Nature最新文献

Given the high recurrence rates of hepatocellular carcinoma (HCC) post-resection^1,2,3, improved early identification of patients at high risk for post-resection recurrence would help to improve patient outcomes and prioritize healthcare resources^4,5,6. Here we observed a spatial and HCC recurrence-associated distribution of natural killer (NK) cells in the invasive front and tumour centre from 61 patients. Using extreme gradient boosting and inverse-variance weighting, we developed the tumour immune microenvironment spatial (TIMES) score based on the spatial expression patterns of five biomarkers (SPON2, ZFP36L2, ZFP36, VIM and HLA-DRB1) to predict HCC recurrence risk. The TIMES score (hazard ratio = 88.2, P < 0.001) outperformed current standard tools for patient risk stratification including the TNM and BCLC systems. We validated the model in 231 patients from five multicentred cohorts, achieving a real-world accuracy of 82.2% and specificity of 85.7%. The predictive power of these biomarkers emerged through the integration of their spatial distributions, rather than individual marker expression levels alone. In vivo models, including NK cell-specific Spon2-knockout mice, revealed that SPON2 enhances IFNγ secretion and NK cell infiltration at the invasive front. Our study introduces TIMES, a publicly accessible tool for predicting HCC recurrence risk, offering insights into its potential to inform treatment decisions for early-stage HCC.

Many natural motor skills, such as speaking or locomotion, are acquired through a process of trial-and-error learning over the course of development. It has long been hypothesized, motivated by observations in artificial learning experiments, that dopamine has a crucial role in this process. Dopamine in the basal ganglia is thought to guide reward-based trial-and-error learning by encoding reward prediction errors¹, decreasing after worse-than-predicted reward outcomes and increasing after better-than-predicted ones. Our previous work in adult zebra finches—in which we changed the perceived song quality with distorted auditory feedback—showed that dopamine in Area X, the singing-related basal ganglia, encodes performance prediction error: dopamine is suppressed after worse-than-predicted (distorted syllables) and activated after better-than-predicted (undistorted syllables) performance². However, it remains unknown whether the learning of natural behaviours, such as developmental vocal learning, occurs through dopamine-based reinforcement. Here we tracked song learning trajectories in juvenile zebra finches and used fibre photometry³ to monitor concurrent dopamine activity in Area X. We found that dopamine was activated after syllable renditions that were closer to the eventual adult version of the song, compared with recent renditions, and suppressed after renditions that were further away. Furthermore, the relationship between dopamine and song fluctuations revealed that dopamine predicted the future evolution of song, suggesting that dopamine drives behaviour. Finally, dopamine activity was explained by the contrast between the quality of the current rendition and the recent history of renditions—consistent with dopamine’s hypothesized role in encoding prediction errors in an actor–critic reinforcement-learning model^4,5. Reinforcement-learning algorithms⁶ have emerged as a powerful class of model to explain learning in reward-based laboratory tasks, as well as for driving autonomous learning in artificial intelligence⁷. Our results suggest that complex natural behaviours in biological systems can also be acquired through dopamine-mediated reinforcement learning.

Global ocean surface temperatures were at record levels for more than a year from April 2023 onwards, exceeding the previous record in 2015–2016 by 0.25 °C on average between April 2023 and March 2024¹. The nearly global extent and unprecedented intensity of this event prompted questions about how exceptional it was and whether climate models can represent such record-shattering jumps in surface ocean temperatures². Here we construct observation-based synthetic time series to show that a jump in global sea surface temperatures that breaks the previous record by at least 0.25 °C is a 1-in-512-year event under the current long-term warming trend (1-in-205-year to 1-in-1,185-year event; 95% confidence interval). Without a global warming trend, such an event would have been practically impossible. Using 270 simulations from a wide range of fully coupled climate models, we show that these models successfully simulate such record-shattering jumps in global ocean surface temperatures, underpinning the models’ usefulness in understanding the characteristics, drivers and consequences of such events. These model simulations suggest that the record-shattering jump in surface ocean temperatures in 2023–2024 was an extreme event after which surface ocean temperatures are expected to revert to the expected long-term warming trend.

Habitat fragmentation generally reduces biodiversity at the patch scale (α diversity)¹. However, there is ongoing debate about whether such negative effects can be alleviated at the landscape scale (γ diversity) if among-patch diversity (β diversity) increases as a result of fragmentation^2,3,4,5,6. This controversial view has not been rigorously tested. Here we use a dataset of 4,006 taxa across 37 studies from 6 continents to test the effects of fragmentation on biodiversity across scales by explicitly comparing continuous and fragmented landscapes. We find that fragmented landscapes consistently have both lower α diversity and lower γ diversity. Although fragmented landscapes did tend to have higher β diversity, this did not translate into higher γ diversity. Our findings refute claims that habitat fragmentation can increase biodiversity at landscape scales, and emphasize the need to restore habitat and increase connectivity to minimize biodiversity loss at ever-increasing scales.

Frictional interfaces are found in systems ranging from biological joints to earthquake faults. When and how these interfaces slide is a fundamental problem in geosciences and engineering^{1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20}. It is believed that there exists a threshold shear force, called static friction, below which the interface is stationary^4,10, despite many studies suggesting that this concept is outdated^{1,21,22,23,24,25,26,27,28}. By contrast, rate-and-state friction formulations^1,26,27 predict that interfaces are always sliding²⁹, but this feature is often considered an artefact that calls for modifications³⁰. Here we show that nominally stationary interfaces subjected to constant shear and normal loads, with a driving force that is notably below the classically defined static friction for which creep is known to occur^9,27,28,29, are sliding, but with diminishingly small rates down to 10⁻¹² m s⁻¹. Our precise measurements directly at the interface are enabled by digital image correlation^18,31,32. This behaviour contradicts classical models of friction but confirms the prediction of rate-and-state friction^1,26,27. The diminishing slip rates of nominally stationary interfaces reflect interface healing, which would manifest itself in higher peak friction in subsequent slip events^15,27,33, such as earthquakes and landslides, substantially modifying their nucleation and propagation and hence their hazard^3,12,13,34.