Search Results (1,680)

To address the limitations of the traditional Balancing Composite Motion Optimization (BCMO) algorithm—namely weak directional global exploration, insufficient local exploitation accuracy, and a tendency to fall into local optima with reduced population diversity in feature selection tasks—this paper proposes a Multi-Strategy Enhanced Balancing Composite Motion Optimization algorithm (MEBCMO). From a symmetry perspective, MEBCMO exploits the symmetric and asymmetric relationships among candidate solutions in the search space to achieve a better balance between exploration and exploitation. The performance of MEBCMO is enhanced through three complementary strategies. First, an adaptive heat-conduction search mechanism is introduced to simulate thermal transmission behavior, where a Sigmoid function adjusts the heat-conduction coefficient α_T from 0.9 to 0.2 during iterations. By utilizing the symmetric fitness–distance relationship between the current solution and the global best, this mechanism improves the directionality and efficiency of global exploration. Second, a quadratic interpolation search strategy is designed. By constructing a quadratic model based on the current individual, a randomly selected individual, and the global best, the algorithm exploits local symmetric characteristics of the fitness landscape to strengthen local exploitation and alleviate performance degradation in high-dimensional spaces. Third, an elite population genetic strategy is incorporated, in which the top three individuals generate new candidates through symmetric linear combinations with non-elite individuals and Gaussian perturbations, preserving population diversity and preventing premature convergence. To evaluate MEBCMO, extensive global optimization experiments are conducted on the CEC2017 benchmark suite with dimensions of 30, 50, and 100, and comparisons are made with eight mainstream algorithms, including PSO, DE, and GWO. Experimental results demonstrate that MEBCMO achieves superior performance across unimodal, multimodal, hybrid, and composite functions. Furthermore, MEBCMO is combined with LightGBM to form the MEBCMO-LightGBM model for feature selection on 14 public datasets, yielding lower fitness values, higher classification accuracy, and fewer selected features. Statistical tests and convergence analyses confirm the effectiveness, stability, and rapid convergence of MEBCMO in symmetric and complex optimization landscapes. Full article

The Xinjiang Brown cattle (XJBC) is one of China’s five major dual-purpose dairy and beef breeds. Analyzing the genetic diversity of the Xinjiang Brown cattle population lays the theoretical groundwork for identifying and conserving its genetic resources. This study employed the Illumina Bovine SNP 150K chip to analyze genetic diversity, inbreeding coefficient, kinship, and genetic distance in a population of 750 Xinjiang Brown cattle from three breeding farms in Xinjiang. Genetic diversity was assessed by calculating minimum allele frequency (MAF), observed heterozygosity (Ho), expected heterozygosity (He), polymorphic information content (PIC), and linkage disequilibrium (LD). Population structure was analyzed using PCA. ROH was calculated to derive ROH-based inbreeding coefficients, pedigree-based inbreeding coefficients (F_PED) were estimated using CFC software for comparison, and candidate genes within high-frequency ROH regions in Xinjiang Brown cattle were identified. A G matrix was constructed to analyze population kinship. Results revealed 94,173 high-quality SNP loci in Xinjiang Brown cattle, with an average MAF of 0.276, PIC of 0.376, Ho of 0.345, and He of 0.376. Breeding farm 3 exhibited the fastest LD decay, indicating relatively high genetic diversity across Xinjiang Brown cattle populations, with farm 3 demonstrating greater diversity. The IBS genetic distance was 0.313. The G matrix results aligned with the IBS distance matrix, both indicating close kinship among some individuals within the Xinjiang Brown cattle population. The ranges for average F_PED and average F_ROH across farms were 0.0017–0.0189 and 0.0609–0.0878, respectively. Short ROH segments (0.5–2 Mb) constituted the largest proportion (51.31%) of all ROHs. Within high-frequency ROH enrichment regions, 61 genes, including LCORL, FAM110B, NR4A1, and PER2, were identified as potentially associated with economic traits in Xinjiang Brown cattle. These findings provide relevant marker sites for genomic selection in Xinjiang Brown cattle and lay a theoretical foundation for subsequent research. Full article

The integration of edge and cloud computing is critical for resource-intensive applications which require low-latency communication, high reliability, and efficient resource utilisation. The service placement problem in these environments poses significant challenges owing to dynamic network conditions, heterogeneous resource availability, and the necessity for real-time decision-making. Because determining an optimal service placement in such networks is an NP-complete problem, the existing solutions rely on fast but suboptimal heuristics or computationally intensive metaheuristics. Neither approach meets the real-time demands of online scenarios, owing to its inefficiency or high computational overhead. In this study, we propose a lightweight learning-based approach for the online placement of services with multi-version components in edge-to-cloud computing. The proposed approach utilises a Shallow Neural Network (SNN) with both weight and power coefficients optimised using a Genetic Algorithm (GA). The use of an SNN ensures low computational overhead during the training phase and almost instant inference when deployed, making it well suited for real-time and online service placement in edge-to-cloud environments where rapid decision-making is crucial. The proposed method (SNN-GA) is specifically evaluated in AR/VR-based remote repair and maintenance scenarios, developed in collaboration with our industrial partner, and demonstrated robust performance and scalability across a wide range of problem sizes. The experimental results show that SNN-GA reduces the service response time by up to 27% compared to metaheuristics and 55% compared to heuristics at larger scales. It also achieves over 95% platform reliability, outperforming heuristics (which remain below 85%) and metaheuristics (which decrease to 90% at larger scales). Full article

Callus induction is a complex procedure in plant organ, cell, and tissue culture that underpins processes such as metabolite production, regeneration, and genetic transformation. It is important to monitor callus formation alongside subjective evaluations, which require labor-intensive care. In this research, the first curated lentil (Lens culinaris) callus dataset for instance segmentation was experimentally generated using three genotypes as one data set: Firat-87, Cagil, and Tigris. Leaf explants were cultured on MS medium fortified with different concentrations of gross regulators of BA and NAA to induce callus formation. Three biologically relevant stages, the leaf stage, the green callus, and the necrosis callus, were produced. During this process, 122 high-resolution images were obtained, resulting in 1185 total annotations across them. The dataset was evaluated across four successive generations (v5/7/8/11) of YOLO deep learning models under identical conditions using $mAP$, $Dice$ coefficient, $Precision$, $Recall$, and $IoU$, together with efficiency metrics including parameter counts, FLOPs, and inference speed. The results show that anchor-based variants (YOLOv5/7) relied on predefined priors and showed limited boundary precision, whereas anchor-free designs (YOLOv8/11) used decoupled heads and direct center/boundary regression that provided clear advantages for callus structures. YOLOv8 reached the highest instance segmentation precision with mAP50@0.855, while it matched the accuracy with greater efficiency and achieved real-time inference with 166 FPS. Full article

Conservation programmes for two local dairy cattle breeds—Latvian Brown old type (BV) and Latvian Blue (LZ)—commenced in 2004. The aim of this study was to evaluate genetic diversity in the BV and LZ local cattle populations using SNP data. This study was based on genotype data from 96 BV and 75 LZ cows and 20 BV and 18 LZ bulls. The SNPs were determined using the GGP 100K bovine SNP BeadChip. Quality control (QC) and genotype data analysis were performed using PLINK v1.9. The observed heterozygosity was moderate, at around 0.4, for both breeds. Inbreeding coefficients were estimated based on homozygosity runs (F_ROH) to compare recent and ancient inbreeding in the BV and LZ populations. Therefore, the ROH segments were divided into segments with the four classes (1–4 Mb, 4–8 Mb, 8–16 Mb, and above 16 Mb). Shorter ROH regions (ROH < 4 Mb) predominated in the genome. ROH regions with lengths above 16 Mb covers 4–6% of the genome in BV and 11% in LZ population. The average inbreeding coefficient for approximately three generations (F_ROH>16) was 2.30% and 4.87% for BV and LZ cows (p < 0.05), respectively, and 2.59% and 3.85% for BV and LZ bulls, respectively. This study demonstrates that inbreeding has increased from generation to generation (F_ROH>16 is higher compared with F_ROH<16) in both populations. The level of current inbreeding in LZ is higher compared with that in the BV breed. The overall level of inbreeding in the BV and LZ populations is low, but there is a high level of inbreeding among a few animals. The impact of inbreeding on cow productivity has been observed in the LZ and BV cow populations. As a result, breeding organisations need to monitor and control the level of inbreeding and prevent the loss of genetic diversity in these animal populations. Breeders should minimize mating among close relatives; introduce genetically unrelated animals, use pedigree, and genomic information in controlling rates of inbreeding. Full article

(1) Background: In the Sudano-Sahelian zone of Cameroon, which is affected by drought and forest decline, Vachellia nilotica leaves and seeds are fodder for livestock. (2) Methods: A provenance and progeny study on growth performance and heritability of V. nilotica was carried out to provide a reliable database for tree selection, improvement programs, and the creation of future forested areas in this region. Open-pollinated seeds from 120 mother trees (10 half-sib families per provenance) representing twelve provenances, 50–100 km apart, were used for a progeny trial near Maroua, the Far North region of Cameroon. The experimental design was a Fisher block. (3) Results: The results reveal significant differences among provenances only for the number of leaves, and the variability was marked by coefficients of variation ranging from 0.24−0.63. Narrow-sense heritability was measured, varying from 0.01 ± 0.009 to 0.74 ± 0.02, and genetic gain reached 21.83 at the selection intensity of 5% for the number of leaves per plant. The phenotypic coefficient of variation varied between 14% and 90%. Half-sib families were classified into three subgroups using hierarchical ascending classification, and provenances were grouped into five groups using principal component analysis. (4) Conclusions: These results could contribute to initiating tree selection, but more provenances, longer-term experiments, and molecular genetic testing are needed to complement these nursery-level observations. Full article

Integrating perennial grain crops into agricultural systems can become a key milestone for increasing the provision of ecosystem services of food production systems. Intermediate wheatgrass is a novel perennial grain and forage crop that is undergoing domestication. Potential trade-offs between resource allocation and reproductive and vegetative plant structures can challenge the response to selection for both grain and forage production under dual-purpose use. Our goal was to understand the genetic relationship between grain and forage yield components, quantify potential trade-offs between vegetative and reproductive allocation, and optimize the response to selection under dual-purpose management. Phenological, grain, and forage traits were evaluated in 30 half-sib families across two field experiments conducted over three years. No trade-offs were detected between grain and forage yield traits, indicating that the simultaneous improvement of both traits is feasible. Grain yield per spike and spikes per plant are promising secondary traits for indirect selection, given their moderate-to-high heritability (h² = 0.58 and 0.41) and strong Pearson correlation coefficients with grain yield per plant (0.68 and 0.82). These traits could be assessed in the first year, increasing genetic gain per unit time. Intermediate wheatgrass germplasm could therefore be efficiently developed by shortening the time to first evaluation, using secondary traits, and performing selection under dual-purpose management. Full article

Rumex crispus L. is a globally distributed invasive species that has naturalized in South Korea, where its use as a medicinal, edible, and ecological restoration resource continues to expand. However, its genetic background remains insufficiently understood, underscoring the need for species-specific molecular markers to enable accurate assessments of intraspecific genetic diversity and population structure. Using 19 newly developed microsatellite markers, we analyzed 120 plants from 6 populations in the riparian zone. A total of 166 alleles were detected, with a mean polymorphism information content of 0.637. Across the six populations, genetic diversity analysis showed mean observed (Ho = 0.304) and expected (He = 0.588) heterozygosity values indicative of heterozygote deficiency (inbreeding coefficient F_IS = 0.456–0.559). Genetic differentiation was low in AMOVA (10%) and F_ST (0.048–0.120) but higher in Jost’s D (0.096–0.342). STRUCTURE analysis identified two major genetic clusters (ΔK = 2), and spatial Bayesian clustering revealed six distinct genetic units (K = 6), suggesting that partial barriers to gene flow may have influenced population structure. These findings provide essential genetic insights that can support the effective control of R. crispus spread and its potential use as a valuable plant resource. Full article

Cherimoya is a fruit tree native to the Andean regions of South America, also in Central America, prized for its flavor, nutritional properties, and medicinal potential. Despite its economic relevance, in situ assessments of phenotypic diversity are limited, particularly in southern Ecuador, a key center of domestication. This study evaluated the morphological and ecogeographic diversity of 270 native trees across eight cantons in Loja province, Ecuador, using 34 qualitative and quantitative descriptors of leaves, flowers, fruits, and seeds. High phenotypic variability was observed, with coefficients of variation exceeding 40% for key traits, including mature fruit weight (48.15%), pulp weight (55.33%) and pulp-to-seed ratio (64.23%). Principal component analysis revealed three major axes of variation associated with productivity, floral morphology, and organoleptic quality. Cluster analysis identified four groups, with one distinguished by a favorable pulp-to-seed ratio and sugar–acid content. Species distribution modeling, which included bioclimatic and soil variables, showed that Gonzanamá, Quilanga and Espíndola possess the highest ecological suitability for cherimoya. These findings highlight priority areas for in situ conservation and phenotype selection, providing a foundation for sustainable use, genetic improvement, and the preservation of locally adapted germplasm to support climate-resilient agricultural systems. Full article

Low temperature exerts severe adverse effects on maize growth, particularly during the seedling stage. Screening for cold-tolerant maize genotypes is highly significant for identifying genes associated with cold tolerance and enhancing maize performance under low, suboptimal temperature conditions. The identification of representative cold tolerance-related genes is of great significance for the breeding of cold-resistant maize varieties. In this study, a diversity panel of 205 materials was evaluated and classified for cold tolerance at the seedling stage. The coefficients of variation of all materials ranged from 14.53% to 35.71%, reflecting considerable genetic diversity within the panel. The correlation coefficients for each phenotypic trait between the cold-treated (CT) and control (CK) maize materials ranged from 0.60 to 0.90, further indicating that all traits displayed varying degrees of sensitivity to cold stress. A comprehensive evaluation of cold tolerance using the D value was conducted. The D values of all materials ranged from 0.355 to 0.863, with a mean value of 0.64. A hierarchical clustering analysis was performed to classify all materials into five categories based on their cold tolerance. Further, 17 SNPs were identified using GWAS analysis, and 12 candidate genes were located within the regions related to the SNPs. Some candidate genes were closely associated with cold tolerance, such as genes encoding MYB and GRAS transcription factors, leucine-rich repeat (LRR) proteins, and protein kinases. Validation by qRT-PCR confirmed that the expression of some genes was induced under cold stress conditions. These findings lay a crucial foundation for breeding cold-tolerant maize varieties and for further exploration of genes associated with cold tolerance. Full article

Walnut (Juglans regia L.), a critical economic species, experiences substantial declines in fruit quality and yield due to Walnut Leaf Scorch (WLS). This issue is particularly severe in the Xinjiang Uygur Autonomous Region (XUAR)—one of Asia’s leading walnut-producing regions. To mitigate the disease, timely and efficient monitoring approaches for detecting infected trees and quantifying their disease severity are in urgent demand. In this study, we explored the feasibility of developing a predictive model for the precise quantification of WLS severity. First, five 4-mu (1 mu = 0.067 ha) sample plots were established to identify infected individual trees, from which the WLS Disease Index (DI) was calculated for each tree. Concurrently, hyperspectral data of individual trees were acquired via an unmanned aerial vehicle (UAV) platform. Second, DI estimation models were developed based on the Random Forest (RF) and Support Vector Machine (SVM) algorithms, with each algorithm optimized using either Grid Search (GS) or a Genetic Algorithm (GA). Finally, four integrated models (GS-RF, GA-RF, GS-SVM, and GA-SVM) were constructed and systematically compared. The results showed that the Genetic Algorithm-optimized SVM model (GA-SVM) exhibited the highest predictive accuracy and robustness, achieving a coefficient of determination (R²) of 0.6302, a Root Mean Square Error (RMSE) of 0.0629, and a Mean Absolute Error (MAE) of 0.0480. Our findings demonstrate the great potential of integrating UAV-based hyperspectral remote sensing with optimized machine learning algorithms for WLS monitoring, thus offering a novel technical approach for the macroscopic, rapid, and non-destructive surveillance of this disease. Full article

The average pairwise dissimilarity (APD) between one plant sample and other assayed samples based on genetic markers was developed in 2006 to assess genetic distinctness and genetic redundancy in a plant germplasm collection. With the availability of abundant genomic variants across a genome, APD can be expanded to measure individual genomic distinctness. This study was conducted to assess the applicability of APD estimates in measuring the individual genomic distinctness of 1789 indica and 854 japonica rice samples based on published genome-wide single-nucleotide polymorphism (SNP) and structural variant (SV) data. It was found that the acquired APD estimates were weakly or not correlated between the SNP and SV data sets in the indica or japonica samples, respectively. For the indica samples, the APD estimates based on the SNP and SV data ranged from 0.1779 to 0.3277 and from 0.2297 to 0.4096, respectively. For the japonica samples, the SNP-based and SV-based APD estimates varied from 0.1774 to 0.3029 and from 0.1534 to 0.3459, respectively. These APD estimates were highly negatively correlated with the estimates of individual inbreeding coefficients and can identify the most genomically distinct rice germplasm that are compatible with those revealed through principal component analysis. Also, a reliable APD estimation was found to require 5000 to 10,000 random genomic SNPs or SVs. These findings together are significant, not only in demonstrating the informativeness of APD estimates in the identification of individuals with variable genomic distinctness, but also in providing guidance for APD applications to measure individual genomic distinctness. Full article

Developing reliable in vitro propagation methods for rowanberry genotypes is essential for their use in breeding and horticultural programs. While different rowanberry species and cultivars are primarily valued for their ornamental and forestry properties, poor seed germination and the low success rate of conventional vegetative techniques constrain their propagation. Micropropagation offers a practical approach to obtaining uniform, disease-free plant material for selection, hybridization, and the subsequent horticultural use of particular valuable genotypes. Shoot multiplication of a prospective sweet rowanberry cultivar ‘Discolor’ was studied on a Murashige and Skoog (MS) medium supplemented with 6-benzylaminopurine (BAP) at concentrations of 1, 2, and 4 mg L⁻¹, thidiazuron (TDZ) at concentrations of 0.5 and 1 mg L⁻¹, and 6-(γ,γ-dimethylallylamino)purine (2iP) at a concentration of 10 mg L⁻¹. Root induction was evaluated on a half-strength MS medium (50% MS) supplemented with 1 mg L⁻¹ of naphthaleneacetic acid (NAA), indole-3-butyric acid (IBA), or indole-3-acetic acid (IAA). TDZ at 1 mg L⁻¹ yielded the highest multiplication coefficient. However, media with TDZ at a lower concentration (0.5 mg L⁻¹) or BAP (2–4 mg L⁻¹) provided the best balance between proliferation rate and shoot quality. These media promoted the growth of vigorous, well-elongated shoots with minimal callus formation. In contrast, the phytohormone 2iP did not elicit physiological response in the in vitro multiplication of explants. The best rooting results were obtained using a 50% MS medium supplemented with 1 mg L⁻¹ IAA, which provided the highest rooting percentage and root quality. IBA produced slightly lower, though comparable, results, while NAA resulted in weak, sporadic root formation. The established protocol enables the efficient in vitro propagation of the studied cultivar. This system supports its application in breeding and fruit production programs, as well as in maintaining valuable genetic resources within the genus Sorbus. Full article

Eucalyptus urophylla is a core tree species for short-rotation industrial timber plantations in South and Southwest China. However, the dynamic correlation rules of its growth traits during the full rotation period remain unclear, and the theoretical research on early selection is insufficient. In this study, 12 pure E. urophylla clones (including U6 and MLA as controls) were used as plant materials. Based on the data of tree height (H), diameter at breast height (DBH, D), and individual tree volume (V) from 0.5 to 7.5 years old, the correlation rules of early and late growth traits were explored, core predictive traits were screened, and the optimal selection age was determined through rank correlation, phenotypic and genetic correlation analyses, combined with regression modeling and selection efficiency calculation. Early selection of E. urophylla clones was feasible: after 3.5 years, the early–late phenotypic and genetic correlation coefficients of H, D, and V all reached significant or highly significant levels, and the genetic correlation coefficients were greater than the phenotypic ones, indicating that genetic factors dominated trait correlations with little environmental interference. All five established early selection regression models passed the highly significant test. Among them, the models of D-early versus D-late, V-early versus V-late, and D-early versus V-late had the highest coefficients of determination (0.9293–0.9385), making them the optimal selection traits; the models of H-early versus H-late and H-early versus V-late had lower coefficients of determination (0.8010–0.8364) due to errors in height measurement. The best selection effect was achieved within 1/2–2/3 of the rotation period: for a 6-year rotation period (pulpwood), the optimal selection age was 3.5 years old (annual efficiency 1.318); for an 8-year rotation period (medium-diameter timber), it was 4.5 years old (annual efficiency 1.345); and for a 12-year rotation period (large-diameter timber), it was 6.5 years old (annual efficiency 1.379). This study not only fills the theoretical gap in early selection of E. urophylla during the full rotation period but also constructs an integrated early selection technology system of “trait screening—model prediction—age determination”. It provides key support for shortening the breeding cycle of E. urophylla and achieving precise control of breeding costs and offers important references for early selection research on fast-growing broad-leaved tree species worldwide. Full article

Accurate EMA assessment in live sheep is crucial for genetic breeding and production management within the meat sheep industry. However, the segmentation accuracy and reliability of existing automated methods are limited by challenges inherent to B-mode ultrasound images, such as low contrast and noise interference. To address these challenges, we present MPG-SwinUMamba, a novel deep learning-based segmentation network. This model uniquely combines the state-space model with a U-Net architecture. It also integrates an edge-enhancement multi-scale attention module (MSEE) and a pyramid attention refinement module (PARM) to improve the detection of indistinct boundaries and better capture global context. The global context aggregation decoder (GCAD) is employed to precisely reconstruct the segmentation mask, enabling automated measurement of the EMA. Compared to 12 other leading segmentation models, MPG-SwinUMamba achieved superior performance, with an intersection-over-union of 91.62% and a Dice similarity coefficient of 95.54%. Additionally, automated measurements show excellent agreement with expert manual assessments (correlation coefficient r = 0.9637), with a mean absolute percentage error of only 4.05%. This method offers non-invasive and efficient and objective evaluation of carcass performance in live sheep, with the potential to reduce measurement costs and enhance breeding efficiency. Full article