An Improved Method for Broiler Weight Estimation Integrating Multi-Feature with Gradient Boosting Decision Tree.

Broiler weighing is essential in the broiler farming industry. Camera-based systems can economically weigh various broiler types without expensive platforms. However, existing computer vision methods for weight estimation are less mature, as they focus on young broilers. In effect, the estimation error increases with the age of the broiler. To tackle this, this paper presents a novel framework. First, it employs Mask R-CNN for instance segmentation of depth images captured by 3D cameras. Next, once the images of either a single broiler or multiple broilers are segmented, the extended artificial features and the learned features extracted by Customized Resnet50 (C-Resnet50) are fused by a feature fusion module. Finally, the fused features are adopted to estimate the body weight of each broiler employing gradient boosting decision tree (GBDT). By integrating diverse features with GBTD, the proposed framework can effectively obtain the broiler instance among many depth images of multiple broilers in the visual field despite the complex background. Experimental results show that this framework significantly boosts accuracy and robustness. With an MAE of 0.093 kg and an R2 of 0.707 in a test set of 240 63-day-old bantam chicken images, it outperforms other methods.

Automatic Penaeus Monodon Larvae Counting via Equal Keypoint Regression with Smartphones.

Today, large-scale Penaeus monodon farms no longer incubate eggs but instead purchase larvae from large-scale hatcheries for rearing. The accurate counting of tens of thousands of larvae in these transactions is a challenging task due to the small size of the larvae and the highly congested scenes. To address this issue, we present the Penaeus Larvae Counting Strategy (PLCS), a simple and efficient method for counting Penaeus monodon larvae that only requires a smartphone to capture images without the need for any additional equipment. Our approach treats two different types of keypoints as equip keypoints based on keypoint regression to determine the number of shrimp larvae in the image. We constructed a high-resolution image dataset named Penaeus_1k using images captured by five smartphones. This dataset contains 1420 images of Penaeus monodon larvae and includes general annotations for three keypoints, making it suitable for density map counting, keypoint regression, and other methods. The effectiveness of the proposed method was evaluated on a real Penaeus monodon larvae dataset. The average accuracy of 720 images with seven different density groups in the test dataset was 93.79%, outperforming the classical density map algorithm and demonstrating the efficacy of the PLCS.

Exploring the RNA Editing Events and Their Potential Regulatory Roles in Tea Plant (Camellia sinensis L.).

RNA editing is a post-transcriptional modification process that alters the RNA sequence relative to the genomic blueprint. In plant organelles (namely, mitochondria and chloroplasts), the most common type is C-to-U, and the absence of C-to-U RNA editing results in abnormal plant development, such as etiolation and albino leaves, aborted embryonic development and retarded seedling growth. Here, through PREP, RES-Scanner, PCR and RT-PCR analyses, 38 and 139 RNA editing sites were identified from the chloroplast and mitochondrial genomes of Camellia sinensis, respectively. Analysis of the base preference around the RNA editing sites showed that in the -1 position of the edited C had more frequent occurrences of T whereas rare occurrences of G. Three conserved motifs were identified at 25 bases upstream of the RNA editing site. Structural analyses indicated that the RNA secondary structure of 32 genes, protein secondary structure of 37 genes and the three-dimensional structure of 5 proteins were altered due to RNA editing. The editing level analysis of matK and ndhD in six tea cultivars indicated that matK-701 might be involved in the color change of tea leaves. Furthermore, 218 PLS-CsPPR proteins were predicted to interact with the identified RNA editing sites. In conclusion, this study provides comprehensive insight into RNA editing events, which will facilitate further study of the RNA editing phenomenon of the tea plant.

Fine-Grained Butterfly Recognition via Peer Learning Network with Distribution-Aware Penalty Mechanism.

Automatic species recognition plays a key role in intelligent agricultural production management and the study of species diversity. However, fine-grained species recognition is a challenging task due to the quite diverse and subtle interclass differences among species and the long-tailed distribution of sample data. In this work, a peer learning network with a distribution-aware penalty mechanism is proposed to address these challenges. Specifically, the proposed method employs the two-stream ResNeSt-50 as the backbone to obtain the initial predicted results. Then, the samples, which are selected from the instances with the same predicted labels by knowledge exchange strategy, are utilized to update the model parameters via the distribution-aware penalty mechanism to mitigate the bias and variance problems in the long-tailed distribution. By performing such adaptive interactive learning, the proposed method can effectively achieve improved recognition accuracy for head classes in long-tailed data and alleviate the adverse effect of many head samples relative to a few samples of the tail classes. To evaluate the proposed method, we construct a large-scale butterfly dataset (named Butterfly-914) that contains approximately 72,152 images belonging to 914 species and at least 20 images for each category. Exhaustive experiments are conducted to validate the efficiency of the proposed method from several perspectives. Moreover, the superior Top-1 accuracy rate (86.2%) achieved on the butterfly dataset demonstrates that the proposed method can be widely used for agricultural species identification and insect monitoring.

N-doped carbon fibers in situ prepared by hydrothermal carbonization of Camellia sinensis branches waste for efficient removal of heavy metal ions.

N-doped carbon fibers (NCFs) were in situ prepared by Camellia sinensis branches waste through hydrothermal carbonization with urea/ZnCl2 at 160-280 °C under 0.8-8.9 MPa. The structural characteristics of NCFs were investigated by elemental analysis, SEM, TEM, XRD, XPS, Raman spectra, and BET surface area. The highest N content of NCFs obtained at 280 °C was 8.96%, and the main forms of doped N were pyridinic N, pyrrolic N, and graphitic N. Moreover, NCFs were applied to remove metal ions successfully. The results showed that NCF-240 had the maximum adsorption amounts of 106.52, 125.23, and 153.49 mg/g for Cu2+, Pb2+, and Zn2+, respectively, while NCF-280 had the best removal ability on Cr6+ (145.67 mg/g). Finally, it demonstrated that the adsorption behavior of NCFs was well fitted by the pseudo-second-order kinetic and the Langmuir adsorption isotherm models.

Holstein Cattle Face Re-Identification Unifying Global and Part Feature Deep Network with Attention Mechanism.

In precision dairy farming, computer vision-based approaches have been widely employed to monitor the cattle conditions (e.g., the physical, physiology, health and welfare). To this end, the accurate and effective identification of individual cow is a prerequisite. In this paper, a deep learning re-identification network model, Global and Part Network (GPN), is proposed to identify individual cow face. The GPN model, with ResNet50 as backbone network to generate a pooling of feature maps, builds three branch modules (Middle branch, Global branch and Part branch) to learn more discriminative and robust feature representation from the maps. Specifically, the Middle branch and the Global branch separately extract the global features of middle dimension and high dimension from the maps, and the Part branch extracts the local features in the unified block, all of which are integrated to act as the feature representation for cow face re-identification. By performing such strategies, the GPN model not only extracts the discriminative global and local features, but also learns the subtle differences among different cow faces. To further improve the performance of the proposed framework, a Global and Part Network with Spatial Transform (GPN-ST) model is also developed to incorporate an attention mechanism module in the Part branch. Additionally, to test the efficiency of the proposed approach, a large-scale cow face dataset is constructed, which contains 130,000 images with 3000 cows under different conditions (e.g., occlusion, change of viewpoints and illumination, blur, and background clutters). The results of various contrast experiments show that the GPN outperforms the representative re-identification methods, and the improved GPN-ST model has a higher accuracy rate (up by 2.8% and 2.2% respectively) in Rank-1 and mAP, compared with the GPN model. In conclusion, using the Global and Part feature deep network with attention mechanism can effectively ameliorate the efficiency of cow face re-identification.

Essential Roles of the Linker Sequence Between Tetratricopeptide Repeat Motifs of Ethylene Overproduction 1 in Ethylene Biosynthesis.

Ethylene Overproduction 1 (ETO1) is a negative regulator of ethylene biosynthesis. However, the regulation mechanism of ETO1 remains largely unclear. Here, a novel eto1 allele (eto1-16) was isolated with typical triple phenotypes due to an amino acid substitution of G480C in the uncharacterized linker sequence between the TPR1 and TPR2 motifs. Further genetic and biochemical experiments confirmed the eto1-16 mutation site. Sequence analysis revealed that G480 is conserved not only in two paralogs, EOL1 and EOL2, in Arabidopsis, but also in the homologous protein in other species. The glycine mutations (eto1-11, eto1-12, and eto1-16) do not influence the mRNA abundance of ETO1, which is reflected by the mRNA secondary structure similar to that of WT. According to the protein-protein interaction analysis, the abnormal root phenotype of eto1-16 might be caused by the disruption of the interaction with type 2 1-aminocyclopropane-1-carboxylic acid (ACC) synthases (ACSs) proteins. Overall, these data suggest that the linker sequence between tetratricopeptide repeat (TPR) motifs and the glycine in TPR motifs or the linker region are essential for ETO1 to bind with downstream mediators, which strengthens our knowledge of ETO1 regulation in balancing ACSs.

Genome-Wide Identification and Characterization of the CsFHY3/FAR1 Gene Family and Expression Analysis under Biotic and Abiotic Stresses in Tea Plants (Camellia sinensis).

The FHY3/FAR1 transcription factor family, derived from transposases, plays important roles in light signal transduction, and in the growth and development of plants. However, the homologous genes in tea plants have not been studied. In this study, 25 CsFHY3/FAR1 genes were identified in the tea plant genome through a genome-wide study, and were classified into five subgroups based on their phylogenic relationships. Their potential regulatory roles in light signal transduction and photomorphogenesis, plant growth and development, and hormone responses were verified by the existence of the corresponding cis-acting elements. The transcriptome data showed that these genes could respond to salt stress and shading treatment. An expression analysis revealed that, in different tissues, especially in leaves, CsFHY3/FAR1s were strongly expressed, and most of these genes were positively expressed under salt stress (NaCl), and negatively expressed under low temperature (4 °C) stress. In addition, a potential interaction network demonstrated that PHYA, PHYC, PHYE, LHY, FHL, HY5, and other FRSs were directly or indirectly associated with CsFHY3/FAR1 members. These results will provide the foundation for functional studies of the CsFHY3/FAR1 family, and will contribute to the breeding of tea varieties with high light efficiency and strong stress resistance.

Structural analysis and heavy metal adsorption of N-doped biochar from hydrothermal carbonization of Camellia sinensis waste.

N-doped biochar as adsorption material for heavy metal removal has attracted increasing concern in environmental application due to its unique features. Here, N-doped biochar was prepared by hydrothermal carbonization of Camellia sinensis branch waste using KOH/NH4Cl at 120-280 °C for 2 h under 0.4-6.5 MPa, followed by structural analysis. The results showed that the highest N content determined by elemental analysis could reach up to 6.18% in biochar, and the major N species were involved in graphitic N, pyrrolic N, and pyridinic N. Interestingly, these N-doped biochar exhibited the effective adsorption ability of Cu2+, Pb2+, Zn2+, and Cr6+. The batch adsorption behavior had a better adjustment to the pseudo-second-order kinetic and the Langmuir adsorption isotherm models. In brief, the present results are attributed to develop low-cost adsorbent for removing heavy metal ions.

Urea/ZnCl2 in situ hydrothermal carbonization of Camellia sinensis waste to prepare N-doped biochar for heavy metal removal.

Environmental benefits of biochar require a simple and effective method for preparation of functional N-doped biochar. In this study, urea/ZnCl2 was developed to prepare N-doped biochar via in situ hydrothermal carbonization (HTC) of Camellia sinensis waste at 120-280 °C for 2 h under 1.0-9.8 MPa. Physicochemical and structural properties of the N-doped biochar were investigated by Raman spectra, elemental analysis, BET surface area, SEM, TEM, XRD, and XPS. The results showed that the N content in biochar could reach up to 7.79% at 280 °C. Surface chemistry suggested that pyridinic N, pyrollic N, and graphitic N were the major N species on the biochar. Moreover, the N-doped biochar was successfully employed to remove metal ions Cu2+, Pb2+, Zn2+, and Cr6+. Adsorption data fit closely to the pseudo-second-order kinetic equation and the Langmuir adsorption isotherm model for all metal ions.

The complete mitochondrial genome of Eterusia aedea (Lepidoptera, Zygaenidae) and comparison with other zygaenid moths.

Zygaenidae comprises >1036 species, including many folivorous pests in agriculture. In the present study, the complete mitochondrial genome (mitogenome) of a major pest of tea trees, Eterusia aedea was determined. The 15,196-bp circular genome contained the common set of 37 mitochondrial genes (including 13 protein-coding genes, two rRNA genes, and 22 tRNA genes) and exhibited the similar genomic features to reported Zygaenidae mitogenome. Comparative analyses of Zygaenidae mitogenomes showed a typical evolutionary trend of lepidopteran mitogenomes. In addition, we also investigated the gene order of lepidopteran mitogenomes and proposed that the novel gene order trnA-trnR-trnN-trnE-trnS-trnF from Zygaenidae and Gelechiidae and most other gene rearrangements of this tRNA cluster evolved independently. Finally, the mitogenomic phylogeny of Lepidoptera was reconstructed based on multiple mitochondrial datasets. And all the phylogenetic results revealed the sister relationships of Cossoidea and Zygaenoidea with both BI and ML methods, which is the first stable mitogenomic evidence for this clade.

Alternative splicing affects the targeting sequence of peroxisome proteins in Arabidopsis.

KEY MESSAGE: A systematic analysis of the Arabidopsis genome in combination with localization experiments indicates that alternative splicing affects the peroxisomal targeting sequence of at least 71 genes in Arabidopsis. Peroxisomes are ubiquitous eukaryotic cellular organelles that play a key role in diverse metabolic functions. All peroxisome proteins are encoded by nuclear genes and target to peroxisomes mainly through two types of targeting signals: peroxisomal targeting signal type 1 (PTS1) and PTS2. Alternative splicing (AS) is a process occurring in all eukaryotes by which a single pre-mRNA can generate multiple mRNA variants, often encoding proteins with functional differences. However, the effects of AS on the PTS1 or PTS2 and the targeting of the protein were rarely studied, especially in plants. Here, we systematically analyzed the genome of Arabidopsis, and found that the C-terminal targeting sequence PTS1 of 66 genes and the N-terminal targeting sequence PTS2 of 5 genes are affected by AS. Experimental determination of the targeting of selected protein isoforms further demonstrated that AS at both the 5' and 3' region of a gene can affect the inclusion of PTS2 and PTS1, respectively. This work underscores the importance of AS on the global regulation of peroxisome protein targeting.

[Engineering of a flavonoid 3'-hydroxylase from tea plant (Camellia sinensis) for biosynthesis of B-3',4'-dihydroxylated flavones].

Objective: A flavonoid 3'-hydroxylase from tea plant was engineered to synthesize B-3',4'-dihydroxylated flavones such as eriodictyol, dihydroquercetin and quercetin. Methods: Four articifical P450 constructs harboring both flavonoid 3'-hydroxylase gene from Camellia sinensis (CsF3'H) and P450 reductase gene from Arabidopsis thaliana (ATR1 or ATR2) were introduced into Escherichia coli strains TOP10, DH5α and BL21, resultantly engineering strains S1 to S12. The plasmid pYES-Dest52-CsF3'H harboring CsF3'H gene was introduced into yeast Saccharomyces cerevisiae WAT11 designated as strain S13. The plasmid pES-HIS-CsF3H::AtFLS 9 AA was constructed through fusing flavanone 3-hydroxylase gene from Camellia sinensis (CsF3H) and flavonol synthase gene from Arabidopsis thaliana (AtFLS). Plasmid pES-URA-CsF3'H and pES-HIS-CsF3H::AtFLS 9 AA were then co-introduced into yeast S. cerevisiae WAT11 designated as strain S14. Results: Strain S6 generated highest bioconversion efficiency at 25℃ among all E. coli strains during 24 h fernentation. Supplemented with 1000 µmol/L naringenin, dihydrokaempferol and kaempferol, the maximum amounts of eriodictyol, dihydroquercetin and quercetin produced by strain S13 were 734.32 µmol/L, 446.07 µmol/L and 594.64 µmol/L respectively. Supplemented with 5 mmol/L naringenin, the maximum amounts of eriodictyol, kaempferol, quercetin, dihydroquercetin and dihydrokaempferol produced by strain S14 were 1412.16 µmol/L, 490.25 µmol/L, 445.75 µmol/L, 66.75 µmol/L and 73.50 µmol/L during 36-48 h fermentaion respectively. Conclusion: CsF3'H was engineered for biosynthesis of B-3',4'-dihydroxylated flavone.

Synthesis and characterization of carboxymethyl xylan-g-poly(propylene oxide) and its application in films.

Carboxymethyl xylan-g-poly(propylene oxide) (CMX-g-PPO) was successfully synthesized by grafting poly(propylene oxide) chains onto xylan from bamboo using the Al(Oi-Pr)3 initiated ring-opening polymerization of propylene oxides, followed by carboxymethylation with sodium chloroacetate under microwave irradiation. The synthesized CMX-g-PPO was well characterized by FT-IR, (13)C NMR, and AFM. The AFM imaging showed that the average sizes of xylan were 422.1×67.4×1.2nm, while the average sizes of grafting branches PPO were 128.0×38.5×5.1nm, which firstly provided an irrefutable and visual evidence for the structure of grafted xylan at single molecular level. Subsequently, a serial of CMX-g-PPO/CS films were prepared without addition of any plasticizers. The surface morphologies, wettability, water vapor barrier properties, mechanical properties, and thermal stabilities of the obtained films were characterized and compared with those of the control films by AFM, contact angle, WVP, tensile testing, and TGA, respectively.

Arabidopsis FHY3/CPD45 regulates far-red light signaling and chloroplast division in parallel.

CPD45 (chloroplast division45),which is also known as FHY3 (far-red elongated hypocotyl3), is a key factor in the far-red light signaling pathway in Arabidopsis. We previously showed that FHY3/CPD45 also regulates chloroplast division. Because light is also a regulator of chloroplast development and division, we sought to clarify the relationship between far-red light signaling and chloroplast division pathways. We found that the chloroplast division mutant arc5-3 had no defect in far-red light sensing, and that constitutive overexpression of ARC5 rescued the chloroplast division defect, but not the defect in far-red light signaling, of cpd45. fhy1, which is defective in far-red light signaling, exhibited normal chloroplast division. Constitutive overexpression of FHY1 rescued the far-red light signaling defect, but not the chloroplast division defect, of cpd45. Moreover, ARC5 and FHY1 expression were not affected in fhy1 and arc5-3, respectively. Based on these results, we propose that FHY3/CPD45 regulates far-red light signaling and chloroplast division in parallel by activating the expression of FHY1 and ARC5 independently. This work demonstrates how relationships between different pathways in a gene regulatory network can be explored.

Arabidopsis FRS4/CPD25 and FHY3/CPD45 work cooperatively to promote the expression of the chloroplast division gene ARC5 and chloroplast division.

ARC5 is a dynamin-related GTPase essential for the division of chloroplasts in plants. The arc5 mutant frequently exhibits enlarged, dumbbell-shaped chloroplasts, indicating a role for ARC5 in the constriction of the chloroplast division site. In a screen for chloroplast division mutants with a phenotype similar to arc5, two mutants, cpd25 and cpd45, were obtained. CPD45 was identified as being the same gene as FHY3, a key regulator of far-red light signaling recently shown to be involved in the regulation of ARC5. CPD25 was previously named FRS4 and is homologous to FHY3. We found that CPD25 is also required for the expression of ARC5, suggesting that its function is not redundant to that of FHY3. Moreover, cpd25 does not have the far-red light-sensing defect present in fhy3 and far1. Both FRS4/CPD25 and FHY3/CPD45 could bind to the FBS-like 'ACGCGC' motifs in the promoter region of ARC5, and the binding efficiency of FRS4/CPD25 was much higher than that of FHY3/CPD45. Unlike FHY3/CPD45, FRS4/CPD25 has no ARC5 activation activity. Our data suggest that FRS4/CPD25 and FHY3/CPD45 function as a heterodimer that cooperatively activates ARC5, that FRS4/CPD25 plays the major role in promoter binding, and that FHY3/CPD45 is largely responsible for the gene activation. This study not only provides insight into the mechanisms underlying the regulation of chloroplast division in higher plants, but also suggests a model that shows how members of a transcription factor family can evolve to have different DNA-binding and gene activation features.

Genetic mapping and isolation of two arc3 alleles in Arabidopsis.

KEY MESSAGE : Two new alleles of arc3 in Arabidopsis thaliana, arc3-4 and arc3-5, were isolated in the Columbia-0 ecotype. The mutants were characterized in detail using microscopy and molecular techniques. Chloroplasts are essential organelles for photosynthesis in plant cells. Division of chloroplasts is coordinated by the internal division machinery (mainly the tubulin-like FtsZ ring) and the external division machinery (mainly the dynamin-like ARC5 ring). Accumulation and replication of chloroplasts3 (ARC3) is important for the correct positioning of chloroplast division machinery. During evolution, ARC3 has probably replaced minicellC (MinC), an important factor involved in positioning of the division site in bacteria. However, the working mechanism of ARC3 is still unclear. Using forward genetic approaches, we isolated two new alleles of arc3 in Arabidopsis thaliana, arc3-4 and arc3-5, in which mutant loci differed from those of previously reported arc3 mutants. Microscopy analyses showed more detailed, and some new, phenotypes of arc3 mutants. Reverse-transcription polymerase chain reaction (RT-PCR) and real-time quantitative RT-PCR (qRT-PCR) results indicated that the mRNA of the ARC3 gene was unstable in arc3-4 and arc3-5 mutant plants. Also, RNA secondary structures of the ARC3 gene were predicted to differ between these two arc3 mutants and wild type. Our studies increase our understanding of the function of ARC3 in chloroplast division.