Research on Automatic Recognition of Dairy Cow Daily Behaviors Based on Deep Learning.

Dairy cow behavior carries important health information. Timely and accurate detection of behaviors such as drinking, feeding, lying, and standing is meaningful for monitoring individual cows and herd management. In this study, a model called Res-DenseYOLO is proposed for accurately detecting the individual behavior of dairy cows living in cowsheds. Specifically, a dense module was integrated into the backbone network of YOLOv5 to strengthen feature extraction for actual cowshed environments. A CoordAtt attention mechanism and SioU loss function were added to enhance feature learning and training convergence. Multi-scale detection heads were designed to improve small target detection. The model was trained and tested on 5516 images collected from monitoring videos of a dairy cowshed. The experimental results showed that the performance of Res-DenseYOLO proposed in this paper is better than that of Fast-RCNN, SSD, YOLOv4, YOLOv7, and other detection models in terms of precision, recall, and mAP metrics. Specifically, Res-DenseYOLO achieved 94.7% precision, 91.2% recall, and 96.3% mAP, outperforming the baseline YOLOv5 model by 0.7%, 4.2%, and 3.7%, respectively. This research developed a useful solution for real-time and accurate detection of dairy cow behaviors with video monitoring only, providing valuable behavioral data for animal welfare and production management.

Development of an Alternative In Vitro Rumen Fermentation Prediction Model.

The aim of this study is to identify an alternative approach for simulating the in vitro fermentation and quantifying the production of rumen methane and rumen acetic acid during the rumen fermentation process with different total mixed rations. In this experiment, dietary nutrient compositions (neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), and dry matter (DM)) were selected as input parameters to establish three prediction models for rumen fermentation parameters (methane and acetic acid): an artificial neural network model, a genetic algorithm-bp model, and a support vector machine model. The research findings show that the three models had similar simulation results that aligned with the measured data trends (R2 ≥ 0.83). Additionally, the root mean square errors (RMSEs) were ≤1.85 mL/g in the rumen methane model and ≤2.248 mmol/L in the rumen acetic acid model. Finally, this study also demonstrates the models' capacity for generalization through an independent verification experiment, as they effectively predicted outcomes even when significant trial factors were manipulated. These results suggest that machine learning-based in vitro rumen models can serve as a valuable tool for quantifying rumen fermentation parameters, guiding the optimization of dietary structures for dairy cows, rapidly screening methane-reducing feed options, and enhancing feeding efficiency.

Silibinin reduces in vitro methane production by regulating the rumen microbiome and metabolites.

This study used Silibinin as an additive to conduct fermentation experiments, wherein its effects on rumen gas production, fermentation, metabolites, and microbiome were analyzed in vitro. The silibinin inclusion level were 0 g/L (control group), 0.075 g/L, 0.15 g/L, 0.30 g/L, and 0.60 g/L (experimental group). Fermentation parameters, total gas production, carbon dioxide (CO2), methane (CH4), hydrogen (H2), and their percentages were determined. Further analysis of the rumen microbiome's relative abundance and α/ß diversity was performed on the Illumina NovaSeq sequencing platform. Qualitative and quantitative metabolomics analyses were performed to analyze the differential metabolites and metabolic pathways based on non-targeted metabolomics. The result indicated that with an increasing dose of silibinin, there was a linear reduction in total gas production, CO2, CH4, H2 and their respective percentages, and the acetic acid to propionic acid ratio. Concurrent with a linear increase in pH, when silibinin was added at 0.15 g/L and above, the total volatile fatty acid concentration decreased, the acetic acid molar ratio decreased, the propionic acid molar ratio increased, and dry matter digestibility decreased. At the same time, the relative abundance of Prevotella, Isotricha, Ophryoscolex, unclassified_Rotifera, Methanosphaera, Orpinomyces, and Neocallimastix in the rumen decreased after adding 0.60 g/L of silibinin. Simultaneously, the relative abundance of Succiniclasticum, NK4A214_group, Candidatus_Saccharimonas, and unclassified_Lachnospiraceae increased, altering the rumen species composition, community, and structure. Furthermore, it upregulated the ruminal metabolites, such as 2-Phenylacetamide, Phlorizin, Dalspinin, N6-(1,2-Dicarboxyethyl)-AMP, 5,6,7,8-Tetrahydromethanopterin, Flavin mononucleotide adenine dinucleotide reduced form (FMNH), Pyridoxine 5'-phosphate, Silibinin, and Beta-D-Fructose 6-phosphate, affecting phenylalanine metabolism, flavonoid biosynthesis, and folate biosynthesis pathways. In summary, adding silibinin can alter the rumen fermentation parameters and mitigate enteric methane production by regulating rumen microbiota and metabolites, which is important for developing novel rumen methane inhibitors.

Potentially migrating and residual components of biochar: Effects on phosphorus adsorption performance and storage capacity of black soil.

Biochar has great potential to increase the soil nutrient storage capacity. However, with aging, biochar gradually disintegrates and releases fractions with migration potential, resulting in unknown effects on soil nutrient regulation. Based on this problem, we used ultrasound to separate original biochar (TB) into potentially migrating biochar (DB) and residual biochar (RB). The elemental composition and pore characteristics of TB, DB and RB were analyzed. Different fractions of biochar were applied to black soil, and the kinetic model and isothermal adsorption models were used to explore the adsorption characteristics of different treatments. Then, the effects of initial pH and coexisting ions on adsorption were compared. The adsorption mechanism and potential leaching process of phosphorus in soil were investigated. The results showed that RB had higher O and H contents and was less stable than TB, while RB was more aromatic. The phosphorus adsorption capacity of different treatments was SRB (1.3318 mg g-1) > STB (1.2873 mg g-1) > SDB (1.3025 mg g-1) > SCK (1.1905 mg g-1). SRB had optimal phosphorus adsorption performance and storage capacity, with a maximum adsorption capacity of 1.6741 mg g-1 for the Langmuir isotherm, and it also showed excellent applicability in a pH gradient and with coexisting ions. The main adsorption mode of phosphorus by different treatments was monolayer chemisorption, related to electrostatic repulsion and oxygen-containing functional groups. DB was less effective in inhibiting soil phosphorus migration, with the cumulative leaching of SDB reaching 8.99 mg and the percentage of phosphorus in the 0-6 cm soil layer reaching only 15.42%. Overall, the results can help elucidate potential trends in the adsorption performance and migration process of soil phosphorus by biochar, and improve the comprehensive utilization efficiency of biochar.

Effect of freeze-thaw cycles and biochar coupling on the soil water-soil environment, nitrogen adsorption and N2O emissions in seasonally frozen regions.

Freeze-thaw cycles (FTCs) usually occur in the nongrowing season of crops, and the temporal mismatch between soil nitrogen (N) supply and crop N utilization increases the risk of N loss. Crop straw burning is a seasonal air pollution source, and biochar provides new alternatives for waste biomass recycling and soil pollution remediation. To investigate the effect of biochar on N loss and N2O emissions under frequent FTCs, different biochar content treatments (0 %, 1 %, 2 %) were set, and laboratory simulated soil column FTC tests were conducted. Based on the Langmuir and Freundlich models, the surface microstructure evolution and N adsorption mechanism of biochar before and after FTCs were analyzed, and the change characteristics of the soil water-soil environment, available N and N2O emissions under the interactive effect of FTCs and biochar were studied. The results showed that FTCs increased the oxygen (O) content by 19.69 % and the N content by 17.75 % and decreased the carbon (C) content by 12.39 % of biochar. The increase in the N adsorption capacity of biochar after FTCs was related to changes in surface structure and chemical properties. Biochar can improve the soil water-soil environment, adsorb available nutrients, and reduce N2O emissions by 35.89 %-46.31 %. The water-filled pore space (WFPS) and urease activity (S-UE) were the main environmental factors determining N2O emissions. Ammonium nitrogen (NH4+-N) and microbial biomass nitrogen (MBN), as substrates of N biochemical reactions, significantly affected N2O emissions. The interaction of biochar content and FTCs in different treatments had significant effects on available N (p < 0.05). The application of biochar is an effective way to reduce N loss and N2O emissions under the action of frequent FTCs. These research results can provide a reference for the rational application of biochar and efficient utilization of soil hydrothermal resources in seasonally frozen soil areas.

Effects of biochar and freeze‒thaw cycles on the bacterial community and multifunctionality in a cold black soil area.

Global climate change has altered soil freeze‒thaw cycle events, and little is known about soil microbe response to and multifunctionality regarding freeze‒thaw cycles. Therefore, in this study, biochar was used as a material to place under seasonal freeze-thaw cycling conditions. The purpose of this study was to explore the ability of biochar to regulate the function of freeze-thaw soil cycles to ensure spring sowing and food production. The results showed that biochar significantly increased the richness and diversity of soil bacteria before and after freezing-thawing. In the freezing period, the B50 treatment had the greatest improvement effect (2.6% and 5.5%, respectively), while in the thawing period, the B75 treatment had the best improvement effect. Biochar changed the composition and distribution characteristics of the bacterial structure and enhanced the multifunctionality of freeze-thaw soil and the stability of the bacterial symbiotic network. Compared with the CK treatment, the topological characteristics of the bacterial ecological network of the B50 treatment increased the most. They were 0.89 (Avg.degree), 9.79 (Modularity), 9 (Nodes), and 255 (Links). The freeze-thaw cycle decreased the richness and diversity of the bacterial community and changed the composition and distribution of the bacterial community, and the total bacterial population decreased by 658 (CK), 394 (B25), 644 (B50) and 86 (B75) during the thawing period compared with the freezing period. The soil multifunctionality in the freezing period was higher than that during the thawing period, indicating that the freeze-thaw cycle reduced soil ecological function. From the perspective of abiotic analysis, the decrease in soil multifunctionality was due to the decrease in soil nutrients, enzyme activities, soil basic respiration and other singular functions. From the perspective of bacteria, the decrease in soil multifunctionality was mainly due to the change in the Actinobacteriota group. This work expands the understanding of biochar ecology in cold black soil. These results are conducive to the sustainable development of soil ecological function in cold regions and ultimately ensure crop growth and food productivity.

Rumen Fermentation Parameters Prediction Model for Dairy Cows Using a Stacking Ensemble Learning Method.

Volatile fatty acids (VFAs) and methane are the main products of rumen fermentation. Quantitative studies of rumen fermentation parameters can be performed using in vitro techniques and machine learning methods. The currently proposed models suffer from poor generalization ability due to the small number of samples. In this study, a prediction model for rumen fermentation parameters (methane, acetic acid (AA), and propionic acid (PA)) of dairy cows is established using the stacking ensemble learning method and in vitro techniques. Four factors related to the nutrient level of total mixed rations (TMRs) are selected as inputs to the model: neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), and dry matter (DM). The comparison of the prediction results of the stacking model and base learners shows that the stacking ensemble learning method has better prediction results for rumen methane (coefficient of determination (R2) = 0.928, root mean square error (RMSE) = 0.968 mL/g), AA (R2 = 0.888, RMSE = 1.975 mmol/L) and PA (R2 = 0.924, RMSE = 0.74 mmol/L). And the stacking model simulates the variation of methane and VFAs in relation to the dietary fiber content. To demonstrate the robustness of the model in the case of small samples, an independent validation experiment was conducted. The stacking model successfully simulated the transition of rumen fermentation type and the change of methane content under different concentrate-to-forage (C:F) ratios of TMR. These results suggest that the rumen fermentation parameter prediction model can be used as a decision-making basis for the optimization of dairy cow diet compositions, rapid screening of methane emission reduction, feed beneficial to dairy cow health, and improvement of feed utilization.

Enrichment of low abundance DNA/RNA by oligonucleotide-clicked iron oxide nanoparticles.

Detection of low abundance target DNA/RNA for clinical or research purposes is challenging because the target sequences can be hidden under a large background of human genomic or non-human metagenomic sequences. We describe a probe-based capture method to enrich for target sequences with DNA-clicked iron oxide nanoparticles. Our method was tested against commercial capture assays using streptavidin beads, on a set of probes derived from a common genotype of the hepatitis C virus. We showed that our method is more specific and sensitive, most likely due to the combination of an inert silica coating and a high density of DNA probes clicked to the nanoparticles. This facilitates target capture below the limits of detection for TaqMan qPCR, and we believe that this method has the potential to transform management of infectious diseases.

Biomining of MoS2 with Peptide-based Smart Biomaterials.

Biomining of valuable metals using a target specific approach promises increased purification yields and decreased cost. Target specificity can be implemented with proteins/peptides, the biological molecules, responsible from various structural and functional pathways in living organisms by virtue of their specific recognition abilities towards both organic and inorganic materials. Phage display libraries are used to identify peptide biomolecules capable of specifically recognizing and binding organic/inorganic materials of interest with high affinities. Using combinatorial approaches, these molecular recognition elements can be converted into smart hybrid biomaterials and harnessed for biotechnological applications. Herein, we used a commercially available phage-display library to identify peptides with specific binding affinity to molybdenite (MoS2) and used them to decorate magnetic NPs. These peptide-coupled NPs could capture MoS2 under a variety of environmental conditions. The same batch of NPs could be re-used multiple times to harvest MoS2, clearly suggesting that this hybrid material was robust and recyclable. The advantages of this smart hybrid biomaterial with respect to its MoS2-binding specificity, robust performance under environmentally challenging conditions and its recyclability suggests its potential application in harvesting MoS2 from tailing ponds and downstream mining processes.

Peptide-functionalized iron oxide magnetic nanoparticle for gold mining.

Here, we present our work on preparing a novel nanomaterial composed of inorganic binding peptides and magnetic nanoparticles for inorganic mining. Two previously selected and well-characterized gold-binding peptides from cell surface display, AuBP1 and AuBP2, were exploited. This nanomaterial (AuBP-MNP) was designed to fulfill the following two significant functions: the surface conjugated gold-binding peptide will recognize and selectively bind to gold, while the magnetic nano-sized core will respond and migrate according to the applied external magnetic field. This will allow the smart nanomaterial to mine an individual material (gold) from a pool of mixture, without excessive solvent extraction, filtration, and concentration steps. The working efficiency of AuBP-MNP was determined by showing a dramatic reduction of gold nanoparticle colloid concentration, monitored by spectroscopy. The binding kinetics of AuBP-MNP onto the gold surface was determined using surface plasmon resonance (SPR) spectroscopy, which exhibits around 100 times higher binding kinetics than peptides alone. The binding capacity of AuBP-MNP was demonstrated by a bench-top mining test with gold microparticles. Graphical abstract.