How do multi-scale features and attention mechanisms optimize apple disease identification?
Peer-Reviewed Publication
Updates every hour. Last Updated: 27-Jul-2025 13:10 ET (27-Jul-2025 17:10 GMT/UTC)
Research team used ProteinMPNN to expand the sequence space of synthetic binding proteins (SBPs), improving their solubility and stability, and showed ProteinMPNN-designed proteins outperform classical methods.
A review on machine learning-based prediction methods for drug side effects sorts out methods for predicting side effects caused by single drugs and DDIs, highlights the prediction of side effect frequency and severity, and discusses current challenges and future directions.
B lymphocytes exhibit dual roles in tumorigenesis, acting as both allies and adversaries in the tumor microenvironment (TME). Their anti-tumor functions include recognizing tumor-associated antigens, producing antibodies, activating cytotoxic immune responses, and forming tertiary lymphoid structures (TLS) that enhance immune cell coordination. Tumor-infiltrating B cells (TIL-Bs) within TLS contribute to improved patient survival and immunotherapy responses by facilitating antibody class switching, somatic hypermutation, and cytokine secretion that recruit and activate T cells, natural killer (NK) cells, and dendritic cells (DCs). Antibodies from B cells mediate complement-dependent cytotoxicity (CDC), antibody-dependent cellular phagocytosis (ADCP), and antibody-dependent cell-mediated cytotoxicity (ADCC), directly eliminating tumor cells. Additionally, B cells present antigens to T cells and secrete cytokines like IFN-γ and CXCL13, amplifying anti-tumor immunity. However, regulatory B cells (Bregs) and other subsets suppress immune responses by secreting IL-10, TGF-β, and VEGF, promoting angiogenesis, recruiting immunosuppressive cells like myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs), and expressing immune checkpoints like PD-L1. This duality underscores the complexity of targeting B cells in cancer therapy.
Recent advances in cancer research have underscored the critical role of myeloid cells in shaping tumor microenvironments (TME), influencing tumor progression, immune evasion, and therapeutic resistance. Myeloid cells, including tumor-associated macrophages (TAMs) and myeloid-derived suppressor cells (MDSCs), exhibit functional plasticity driven by interactions with tumor cells, stromal components, and metabolic adaptations. These cells not only directly promote tumor growth by enhancing angiogenesis, matrix remodeling, and metastasis but also suppress anti-tumor immunity through nutrient deprivation, oxidative stress, and cytokine-mediated inhibition of T and NK cells. Their dual roles—both pro-tumorigenic and occasionally anti-tumorigenic—highlight their complexity and context-dependent behavior across cancer types.
Researchers from Shanghaitech University, Shanghai Institute of Microsystem and Information Technology, Institute of Geology and Geophysics, Chinese Academy of Sciences have conducted measurements of X-ray fluorescence spectra excited by X-rays and electrons based on microcalorimeters. It was demonstrated that there are obvious differences between the fluorescence spectral lines excited by electrons and photons. This is a good example for elemental analysis based on fluorescence spectra in electron scanning microscopes. It also provides clear guidance for choosing X-ray excitation sources (such as Fe55 or X-ray tubes) or electron excitation sources (such as Sr90) in space exploration.
In the exploration of celestial bodies, such as Mars, the Moon, and asteroids, X-ray fluorescence analysis has emerged as a critical tool for elemental analysis. However, the varying selection rules and excitation sources introduce complexity. Specifically, these discrepancies can cause variations in the intensities of the characteristic spectral lines emitted by identical elements. These variations, compounded by the minimal energy spacing between these spectral lines, pose substantial challenges for conventional silicon drift detectors (SDD), hindering their ability to accurately differentiate these lines and provide detailed insights into the material structure. To overcome this challenge, a cryogenic X-ray spectrometer based on transition-edge sensor (TES) detector arrays is required to achieve precise measurements. This study measured and analyzed the K-edge characteristic lines of copper and the diverse L-edge characteristic lines of tungsten using a comparative analysis of the electron and X-ray excitation processes. For the electron excitation experiments, copper and tungsten targets were employed as X-ray sources, as they emit distinctive X-ray spectra upon electron-beam bombardment. In the photon excitation experiments, a molybdenum target was used to produce a continuous spectrum with the prominent Mo Kα lines to emit pure copper and tungsten samples. TES detectors were used for the comparative spectroscopic analysis. The initial comparison revealed no substantial differences in the Kα and Kβ lines of copper across different excitation sources. Similarly, the Lα lines of tungsten exhibited uniformity under different excitation sources. However, this investigation revealed pronounced differences within the Lβ line series. The study found that XRF spectra preferentially excite outer-shell electrons, in contrast to intrinsic spectra, owing to different photon and electron interaction mechanisms. Photon interactions are selection-rule-dependent and involve a single electron, whereas electron interactions can involve multiple electrons without such limitations. This leads to varied excitation transitions, as evidenced in the observed Lβ line series.
A research team has successfully assembled the first high-quality, chromosome-level genome of published findings reveal how specific genes and hormone regulatory networks drive the plant’s exceptional growth rate and stress tolerance, offering insights into its invasive nature, opening new avenues for sweet potato crop improvement.