Biodiversity in the United States faces mounting anthropogenic pressures, with an increasing number of species at risk of extinction. We present the most extensive synthesis of recorded threats confronting U.S. imperiled species to date, considering species assessments from NatureServe and the IUCN Red List for over 16,000 species. Our results demonstrate the value of identifying fine-scale threats species face. Species occupying multiple ecosystems tend to see more recorded threats on average, and threat prevalence differs by ecosystem. Certain taxa (particularly corals, reptiles, amphibians, bivalves, and mammals) face disproportionately high recorded threat numbers. Climate change is both the fastest-growing recorded threat and the most ubiquitous IUCN threat category among species with up-to-date assessments. This analysis also highlights significant data gaps, revealing that most imperiled species have outdated or incomplete threat assessments, and over half lack even an initial evaluation. Gaps are disproportionately found among invertebrates, plants, and fungi.
Finding the ideal biomaterial with all required properties to replace specific tissues is an unmet challenge. In this study, twin-layered fibers with two well-established medical biomaterials, PHAs and PLA, have been produced via pressurized gyration and characterized thoroughly for surface properties, mechanical properties, and in vitro cell responses using a range of cell types for both hard and soft tissue engineering. The diameters of fibers produced ranged from 0.9 to 1.2 μm. The PHA sheath fibers increased cell attachment of live viable cells compared to PLA sheath fibers, due to their relatively higher biocompatibility, increased surface roughness, and protein adsorption. PLA, as a core, increased the stiffness and tensile strength of the fibers. This approach produced fibers in a cost-effective, scalable, and sustainable manner, providing a highly effective solution to the challenging clinical need of the regeneration of tissues.
Mining and smelting, combined with high-geochemical-backgrounds, cause heavy metal accumulation in soil-crop systems, endangering food-chain health. Topsoil (0–20 cm) and three local crops were sampled from Yunnan’s Huize lead-zinc mining area. Pb, Zn, Cu, Cd contamination, spatial patterns, source contributions and risks were assessed via ArcGIS interpolation, PN, RI, APCS-MLR, PMF, and USEPA health risk models. All soils exceeded Yunnan background values (100% exceedance), presenting extreme ecological risks mainly driven by Cd (92.94%). Cd has the highest bioavailability, while Pb existed mostly potentially bioavailable. Leafy crops show 100% over-standard rates for Pb and Cd, maize exceeds Pb limits. Four pollution sources were identified. Ingestion was the primary exposure pathway. Children suffered prominent non-carcinogenic Pb risks, while all populations encountered unacceptable Cd carcinogenic risks. Mismatches between soil metal contents and crop safety reveals uniform soil-standard limitations, offering references for pollution control, crop-layout optimization and differentiated risk management in analogous mining areas.
Conservation of rare and endangered species requires scalable monitoring solutions to address accelerating biodiversity loss. Deep learning has transformed wildlife monitoring, enabling automated species identification, individual tracking, demographic inference, and behavioral analysis from camera traps, drones, and video surveillance. This review systematically analyzes 159 wildlife monitoring studies (2014–2026) across four technical domains: species recognition, individual re-identification, soft biometrics (age/sex estimation), and behavior/pose analysis. We examine architectural evolution across three architectural paradigms—CNN-based models, Attention & Sequence Architectures, and Foundation Models- documenting performance advances and ecological applications. Key challenges—environmental interference, data scarcity, domain shift, computational constraints—and optimization strategies, including data augmentation, transfer learning, and model compression, are systematically addressed. Case studies demonstrate real-world impact across diverse taxa: giant pandas, Amur tigers, great apes, and marine mammals. We identify critical research gaps and emerging opportunities in foundation model adaptation, multimodal learning, spatiotemporal modeling, and edge deployment. This review bridges computer vision research and conservation practice, providing actionable insights for developing robust, deployable monitoring systems that translate algorithmic advances into conservation impact.
Polycystic ovary syndrome (PCOS) is a prevalent endocrine disorder with complex metabolic and reproductive manifestations. Gut microbiota dysbiosis has emerged as a key factor in PCOS, yet the causal role of microbiota-targeted interventions remains to be fully defined. In a letrozole-induced rat model, we show that gypenosides (GPs)—bioactive saponins from Gynostemma pentaphyllum—alleviate weight gain, hormonal imbalance, and estrous cycle disruption, alongside reducing systemic oxidative stress and inflammation. Integrated ovarian transcriptomics and gut microbiome 16S rRNA sequencing reveal that GPs modulate ovarian gene expression related to inflammation and cellular function while reshaping gut microbiota by enriching beneficial genera including Lactobacillus and Romboutsia. Fecal microbiota transplantation (FMT) establishes causality: PCOS-derived microbiota transfers disease traits to normal recipients, whereas GPs-conditioned microbiota alleviates PCOS phenotypes in recipients. These findings demonstrate that GPs act through a “gut microbiota-oxidative stress-ovary” axis, supporting GPs as a promising microbiota-targeted intervention for PCOS management.
