Search

Abstract Machine learning offers a promising avenue for expediting the discovery of new compounds by accurately predicting their thermodynamic stability. This approach provides significant advantages in terms of time and resource efficiency compared to traditional experimental and modeling methods. However, most existing models are constructed based on specific domain knowledge, potentially introducing biases that impact their performance. Here, we propose a machine learning framework rooted in electron configuration, further enhanced through stack generalization with two additional models grounded in diverse domain knowledge. Experimental results validate the efficacy of our model in accurately predicting the stability of compounds, achieving an Area Under the Curve score of 0.988. Notably, our model demonstrates exceptional efficiency in sample utilization, requiring only one-seventh of the data used by existing models to achieve the same performance. To underscore the versatility of our approach, we present three illustrative examples showcasing its effectiveness in navigating unexplored composition space. We present two case studies to demonstrate that our method can facilitate the exploration of new two-dimensional wide bandgap semiconductors and double perovskite oxides. Validation results from first-principles calculations indicate that our method demonstrates remarkable accuracy in correctly identifying stable compounds.

Non-alcoholic fatty liver disease (NAFLD) is a significant health burden in youths and young adults, and the trend toward younger onset of NAFLD is alarming. Utilizing data from the Global Burden of Disease (GBD) 2021 study, this study assessed the burden and trends of NAFLD in youths and young adults aged 15–39 from 1990 to 2021 and extracted data from this study on the incidence, prevalence, death, and disability-adjusted life-years (DALYs) rates of NAFLD. We evaluated the global temporal trend of NAFLD from 1990 to 2021 with estimated annual percentage change (EAPC) and age-standardized rate (ASR). The Bayesian age-period-cohort (BAPC) model was used to predict future trends of the NAFLD burden to 2035. We found that the global burden of NAFLD in youths and young adults has risen steadily from 1990 to 2021, and projects to increase to 2035, which places enormous pressure on society. To alleviate this burden, implementing measures targeting risk factors such as glycemic control and smoking cessation is necessary.

The mechanisms that concentrated metals in the world’s single largest nickel–copper– platinum-group element (Ni–Cu–PGE) sulfide deposit, the Jinchuan deposit in northwestern China, remain enigmatic. Nickel and Cu isotopes are novel isotopic systems that may fill this knowledge gap as they are major constituents of sulfide melt, the main carrier of economically important metals. Sulfide ores in the Jinchuan deposit are characterized by systematically high δ60/58 Ni and δ65Cu values, which are positively correlated (r² = 0.84), and show well-defined correlations with PGEs, S, Ni, and Cu contents, and Ni/Pd, Cu/Pd, and Pd/Ir ratios. The isotopic signatures can be best explained by the early segregation of sulfide melt at depth prior to ascent of the parental magmas, storing isotopically light Ni and Cu in deep magma chambers. Subsequent sulfide melt segregation and fractionation in shallow portions of the magmatic conduit generated the currently explored sulfide ores, with heavy and positively correlated Ni–Cu isotopic signatures. Therefore, hidden deep orebodies are promising targets for further exploration beneath the Jinchuan deposit. [ABSTRACT FROM AUTHOR]