intelligence

Artificial IntelligenceOctober 20, 2026

Integrating cooperative breeding with reciprocity and culture extends Hamilton's rule to humans.

We study how kinship and reciprocity shape giving and child care support in societies with limited access to markets. Extending Hamilton's rule to incorporate reciprocity, we test it using behavioral data from dictator games and surveys in remote matrilocal and patrilocal villages of northeast India (N = 406 parents of young children). We replicate with data from urban Manila neighborhoods (N = 301) and the Solomon Islands (N = 624). Consistent with predictions, transfers to recipients increase

Artificial IntelligenceOctober 14, 2026

A microexon regulatory code reveals autism-linked genetic variation.

Neuronal microexons are a highly conserved class of alternative splicing events enriched in genes with critical functions in nervous system biology 1,2 . Microexon splicing is often disrupted in the brains of autistic individuals, and mice with microexon deficiencies display autism-like phenotypes 1,3,4 . However, the full landscape of microexons is not known, and the impact of genetic variation on these exons has not been systematically investigated. Here, through extensive transcriptome profil

Artificial IntelligenceOctober 13, 2026

Deep learning of fossil pollen morphology reveals 25,000 y of ecological change in eastern African grasslands.

Grass (Poaceae) pollen is largely overlooked in investigations of grassland evolution because the pollen of most species cannot be differentiated using traditional optical microscopy. However, the combination of superresolution microscopy and deep learning enables the capture and quantification of distinct morphological variation across the pollen of grass species. Using a semisupervised deep-learning strategy, we trained convolutional neural networks (CNNs) on superresolution images of known mo

Artificial IntelligenceOctober 13, 2026

Tyrosine phosphorylation unfolds nucleophosmin and disrupts its integration into the nucleolus.

Nucleophosmin (NPM1) is a multifunctional nucleolar protein essential for ribosome biogenesis, genome stability, and stress responses. Its integration into the nucleolus depends on its oligomerization and multivalent interactions that enable liquid-liquid phase separation (LLPS). Here, we investigate how tyrosine phosphorylation at Tyr17, Tyr29, and Tyr67 located within the interface between monomers at the N-terminal oligomerization domain regulates NPM1 structure and function. Replacing tyrosi

Artificial IntelligenceOctober 13, 2026

Human-artificial intelligence collaboration for comanaging epilepsy in real time.

One-third of the world's 70 million people with epilepsy have seizures uncontrolled by medication, and implantable devices are a promising treatment. These devices can detect impending seizures, missed medication and impaired cognition, but, like most medical devices, they cannot share this information with patients in real time. This is a missed opportunity for devices to learn from clinical events and teach patients how their behaviors affect their health. Here we present a device platform tha

Artificial IntelligenceOctober 12, 2026

A health-system-scale, episode-resolved multimodal atlas of cancer cachexia.

Cancer cachexia is a wasting syndrome with outsized impact on morbidity and mortality, yet its drivers and systemic consequences remain poorly defined. In this study, we repurposed longitudinal clinicogenomic data from 52,236 patients with cancer profiled by MSK-IMPACT (2010-2024) at Memorial Sloan Kettering Cancer Center to define episode-resolved trajectories of consensus-defined cachectic weight loss and linked these events to outcomes, serology and tumor genotypes. Cachectic episodes cluster

Physical/TechOctober 12, 2026

Multidimensional vulnerabilities of informal settlements in the Urban South

Informal settlements are widespread and rapidly growing, often associated with high vulnerabilities. However, global-scale understanding on their spatial distribution and complexity of vulnerability is limited. Here, we delineate the informal settlements across a sample of 1,914 cities of the Global South by integrating Sentinel-2 imagery with deep learning models, and we explore the heterogeneity of vulnerability using Urban Informal Settlement Vulnerability Index that encompasses living condit

Artificial IntelligenceOctober 8, 2026

Shared patterns of human milk composition link mammary gland function to infant growth.

The mammary gland produces nutrient-rich human milk (HM), yet how mammary functional state shapes HM composition and infant outcomes remains poorly understood. We used HM multi-omics - metabolomics, proteomics, micronutrients, macronutrients, and HM oligosaccharides - across 1,543 samples from three cohorts, including two randomized trials, as a noninvasive readout of mammary functional state. Trajectory modeling and multi-omic integration showed that maternal supplementation improved recovery f

Artificial IntelligenceOctober 8, 2026

Conversational diagnostic artificial intelligence in ambulatory primary care: a prospective feasibility study.

Artificial intelligence (AI)-based systems show promise for assisting primary care providers (PCPs) with patient care. We aimed to evaluate the safety and quality of clinical conversations of a patient-facing conversational AI system, which engaged in real-world urgent primary care appointments. In this prospective, single-centre, single-arm feasibility study, English-speaking patients aged at least 18 years interacted with the Articulate Medical Intelligence Explorer (AMIE) up to 5 days before

Artificial IntelligenceOctober 8, 2026

Advancing mathematics research with AI-driven formal proof search

Large language models (LLMs) increasingly excel at mathematics tasks, but their unreliability limits their utility in mathematics research. A mitigation is to use LLMs to generate formal proofs in languages such as Lean, in which the compiler verifies every proof step. We present the first demonstration of this method’s value in solving open problems at scale. We built an artificial intelligence agent for formal proof search that autonomously resolved nine of 353 open Erdős problems, proved 44/4

Artificial IntelligenceOctober 6, 2026

Large-scale semidefinite programming with graphics processing units.

Semidefinite programming (SDP) provides a powerful framework in applied mathematics with applications spanning optimization, machine learning, quantum computing, and beyond. However, the computational cost of solving large-scale SDP problems remains a significant practical limitation. We break this long-standing computational bottleneck through a synergistic codesign of low-rank algorithms and graphics processing unit (GPU) architectures, developing accelerated first-order methods that leverage

Artificial IntelligenceOctober 6, 2026

Genome stability disrupted by the circRBM39(4,5,6)-RPA2 axis represses breast tumorigenesis.

Genome instability (GI) is a hallmark of cancer. GI is associated with accumulative DNA damage and tumor-specific defects in DNA repair. Multiple antitumor drugs have been developed to promote persistent GI and to drive DNA damage beyond a threshold that tumor cells can survive. However, regulatory roles of circular RNAs (circRNAs) in GI and DNA damage remain elusive. Through circRNA profiling and a small interfering RNA (siRNA)-mediated screen, we have identified a circRNA termed circRBM39(4,5,

Artificial IntelligenceOctober 6, 2026

α-aminobutyric acid promotes tumor progression by polarizing macrophages through the ASNS-mTORC2 axis.

Tumor-associated macrophages (TAMs), the most abundant immune cell subset in the tumor microenvironment (TME), exhibit phenotypic plasticity and exert critical roles in tumor progression and antitumor immunity. Targeting TAM polarization has emerged as a promising strategy for cancer immunotherapy, yet the key regulators governing this process remain incompletely defined. Here, we identified α-aminobutyric acid (AABA) as a pro-tumor metabolite that drives M2-like polarization of TAMs to pr

Artificial IntelligenceOctober 6, 2026

An open vision-language model for diverse medical applications

Abstract Artificial intelligence has high potential for impact in healthcare applications, but its training and deployment are challenging due to diverse data, a complex spectrum of possible tasks and important privacy needs. High-performing foundation models that enable data-efficient fine-tuning for diverse downstream tasks can meaningfully accelerate development in this domain. We introduce MedGemma, a collection of medical vision-language foundation models based on Gemma 3. MedGemma demonstr

Artificial IntelligenceOctober 1, 2026

Keep AI-augmented science contestable

Artificial intelligence (AI) is moving upstream in science. Systems such as AlphaFold have already transformed protein-structure prediction, while newer agentic systems can search literatures, generate hypotheses, write and execute code, interpret results, and automate much of the path from idea to paper. Recent debates have asked how such systems may change scientific progress, surprise, and human understanding. Yet they also change something more fundamental: who—or what—participates in decidi

Artificial IntelligenceSeptember 29, 2026

Plant fiber exploitation contributed to the emergence of ground-edged cutting tools in North China.

Ground-edged cutting tools are widely regarded as technological hallmarks of early agriculture in North China, commonly hypothesized to have been developed primarily for cereal harvesting. Yet the functional foundations of this innovation remain insufficiently tested. This study integrates use-wear and microfossil analyses of 140 stone tools from the Peiligang site, with experimental data, to reassess their roles within long-term trajectories of technological change from the Late Paleolithic to

Artificial IntelligenceSeptember 29, 2026

Transient chaos in the self-organization of dissipative optical solitons.

Transient chaos is a hallmark of complex dynamics in nonlinear systems. Governed by chaotic saddles, it manifests as short-lived yet rich chaotic behavior preceding an abrupt transition to a stable attractor. However, real-time capture and quantitative characterization of such inherently unpredictable and nonrepetitive dynamics remain challenging, obscuring their physical origins. Here, we experimentally demonstrate transient chaos during the self-organization of dissipative optical solitons in

Artificial IntelligenceSeptember 28, 2026

Neural network-augmented Pfaffian wave-functions for scalable simulations of interacting fermions.

Developing accurate numerical methods for strongly interacting fermions is crucial for improving our understanding of various quantum many-body phenomena, especially unconventional superconductivity. Recently, neural quantum states have emerged as a promising approach for studying correlated fermions, highlighted by the hidden fermion and backflow methods, which use neural networks to model corrections to fermionic quasiparticle orbitals. In this work, we expand these ideas to the space of Pfaff

Artificial IntelligenceSeptember 26, 2026

Why friends in common reveal network stars.

The Friendship Paradox states that, on average, your friends have more friends than you do. We extend this to common friends-those who appear in multiple people's friend lists. We show that the more people who share a common friend, the more connected that person tends to be, and we derive an expression quantifying this progression. In a regional Facebook network, a common friend to three randomly sampled individuals has on average more friends than 99.9% of the network. In a citation network, a

Artificial IntelligenceSeptember 26, 2026

Sensory context improves language prediction in humans and LLMs.

Language is a fundamental human capacity. Large language models (LLMs) have presented the first viable model of language outside of humans, yet how these models learn and use language differs significantly from humans. Here, we compare LLMs and humans predicting language with varying levels of sensory information-from disembodied written text to audiovisual videos of speakers-to demonstrate that, in both humans and LLMs, sensory context is critical for optimal performance. We asked human partici

Artificial IntelligenceSeptember 23, 2026

TRAM promotes Toll-like-receptor-free myddosome signal transduction.

In the Toll-like receptor (TLR) signal transduction pathways, activated receptors are present for mere minutes within the initial, membrane-proximal signalling complexes called proto-myddosomes 1 . Proto-myddosomes are rapidly released from TLRs to self-assemble (that is, mature) into enzyme-rich, cytosolic supramolecular organizing centres called myddosomes 2-4 . Myddosomes induce hours-long inflammatory gene expression 5-10 . The mechanism controlling proto-myddosome release and its effect on

Artificial IntelligenceSeptember 22, 2026

Multimodal integration supports neural processing of grammatical negation.

Negation is a fundamental aspect of human cognition and grammar, expressed across different modalities. Yet, little is known about multimodal processing of negation, especially grammatical negation (e.g., "I did not sleep"), and the contribution of gestures and prosodic markers to its neural processing remains unexplored. This study investigates the neurocognition of multimodal negation in Turkish, a verb-final language where standard verbal negation is expressed morphologically with a postverba

Artificial IntelligenceSeptember 21, 2026

Chemist-aligned retrosynthesis by ensembling diverse inductive bias models.

Chemical synthesis remains a critical bottleneck in the discovery and manufacture of functional small molecules 1-3 . While AI-assisted synthesis planning has proliferated in recent years, a detailed understanding of its failure modes has not been achieved, and models still struggle with predicting less frequent, yet strategically critical reactions, as well as hallucinated, incorrect predictions misaligned with chemists' expectations 4-12 . In this work, we analyze the failure modes of current

Artificial IntelligenceSeptember 21, 2026

Using Data Science Tools to Explore Rate Matching in a Nickel-Catalyzed Cross-Electrophile Coupling of Alkyl and Aryl Halides (Cl, Br) with a Tridentate Monoanionic Ligand.

Nickel-catalyzed cross-electrophile coupling (XEC) between aryl and alkyl electrophiles has emerged as an enabling synthetic method, yet predicting the outcomes of specific electrophile pairs is difficult due to the complexity of the mechanistic pathways. It has been proposed that "rate matching" between the oxidative addition of the aryl halide and the rate of formation of the alkyl radical, in addition to the stability of the Ni(II)-Ar complex, dictates the yield of the cross-coupling product.

Artificial IntelligenceSeptember 17, 2026

The Virtual Biotech: A multi-agent AI framework for therapeutic discovery and development

Drug development requires evidence integration across biological scales and modalities, but relevant tools are fragmented. We introduce the Virtual Biotech, an organization of artificial intelligence (AI) agents modeled on a drug-development company, with agentic divisions spanning target discovery, safety assessment, modality selection, and clinical development. We demonstrate its utility at three drug-development decision points. First, over 37,000 agents annotated outcomes from 55,984 trials

Artificial IntelligenceSeptember 16, 2026

Rapid patient-specific neural networks for X-ray to volume registration.

Advanced navigation techniques in image-guided interventions and surgical robotics require the rapid and precise alignment of three-dimensional (3D) preoperative volumes (such as computed tomography and magnetic resonance imaging) to two-dimensional (2D) intraoperative images (such as X-ray fluoroscopy) 1,2 . However, existing 2D/3D registration methods fail to generalize across the broad spectrum of fluoroscopy-guided procedures: intensity-based optimizers require per-individual hyperparameter

Artificial IntelligenceSeptember 16, 2026

Reimagining research papers as interactive and reliable AI agents.

Here we introduce Paper2Agent, an automated framework that converts research papers into artificial intelligence (AI) agents. Paper2Agent transforms research output from passive artefacts into active systems that accelerate use and discovery. Conventional research papers require readers to understand and adapt the paper's code, data and methods to their work, creating barriers to dissemination and reuse. Paper2Agent addresses this challenge by converting a paper into an AI agent that functions a

Artificial IntelligenceSeptember 10, 2026

Treatment Decisions in Multiple Myeloma.

Revolutions in transplantation and targeted and immune therapies have transformed multiple myeloma from a disease with an associated survival of a few years into one for which functional cure is an emerging goal. This abundance of effective therapies has created clinical complexity. Here we provide a practical framework, anchored in trial evidence and informed by emerging biologic discoveries, for the navigation of treatment decisions across the disease spectrum. We outline how cytogenetic and g

Artificial IntelligenceSeptember 8, 2026

Systematic discovery of circular permutations across the protein universe using CIRPIN.

Protein structure search has been revolutionized by deep learning methods that can rapidly search massive databases. However, current structure search tools often miss proteins related by topological rearrangements, particularly circular permutation, wherein proteins share highly similar structure but differ in the positioning of their termini. We introduce a circular permutation-invariant graph neural network (CIRPIN) that addresses this limitation through a data augmentation strategy using syn

Artificial IntelligenceSeptember 3, 2026

High-capacity associative memory in a quantum-optical spin glass

The Hopfield neural network stores memories using all-to-all-coupled spins and recalls those memories through equilibrium dynamics. Storing too many hampers recall because frustration causes an exponential number of spurious patterns to arise as the network becomes a spin glass. Despite this, memory recall can be restored, and even enhanced, under quantum-optical nonequilibrium dynamics because spurious patterns can now serve as reliable memories. We experimentally observe associative memory wit