Advancing Open Science

Background/Objectives: Accurate prediction of complex traits from high-dimensional genomic markers remains difficult in plant breeding. Linear models have limited capacity for non-linear interactions, and many deep learning models scale poorly on long marker sequences. This study introduces BiMSGP, a bidirectional Mamba-based model for scalable genomic prediction. Methods: BiMSGP extends a selective state space model with forward and backward processing so that each marker can integrate sequence context from both directions while retaining linear computational complexity; this is a modeling device for long-range marker dependencies, not a demonstration of biological epistasis. Regression was evaluated on Wheat599, Wheat2000, and SoyBase, and classification on SoyBase, using 10-fold cross-validation with shared fold assignments and no inner validation split; the held-out fold was used for checkpoint selection, learning-rate adjustment, and reporting. Ablation studies compared bidirectional and unidirectional stacks and different depths. Exploratory paired tests of fold-wise PCC differences were used to distinguish numerical rank from a paired difference under this protocol. Results: Under the reported evaluation protocol, BiMSGP attained the numerically highest mean Pearson correlation coefficient (PCC) in all four Wheat599 environments, six of eight Wheat2000 traits, and five of seven quantitative SoyBase traits. Several of these numerical ranks did not have exploratory $p_t<0.05$ against the strongest baseline; GBLUP had a higher mean PCC for SoyBase oil ($p_t=0.025$). Shallow bidirectional designs had higher mean PCC than deeper and unidirectional variants on Wheat599 Environment 1. Training cost was higher than that of lightweight baselines on larger datasets, but inference remained in the millisecond range. Conclusions: Under this protocol, bidirectional selective state space modeling yielded competitive mean scores with a higher training cost than lightweight baselines. The scores are exploratory and may be optimistic; they are not unbiased estimates of generalization. Evaluation that separates model selection from assessment, and tests in independent populations, years, or environments, remains future work.

Skatole (3-methylindole) is a primary malodorous compound emitted from livestock manure, posing significant challenges for odor control in intensive animal production. In this study, a skatole-degrading bacterial strain, designated hhy901, was isolated from fresh cattle manure and identified as Bacillus pumilus through polyphasic taxonomy, including 16S rRNA and gyrB gene phylogeny, morphological characterization, biochemical profiling, and whole-cell fatty acid analysis. Under an initial skatole concentration of 100 mg/L in Bacillus liquid medium, strain hhy901 removed 87.2% of skatole within 96 h, as determined by high-performance liquid chromatography (HPLC). Whole-genome sequencing revealed a genome of approximately 3.60 Mb with a GC content of 41.91%, encoding 3564 predicted protein-coding genes. Functional genomic annotation identified 39 genes associated with xenobiotic biodegradation and metabolism, including putative catechol dioxygenase, aldehyde dehydrogenase, and cytochrome P450 genes, which together suggest a potential skatole catabolic pathway. Furthermore, the genome harbored a broad repertoire of genes associated with resistance to heavy metals, disinfectants, and multiple antibiotic classes, indicating a potential for environmental stress adaptation. Virulence factor profiling predominantly identified genes related to nutrient acquisition and microbial competition, with only a single invasion-associated factor detected, suggesting a relatively low invasive pathogenic potential. However, the presence of multiple antimicrobial resistance determinants warrants cautious interpretation, and phenotypic validation of both degradation function and biosafety is required before environmental application. These results establish B. pumilus hhy901 as a promising microbial candidate for further development of bio-deodorization strategies in livestock production, pending functional and safety validation.

UAV multispectral imagery can support rice yield estimation, but spectral–yield relationships may change with cultivar and planting density. We evaluated three japonica rice cultivars under five planting-density treatments in a split-plot experiment and collected UAV multispectral imagery on 13 dates. We examined five vegetation indices: NDVI, RVI, NDRE, OSAVI, and CLRE. Cultivar significantly affected grain yield and yield components. Planting density had no significant main effect, while the cultivar × density interaction was significant for grain yield. Spectral–yield correlations were strongest during 19–25 August, which broadly matched reproductive development based on field observations. NDRE and CLRE showed the most consistent associations. A four-index PLSR model using 19 August data achieved an outer-LOOCV R² of 0.773 and an RMSE of 530.28 kg ha⁻¹. Tuned RF gave comparable performance (R² = 0.783, RMSE = 518.60 kg ha⁻¹). Adjustment for cultivar, planting density, and block substantially reduced the pooled mid-August correlations. However, the four-index spectral set still improved prediction beyond cultivar, planting density, and their interaction. Grouped validation changed little when density levels or cultivar × density combinations were withheld, but performance declined sharply when an entire cultivar was excluded. These results show that mid-August UAV observations were informative in this experiment, but wider use will require validation across more cultivars, years, and sites.

The facial muscles are crucial for nonverbal communication, physiological functions, and articulation. Unlike skeletal or masticatory muscles, they lack fascia, attach directly to the skin and function without joint-mediated movements or visual feedback. The aim of this study was to reconstruct the facial musculature using high-resolution computerised tomography (CT) and 3 Tesla magnetic resonance imaging (MRI) on 14 body donors, yielding segmented 3D datasets of the facial muscles, as well as bony and soft-tissue head structures. We hypothesise that the facial musculature comprises functional subunits with variable geometries that extend beyond classical anatomical definitions, with attachment points and three-dimensional positioning that differ from conventional descriptions. The most significant finding was that MRI could not resolve individual facial muscles as depicted in anatomical atlases. Instead, the musculature appeared as a single continuous sheet of muscle, embedded in fatty tissue and not attached to the bone. Whether the facial musculature is in fact continuous or consists of several muscles connected by connective tissue remains unclear. The difficulty of segmentation varied depending on the region; the zygomatic muscle was the easiest to separate, whilst the orbicularis oculi, frontalis and platysma muscles proved particularly difficult due to their thin, flat morphology.

The increasing use of generative artificial intelligence (GenAI) in university students’ academic writing is challenging established understandings of authorship, academic integrity, and learning. The study investigated how undergraduates negotiate authorship, responsibility and learning within these emerging writing practices. The study used a qualitatively driven mixed-methods design with undergraduate students enrolled in two sections of an academic writing course at an English-medium university in Cyprus. Data comprised 55 written reflections on students’ use of GenAI while completing a research paper and 27 responses to an anonymous post-course questionnaire. Reflections were analysed using thematic analysis, with questionnaire data providing additional descriptive and contextual evidence. Findings indicate that students positioned GenAI as legitimate support for brainstorming, organising ideas, language refinement and feedback while resisting its use as a substitute for intellectual engagement. Students constructed authorship through human judgement, locating ownership in verification, evaluation, decision-making, and accountability rather than sole textual production. Ethical GenAI use emerged as boundary work distinguishing assistance from substitution, while meaningful learning was associated with maintaining cognitive ownership of the thinking underpinning academic work. These findings suggest that GenAI can be integrated into academic writing while students continue to locate authorship, ethical responsibility, and meaningful learning in human judgement and intellectual engagement.

Climate change is expected to reshape the climatic suitability of Neotropical plant taxa, particularly in topographically complex regions such as the tropical Andes, yet no nationwide assessment exists for the genus Clusia L. (Clusiaceae) in Peru. We modeled its current and future potential distribution using MaxEnt, based on 245 spatially thinned occurrence records (GBIF) and bioclimatic variables from WorldClim 2.1 selected through Pearson correlation and variance inflation factor filtering. Model complexity was tuned with ENMeval using spatial block cross-validation, and future projections were generated under the SSP2-4.5 and SSP5-8.5 scenarios for 2041–2060, 2061–2080, and 2081–2100 across three Global Climate Models. The model showed acceptable, stable performance (training AUC = 0.824, test AUC = 0.825), with elevation and precipitation seasonality (Bio15) as the leading predictors, together with Bio19 and Bio3 accounting for over 80% of the total contribution. Currently, only 16.61% of Peru is climatically suitable (medium–high) for the genus, concentrated discontinuously along Andean slopes and inter-Andean valleys. Both future scenarios project progressive contraction and fragmentation of suitable habitat, most severe under SSP5-8.5, where high-loss areas reach 50.43% of the evaluated territory by 2081–2100; these projections should be interpreted as plausible scenarios rather than validated forecasts, given the absence of independent data for future periods. Habitat gains were restricted to higher elevations, consistent with an altitudinal range shift. These results provide a coarse-scale spatial baseline that highlights broad patterns of climatic vulnerability for the genus, intended to guide exploratory conservation attention and future species-level research in the Andean region, rather than site-specific management decisions.

Oregano essential oil (OEO) was tested as a natural botanical anesthetic for spotted sea bass (Lateolabrax maculatus) through a comprehensive evaluation of anesthetic efficacy, physiological homeostasis, tissue integrity, flesh texture and carvacrol residual dynamics. Gradient treatments (100–300 µL/L) revealed dose-dependent anesthetic performance, and 200 µL/L was validated as the optimal dosage, under which fish reached deep anesthesia within 3 min and regained normal swimming within 5 min, with zero mortality recorded 12 h post-anesthesia. Temporary stress and oxidative stress fluctuations occurred in serum biochemistry and antioxidant enzyme profiles, which approached control levels after 12 h depuration. Only slight histological changes appeared in liver and gill tissues without permanent pathological damage. Minor declines in partial texture attributes recovered completely within 12 h, whereas other muscle quality indicators remained unaffected. Carvacrol was detected in muscle immediately after OEO anesthesia but was below the detection limit at all subsequent sampling times. Functioning as an efficient natural anesthetic with minimal physiological disturbance, OEO shows potential as an alternative to synthetic anesthetics for spotted sea bass live transportation and delivers dependable technical guidance for aquaculture production.

Spontaneous imbibition is an important process in enhanced oil recovery, yet its accurate prediction in nanoscale pores remains a significant challenge due to the breakdown of continuum assumptions at the nanoscale. The classical Lucas-Washburn (L-W) equation, which suggests that imbibition length is proportional to the square root of time, fails to capture the complex early-time dynamics in nanopores. Therefore, in this work, coarse-grained molecular dynamics simulations were employed to systematically investigate two-phase imbibition across a range of pore sizes (1.85–7 nm), fluid types (C4–C20 alkanes), and wettability conditions. The results reveal a distinct transition from continuum-dominated to molecularly dominated flow as pore size decreases, with anomalous kinetics that originate from molecular-scale interactions in sub-2-nm pores. A refined imbibition model is proposed that explicitly incorporates an entrance resistance term in addition to the capillary and viscous forces. This new model successfully resolves the overestimated initial velocity by the L-W equation, revealing a dual-regime physics, including an early-time, almost linear regime influenced by entrance resistance, followed by the classical late-time L-W regime. The model demonstrates excellent agreement with molecular dynamics data for mesoscopic nanopores (r ≥ 3.5 nm), validating its physical basis. Although the proposed model is still not predictive due to the implicit inclusion of imbibition rate in the derivision, the findings provide a robust theoretical foundation for describing imbibition in nanopores and highlight the critical need for molecular-scale approaches in extreme nanoconfinement.