Rongrong Jia

Introduction/Objective: Quantitative Structure–Activity Relationship (QSAR) modeling is a critical computational strategy in drug discovery; however, data heterogeneity and experimental noise frequently compromise model reliability. In this study, QSAR models were developed to predict the inhibitory activity of 9,10-dihydrophenanthrene derivatives against the SARS-CoV-2 3CLpro enzyme using a novel fuzzy Support Vector Regression (FSVR) framework. The proposed approach incorporates a fuzzy membership function to weight activity data according to their reliability and employs both the full molecular descriptor set and a subset refined by Variance Inflation Factor (VIF) analysis. Methods: Four regression models-FSVR-VIF, FSVR-All, SVR-VIF, and SVR-All-were constructed using a radial basis function kernel, with hyperparameters optimized via grid search combined with 10-fold cross-validation. Model performance was evaluated using the coefficient of determination (R²) and the root-mean-square error (RMSE). Results: FSVR consistently outperformed conventional SVR in terms of accuracy and robustness. Among the four models, FSVR-VIF achieved the best cross-validation performance (R²(CV) = 0.9621). The model also demonstrated good predictive performance on the independent external validation set, yielding an R²(ext) of 0.828. Residual error analyses revealed symmetrically distributed residuals with markedly fewer extreme deviations for the fuzzy-enhanced models. Notably, the FSVR-VIF model exhibited a narrow residual spread, with most compounds falling within ±0.5 units across both the training and test sets. Discussion: These findings suggest that the incorporation of fuzzy logic improved model robustness against data uncertainty, while VIF-based descriptor selection enhanced model parsimony and interpretability. Furthermore, descriptor analysis indicated that molecular symmetry and mass distribution (e.g., X23) were positively associated with inhibitory activity, whereas increased methyl substitution and structural branching characteristics (e.g., X7 and X11) tended to be associated with reduced inhibitory potency. Conclusion: Overall, the integration of fuzzy logic with SVR and VIF-based feature selection provided a parsimonious local QSAR model for the investigated series of 9,10-dihydrophenanthrene derivatives, yielding improved predictive performance and descriptor interpretability while maintaining satisfactory predictive capability on an independent external dataset.

In rare-earth orthoferrites (RFeO3), the R3+-Fe3+ exchange interaction plays a crucial role in modulating the magnetic properties of single crystals. Here, we report the effect of Tb3+ substitution on spin switching in an Er0.9Tb0.1FeO3 single crystal. Along the a-axis, Type-II spin switching (SSW-II) is observed under zero-field-cooling warming (ZFC) mode, while multi-step SSW-II occur under field-cooled cooling (FCC) mode. These behaviors are absent in the parent ErFeO3 and TbFeO3 single crystals. Both SSW-II processes are highly sensitive to external fields and can be tuned by magnetic fields below 50 Oe. These two SSW phenomena are attributed to the flipping of the Tb3+ sublattice moment, which is driven by the Tb3+-Fe3+ exchange interaction. We further find that the critical field Hc for the field-induced spin switching reaches a maximum value of 463 Oe at 15.5 K. Compared with parent ErFeO3, 10% Tb3+ substitution reduces the maximum Hc by 84.57% owing to a weakened Er3+-Fe3+ interaction and the consequent suppression of magnetic pinning between the Fe3+ and Er3+ sublattices. These results demonstrate that the competition between (Er3+, Tb3+)-4f and Fe3+-3d electrons in Er0.9Tb0.1FeO3 gives rise to spin switching that is highly tunable by low magnetic fields and temperature, making this material a promising candidate for low-power spintronic devices.

Crystallography is fundamental to elucidating the relationship between microscopic structure and macroscopic material properties, and its progress is closely associated with advances in computational tools and software. This review traces the development of crystallographic software from its origins in the 1960s at Oak Ridge National Laboratory to contemporary integrated platforms. The main aspects covered include the establishment and advancement of structural databases such as CSD and ICSD, the development of single-crystal structure determination programs such as SHELX and Olex2, and the evolution of powder diffraction tools, represented by HighScore Suite and TOPAS, toward integrated platforms. Crystallographic visualization tools have likewise evolved from desktop-based programs such as VESTA and Mercury to modern web-based platforms (JSmol and NGL Viewer). Moreover, we discuss the future directions of crystallographic software, including machine-readable data ecosystems, the application of artificial intelligence in structure determination and prediction, and the development of cloud-based immersive visualization platforms. Overall, this review comprehensively shows the evolution, current situation, and future directions of crystallographic software, which is of certain help to the research and development in related fields.

With the rapid development of the hydrogen energy economy, there is an increasingly urgent demand for low-power-consumption and integrable sensors capable of accurate and real-time monitoring of hydrogen (H₂). In this study, SnO₂ nanoparticles rich in oxygen vacancies were successfully prepared via a metal–organic framework (MOF)-derived method. Based on this, highly dispersed Pt-modified SnO₂ nanoparticles were fabricated using a two-step annealing method, and a MEMS H₂ sensor with high response and low power consumption was developed. Studies have shown that the 0.5%Pt-SnMOF/600-SnO₂ sensor exhibits a high response of 31.5 to 100 ppm of H₂ at an operating temperature of 201 °C, which is 2.7 times that of the SnMOF/600-SnO₂ sensor (with an optimal operating temperature of 244 °C), and its power consumption is only 22.1 mW. Furthermore, this sensor demonstrates excellent comprehensive performance, including an extremely low limit of detection of 73.6 ppb, outstanding selectivity, and good long-term stability. Mechanistic studies indicate that the enhancement of gas-sensing performance is due to a combination of factors: plentiful oxygen vacancies, the role of Pt in promoting oxygen reactions, and its effect on the material’s electronic properties.

Low-cost MnO ₂ efficiently upcycles mixed plastics into high-performance additives in one pot, turning waste into value without costly purification.

Developing high-response, low-cost H₂ sensors is critical for real-time H₂ monitoring in the new energy era. In this work, ultra-low content (0.07 wt%) single-atom Pt-loaded SnO₂/MEMS H₂ sensors were prepared by an extended two-step annealing method, enabling ppb-level H₂ sensing with low power consumption. At the optimal operating temperature of 201 °C, the sensor showed a response of 55.0 to 100 ppm H₂, which is 9.17 times that of the pure SnO₂ sensor. Compared with SnO₂ sensors loaded with Pt via traditional impregnation, its optimal operating temperature is reduced by nearly 30 °C, and its response value is increased by 45.0. Additionally, the sensor exhibited a fast response time of 2.3 s and a limit of detection as low as 36 ppb. Mechanistic studies reveal that,, compared to traditional nanoparticle-modified material, the single-atom Pt-modified material exhibits a higher adsorbed oxygen content and enhanced surface oxidation activity. These results indicate single-atom Pt enhances the active oxygen level, underscoring its critical role in boosting H_2-sensing performance.

Carbon-modified TiO ₂ enables low-cost oxidative upcycling of waste PE under mild conditions, selectively converting PE into dicarboxylic acids with high efficiency and good recyclability.