The heat of formation and band gap are important properties of halide double perovskites, which can dictate the range of application. However, traditional laborious experiments and density functional theory (DFT) calculations are time-consuming, and machine learning methods perform poorly in acquiring hidden knowledge from material samples, which make it difficult to explore promising perovskites with high stability efficiently. Here, we present a new machine learning framework, which could embed material domain knowledge into a machine learning method, and then it is used to predict the heat of formation and band gap. Compared with traditional machine learning methods, we found that the prediction accuracy of the heat of formation and band gap in 540 perovskite materials A₂B⁺B³⁺X₆ is improved by 5.65% and 15.52% (namely 93.58% and 62.65%) respectively. It is shown that material domain knowledge contains a wealth of logical rules and expert experience, which can remedy most of the problems in data-driven materials research, such as data sparse and complex relationships between features.

The mitochondrial displacement loop (D-loop) is known to accumulate mutations and single nucleotide polymorphisms (SNPs) at a higher frequency than other regions of mitochondrial DNA (mtDNA). This is a case-control study. We sequenced SNPs in the D-loop of mtDNA and investigated their association with the risk of chronic kidney disease (CKD). A total of 144 SNPs referring to the positions of the Revised Cambridge Reference Sequence (rCRS) for mitochondrial genome were identified in a case-control study. The minor alleles of nucleotides 73G, 146C, 150T, 194T, 195C and 310C were associated with an increased risk for CKD patients. Analysis of genetic polymorphisms in the mitochondrial D-loop can help identify the people who are at a high risk of developing chronic kidney disease. These SNPs can be considered as potential predictors for CKD.

Rapid antimicrobial resistance (AMR) prediction from MALDI-TOF mass spectrometry (MS) remains challenging, particularly when training artificial intelligence (AI) models under small-sample constraints. Performance is often hampered by the high dimensionality of spectral data and the subtle nature of resistance-related signals: full-spectrum approaches risk overfitting to high-dimensional noise, whereas peak-selection strategies risk discarding structurally informative, low-intensity signals. Here, we propose ReShuffle-MS, a region-guided data augmentation framework for MS data. Each spectrum is partitioned into a Main Discriminative Region (MDR) and a Peripheral Peak Region (PPR). By recombining signals within the PPR across samples of the same class while keeping the MDR intact, ReShuffle-MS generates structure-preserving augmented samples. On a clinical dataset for Escherichia coli (E. coli) levofloxacin resistance prediction, ReShuffle-MS delivered significant and consistent performance gains. It improved the average accuracy of classical machine learning models by 3.7% and enabled a one-dimensional convolutional neural network (CNN) to achieve 83.25% accuracy and 97.28% recall. Visualization using Grad-CAM revealed a shift from sparse, peak-dependent attention toward broader and more meaningful spectral patterns. Validation on the external DRIAMS-C dataset for ceftriaxone resistance further demonstrated that the method generalizes to a distinct laboratory setting and a different antibiotic target. These findings suggest that ReShuffle-MS can enhance the robustness and clinical utility of AI-based AMR prediction from routinely acquired MALDI-TOF spectra.

Although the phase field method is a robust tool for theoretical studies of ferroelectrics, determining the parameters of the Helmholtz free energy in the phase-field model, particularly the Landau coefficients, remains a highly complex and challenging task. Here, we propose a general approach to identify all phase field parameters by developing an artificial intelligence-assisted multi-scale phase field model. This model hierarchically bridges ab initio accuracy with the mesoscale phase field model, linked by the effective Hamiltonian model and deep potential molecular dynamics (DPMD) simulations, effectively overcoming the limitations inherent in relying on a single method. Specifically, Monte Carlo simulations and DPMD calculations are used from first principles to determine temperature-dependent polarization and dielectric constants, which are then fitted to thermodynamic potentials using the particle swarm optimization algorithm. In addition to the Landau–Devonshire energy function, other material properties in the phase field simulation, such as gradient coefficients, electrostriction, and elastic coefficients, are also directly calculated from first-principles calculation, establishing a multi-scale phase field model. To effectively demonstrate the proposed multi-scale model, we have chosen BaxSr1−xTiO3 (BST) solid solutions and two-dimensional (2D) ferroelectric α-In2Se3 as representative examples, showcasing its applicability to both perovskite and 2D ferroelectrics. Based on the multi-scale phase field models developed for BST and α-In2Se3, we performed phase field simulations to explore their thermodynamic properties.