Search

Although deep-learning methods can achieve human-level performance in boundary detection, their improvements mostly rely on larger models and specific datasets, leading to significant computational power consumption. As a fundamental low-level vision task, a single model with fewer parameters to achieve cross-dataset boundary detection merits further investigation. In this study, a lightweight universal boundary detection method was developed based on convolution and a transformer. The network is called a "transformer with difference convolutional network" (TDCN), which implies the introduction of a difference convolutional network rather than a pure transformer. The TDCN structure consists of three parts: convolution, transformer, and head function. First, a convolution network fused with edge operators is used to extract multiscale difference features. These pixel difference features are then fed to the hierarchical transformer as tokens. Considering the intrinsic characteristics of the boundary detection task, a new boundary-aware self-attention structure was designed in the transformer to provide inductive bias. By incorporating the proposed attention loss function, it introduces the direction of the boundary as strongly supervised information to improve the detection ability of the model. Finally, several head functions with multiscale feature inputs were trained using a bidirectional additive strategy. In the experiments, the proposed method achieved competitive performance on multiple public datasets with fewer model parameters. A single model was obtained to realize universal prediction even for different datasets without retraining, demonstrating the effectiveness of the method. The code is available at https://github.com/neulmc/TDCN. [ABSTRACT FROM AUTHOR]

Simple Summary: The Mongolian racerunner (Eremias argus) is a genetically diverse species with a wide distribution across Northeast Asia. The present study is the first to assess the genetic variation in E. argus across its entire geographic range using newly generated mtDNA cyt b and previously published data. We show the complex genetic pattern of this species, where most genetic divergence is not associated with geographic regions. This pattern is the result of multiple influences–robust dispersal ability resulting from ecological features, anthropogenic influence of translocation, and past climate change. Geographic distance has primarily shaped the genetic structure of the species, as indicated by a strong pattern of isolation by distance across all populations. The population growth of E. argus from the Last Interglacial to the Last Glacial Maximum is likely a consequence of an increase in favorable habitats with mild climates following the development of monsoons in East Asia since the mid-Late Pleistocene. Overall, the population genetic structure of E. argus was likely driven by a combination of climatic and geographic changes rather than direct ice sheet coverage. Our results highlight the importance of combining genetic approaches with environmental data when assessing the impact of Pleistocene climatic oscillations. The Mongolian racerunner, Eremias argus, is a small lizard endemic to Northeast Asia that can serve as an excellent model for investigating how geography and past climate change have jointly influenced the evolution of biodiversity in this region. To elucidate the processes underlying its diversification and demography, we reconstructed the range-wide phylogeographic pattern and evolutionary trajectory, using phylogenetic, population genetic, landscape genetic, Bayesian phylogeographic reconstruction and ecological niche modeling approaches. Phylogenetic analyses of the mtDNA cyt b gene revealed eight lineages that were unbounded by geographic region. The genetic structure of E. argus was mainly determined by geographic distance. Divergence dating indicated that E. argus and E. brenchleyi diverged during the Mid-Pliocene Warm Period. E. argus was estimated to have coalesced at~0.4351 Ma (Marine Isotope Stage 19). Bayesian phylogeographic diffusion analysis revealed out-of-Inner Mongolia and rapid colonization events from the end of the Last Interglacial to the Last Glacial Maximum, which is consistent with the expanded suitable range of the Last Glacial Maximum. Pre-Last Glacial Maximum growth of population is presented for most lineages of E. argus. The Glacial Maximum contraction model and the previous multiple glacial refugia hypotheses are rejected. This may be due to an increase in the amount of climatically favorable habitats in Northeast Asia. Furthermore, E. argus barbouri most likely represents an invalid taxon. The present study is the first to report a range-wide phylogeography of reptiles over such a large region in Northeast Asia. Our results make a significant contribution towards understanding the biogeography of the entire Northeast Asia. [ABSTRACT FROM AUTHOR]

Leucaena leucocephala and Acacia nilotica are two arboreal legumes with several uses in agriculture, nutritious forage and livelihoods, fence posts, wood production, biofuel production, charcoal, firewood, shading, hedges, windbreaks, and improvement of soil fertility. The seed dormancy is common in Leucaena and Acacia species, and it creates difficulties in seed testing and planting. The aim of this study is to determine the effective methods of sulfuric acid and hot water due to overcoming dormancy in seeds of those species. Most pretreatments significantly (p

Zintl-phase compounds hold immense potential for thermoelectric applications owing to their intrinsically low lattice thermal conductivity (κL). However, numerous 212-Zintl-phase compounds remain largely unexplored due to the challenges in assessing their thermal and electrical transport properties via traditional trial-and-error approaches. Here, we present a gradient boosting regressor (GBR) machine-learning (ML) model to predict and discover 5 unexplored and promising 212-Zintl-phase compounds with κL lower than 2 W (mK)−1 at 300 K. The model demonstrated excellent predictive capability with a coefficient of determination (R2) of 0.988 and root mean square error (RMSE) of 0.083 W (mK)−1 on the test set using tenfold cross-validation. Notably, the top-ranked compound Ba2ZnBi2 exhibited an ultra-low κL of approximately 1 W (mK)−1 at 300 K, substantially lower than those of other types of Zintl-phase compounds like 122-, and 111-types. Our theoretical calculations further validated the ultra-low κL of Ba2ZnBi2, and revealed that it originates from the large three-phonon scattering rates and the low group velocities due to the weak atomic interaction in the system. Therefore, our study demonstrates the power of combining ML and first-principles calculations to rapidly identify promising candidates for thermoelectric applications. [ABSTRACT FROM AUTHOR]

The intraspecific variations of phenotypic traits in the early life stages, such as seed germination and seedling establishment, are important components affecting species adaptation and differentiation. As one of the most common ways in which seeds are maintained, dry storage usually affects these traits, by either increasing or decreasing variation among populations. However, little is known about how the interaction between dry storage and population variation of a species affects the performance and adaptability of early life-cycle traits. In this study, we conducted experiments with seeds from ten populations of Elymus nutans along an altitudinal gradient on the eastern Qinghai-Tibet Plateau. Fresh seeds and seeds stored dry for 6 months were used to quantify the temperature thresholds for germination, determine seedling emergence and survival in two common gardens and examine the correlation between these traits and environmental conditions of population provenance. Dry storage increased germination percentage, germination speed and seedling emergence, and decreased intraspecific variation in germination traits of seeds (reduced by 33.36, 52.05 and 20.45% for T b, θT(50) and σ θ T , respectively). Dry storage had little effect on the intraspecific variation of seedling emergence and survival. In addition, the temperature threshold for germination cannot be used to predict seedling emergence and survival in either common garden, regardless of whether seeds were stored or not. These results indicated that it is feasible to use dry-stored seeds to evaluate and select suitable provenances in ecological restoration, and using dry-stored seeds in ecological restoration projects is expected to achieve better vegetation restoration results than fresh seeds. On the other hand, the status of seeds (fresh vs. dry stored) should be considered in an evaluation of the adaptive value of plant functional traits, especially in the early life stages, otherwise, inconsistent conclusions may be drawn. [ABSTRACT FROM AUTHOR]

Simple Summary: Northwest China is characterized by unique geological and historical dynamics and endemic biota, but the joint influence of geography and past climate changes on the evolutionary history of endemic animals is poorly understood. We used two mtDNA genes (ND4 and Cytb) and ecological niche modeling (ENM) to explore how the Quaternary climatic fluctuations and complex geography in Northwest China have shaped genetic diversity, genetic structure, and demographic history of the Alashan pit viper (Gloydius cognatus). Our results clearly show that the lineage diversification of G. cognatus is related to the expansions of deserts and/or the early Pleistocene integration of the Yellow River. The strong spatial genetic structure fits an isolation-by-distance model. The signature of demographic and range contractions during the last glacial maximum (LGM) is rejected by both mitochondrial evidence and ENM. In addition, the suitable habitat of G. cognatus is predicted to be decreased in the future, suggesting that conservation and management of evolutionary significant units (ESUs) should be a priority. Our findings provide the first insights on the evolutionary history of G. cognatus in arid Northwest China and adjacent areas throughout the Quaternary. The joint impacts of historical geological events and Quaternary climatic oscillations in Northwest China on species evolution have been examined extensively in plant under a phylogeographic perspective. However, animal phylogeographic analyses in this region are still limited. The Alashan pit viper, Gloydius cognatus, occurs primarily in arid Northwest China and adjacent areas. Based on variation at two mtDNA genes (ND4 and Cytb) in 27 individuals representing 24 populations, the spatial genetic structure and demographic history of G. cognatus were examined across its geographic range. Phylogenetic analyses revealed two well-supported allopatric clades (each with two distinct subclades/lineages), distributed across the southern (Qaidam Basin, Lanzhou Basin, and Zoige Basin [S1]; Loess Plateau [S2]) and northern (Ily Basin [N1]; Junggar Basin and Mongolian Plateau [N2]) regions. AMOVA analysis demonstrated that over 76% of the observed genetic variation was related to these lineage splits, indicating substantial genetic differentiation among the four lineages. A strong pattern of isolation-by-distance across the sampling populations suggested that geographic distance principally shaped the genetic structure. The four lineages diverged by 0.9–2.2% for the concatenated data, which were estimated to have coalesced ~1.17 million years ago (Mya), suggesting that the expansions of the Badain Jaran, Tengger, and Mu Us deserts during the Xixiabangma glaciation likely interrupted gene flow and triggered the observed divergence in the southern and northern regions. Subsequently, the early Pleistocene integration of the Yellow River and associated deserts expansion promoted the differentiation of S1 and S2 lineages (~0.9 Mya). Both mitochondrial evidence and ecological niche modeling (ENM) reject the signature of demographic and range contractions during the LGM for G. cognatus. In addition, ENM predicts that the suitable habitat of G. cognatus will contract in the future. As such, the conservation and management of ESUs should be a priority. Our findings provide the first insights on the lineage diversification and population dynamics of the Alashan pit viper in relation to geological history and Pleistocene climatic oscillations in arid Northwest China. [ABSTRACT FROM AUTHOR]