Search

The authors demonstrate that targeted protein degradation of the dengue virus capsid suppresses infection through capsid-dependent pathways rather than classical inhibition, revealing a new antiviral strategy effective against resistant viral variants.

Papaya is a trioecious species with XX females, XY males, and XY^h hermaphrodites, and the combination of Y and Y^h chromosomes is lethal. Here, the authors identify the degeneration of the YY lethality gene on the Y chromosome as the causal balancing lethal factor that reenforces dioecy and stabilizes balanced sex ratios.

Here the authors present scLong, a billion parameter foundation model trained on 48M single cells, using self attention across all 28k human genes and Gene Ontology knowledge. It captures long range gene context in single Cell transcriptomics.

While therapies targeting type I BRAF mutations have been developed, there are limited options for those with type II and III mutations. Here, the authors identify a subset of BRAF-mutant non-small cell lung cancer patients and characterise the pan-RAF inhibitor exarafenib, demonstrating efficacy in preclinical models and investigating subsequent resistance mechanisms.

The APOE-ε4 allele is the strongest genetic risk factor for late-onset Alzheimer’s disease, but it is not deterministic. Here, the authors show that common genetic variation changes how APOE-ε4 influences cognition.

Biocatalysis of the chemotherapy drug, doxorubicin, relies on the cytochrome P450 DoxA, which is inefficient. Here, the authors ameliorated the biosynthetic limitations by identifying DoxA redox partners and DnrV, which prevents product inhibition, helping improve microbial production.

Humanity’s Last Exam, a multi-modal benchmark at the frontier of human knowledge, is designed to be an expert-level closed-ended academic benchmark with broad subject coverage.

Analysis of the somatic and transcriptomic profile of 123 acral melanoma samples from Mexican patients helps understand tumour origins and prognosis, and highlights the importance of including samples from diverse ancestries in cancer genomics studies.

Analysis combining multiple global tree databases reveals that whether a location is invaded by non-native tree species depends on anthropogenic factors, but the severity of the invasion depends on the native species diversity.

Post-transcriptional regulation of mRNA translation was explored using Ribo-STAMP and single-cell RNA sequencing to reveal cell-type-specific and isoform-specific translation patterns across hippocampal neuronal and non-neuronal cell types, highlighting functional differences between CA1 and CA3.

Trimodal single-cell analysis reveals gene-regulatory architecture in complex tissues.

In one-shot perceptual learning, what we see can be dramatically altered by a single past experience. Using psychophysics, fMRI, iEEG, and DNNs, the authors identify neural and computational mechanisms underlying this remarkable ability in humans.

cellSTAAR advances noncoding rare variant association analysis by integrating single-cell genomics data.

Donahue et al. show that ageing is associated with changes in ER morphology. ER-phagy drives age-associated ER remodelling through tissue-specific factors.

MedHELM, an extensible evaluation framework including a new taxonomy for classifying medical tasks and a benchmark of many datasets across these categories, enables the evaluation of large language models on real-world clinical tasks.

Analysis of cancer genome sequencing data has enabled the discovery of driver mutations. Here, as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium the authors present DriverPower, a software package that identifies coding and non-coding driver mutations within cancer whole genomes via consideration of mutational burden and functional impact evidence.

Researchers studied the blood-based metabolome of over 23,000 people from ten ethnically diverse cohorts. They identified 235 metabolites associated with future risk of type 2 diabetes (T2D). By integrating genetic and modifiable lifestyle factors, their findings provide insights into T2D mechanisms and could improve risk prediction and inform precision prevention.

Spiking neural networks are generally used for sequential information and event data processing but still lack high performance. Through algorithm and hardware co-design, Zhang et al. report a State Space Model based approach to implement on compute-in memory hardware, enabling asynchronous and real-time processing capability with high energy efficiency for event sequences.

Polygenic scores for body mass index and obesity constructed using data from 5 million people with different ancestries were associated with distinct weight gain trajectories from early childhood to adulthood.

The biological clock model LifeClock predicts biological age across all life stages from routine clinical data, revealing distinct pediatric and adult disease risk patterns.

This study reports that de novo germline missense variants in SF3B1, distinct from the somatic variants frequently observed in cancer, cause a neurodevelopmental disorder and disrupt global RNA splicing.

Lipopeptides are used to combat drug-resistant infections, but drawbacks limit their application. Here, Lim et al. contrast the mechanisms of lipopeptides that are superficially similar yet disrupt membranes in distinctly different ways.

Here the authors report that some aspects of clinical heterogeneity in type 2 diabetes vary across populations. Using a deep-learning–based tree model built from over 32,000 patients, they document disease patterns and risks specific for the Chinese population, potentially enabling more precise prediction and personalized care.

The flagship paper of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes Consortium describes the generation of the integrative analyses of 2,658 cancer whole genomes and their matching normal tissues across 38 tumour types, the structures for international data sharing and standardized analyses, and the main scientific findings from across the consortium studies.

With the generation of large pan-cancer whole-exome and whole-genome sequencing projects, a question remains about how comparable these datasets are. Here, using The Cancer Genome Atlas samples analysed as part of the Pan-Cancer Analysis of Whole Genomes project, the authors explore the concordance of mutations called by whole exome sequencing and whole genome sequencing techniques.

MultiSTAAR provides a general and flexible statistical framework for functionally informed multi-trait rare variant analysis of biobank-scale sequencing studies by jointly analyzing multiple traits and incorporating annotation information.

Spatial transcriptomics technologies are still too restrictive for widespread clinical use, and methods that have been designed to bridge them with histopathology carry important limitations. Here, the authors develop MISO, a deep learning framework that allows inferring tissue spatial organisation and gene expression with near single-cell resolution from histopathology images.

Analyses of 2,658 whole genomes across 38 types of cancer identify the contribution of non-coding point mutations and structural variants to driving cancer.

A large genome-wide association study of more than 5 million individuals reveals that 12,111 single-nucleotide polymorphisms account for nearly all the heritability of height attributable to common genetic variants.

Resistance to combination therapies has been reported in rhabdomyosarcoma (RMS). Here, the authors discover that PIK3CA/AKT pathway regulation of multidrug-resistant ABC transporters is involved in the resistance to therapies in RMS, and use of the PI3Kα inhibitor alpelisib re-sensitizes RMS to therapy.

Specific pain-sensing nociceptors and chemosensory tuft cells form a circuit to promote type 2 immune responses in the intestine.

The potential of mammogram-based biological age predictors remains to be fully explored. Here this group introduces a deep learning model to estimate the biological age of the breast using healthy mammograms.

A completely solid-state, single-chip, microwave-frequency surface acoustic wave phonon laser can generate coherent phonons from thermal noise or resonantly amplify injected phonons using only a direct current bias field.

Federated learning (FL) algorithms have emerged as a promising solution to train models for healthcare imaging across institutions while preserving privacy. Here, the authors describe the Federated Tumor Segmentation (FeTS) challenge for the decentralised benchmarking of FL algorithms and evaluation of Healthcare AI algorithm generalizability in real-world cancer imaging datasets.

Using well-calibrated statistical methods the authors jointly analyze Undiagnosed Diseases Network genomes, identifying known and novel disease genes. Software is publicly available to support future cross-cohort rare disease discovery efforts.

There’s an emerging body of evidence to show how biological sex impacts cancer incidence, treatment and underlying biology. Here, using a large pan-cancer dataset, the authors further highlight how sex differences shape the cancer genome.

Multi-omics datasets pose major challenges to data interpretation and hypothesis generation owing to their high-dimensional molecular profiles. Here, the authors develop ActivePathways method, which uses data fusion techniques for integrative pathway analysis of multi-omics data and candidate gene discovery.

Here the authors provide an explanation for 95% of examined predicted loss of function variants found in disease-associated haploinsufficient genes in the Genome Aggregation Database (gnomAD), underscoring the power of the presented analysis to minimize false assignments of disease risk.

This work introduces Deep-STARmap and Deep-RIBOmap, imaging-based sequencing platforms designed to profile transcriptional and translational activity in thick tissue blocks.

Integrative analyses of transcriptome and whole-genome sequencing data for 1,188 tumours across 27 types of cancer are used to provide a comprehensive catalogue of RNA-level alterations in cancer.

Whole-genome sequencing data for 2,778 cancer samples from 2,658 unique donors across 38 cancer types is used to reconstruct the evolutionary history of cancer, revealing that driver mutations can precede diagnosis by several years to decades.

This study reveals that the protein AKAP11 plays a crucial role in regulating neuronal signaling and synaptic function by linking PKA activity to autophagy. Loss of AKAP11 distorts compartment-specific PKA and GSK3α/β activities and impairs neurotransmission, highlighting a shared molecular pathway between bipolar disorder and schizophrenia.

Rudensky and colleagues demonstrate a context-dependent differential requirement for Foxp3 for T_reg cell transcriptional and functional programs.

A pangenome of oat, assembled from 33 wild and domesticated oat lines, sheds light on the evolution and genetic diversity of this cereal crop and will aid genomics-assisted breeding to improve productivity and sustainability.

Species’ traits and environmental conditions determine the abundance of tree species across the globe. Here, the authors find that dominant tree species are taller and have softer wood compared to rare species and that these trait differences are more strongly associated with temperature than water availability.

Multisystem inflammatory syndrome in children (MIS-C) onsets in COVID-19 patients with manifestations similar to Kawasaki disease (KD). Here the author probe the peripheral blood transcriptome of MIS-C patients to find signatures related to natural killer (NK) cell activation and CD8+ T cell exhaustion that are shared with KD patients.