Optimizing multicopy chromosomal integration for stable high-performing strains.

The copy number of genes in chromosomes can be modified by chromosomal integration to construct efficient microbial cell factories but the resulting genetic systems are prone to failure or instability from triggering homologous recombination in repetitive DNA sequences. Finding the optimal copy number of each gene in a pathway is also time and labor intensive. To overcome these challenges, we applied a multiple nonrepetitive coding sequence calculator that generates sets of coding DNA sequence (CDS) variants. A machine learning method was developed to calculate the optimal copy number combination of genes in a pathway. We obtained an engineered Yarrowia lipolytica strain for eicosapentaenoic acid biosynthesis in 6 months, producing the highest titer of 27.5 g l-1 in a 50-liter bioreactor. Moreover, the lycopene production in Escherichia coli was also greatly improved. Importantly, all engineered strains of Y. lipolytica, E. coli and Saccharomyces cerevisiae constructed with nonrepetitive CDSs maintained genetic stability.

Wnt/ß-catenin signaling in the development and therapeutic resistance of non-small cell lung cancer.

Wnt/ß-catenin signaling is a critical pathway that influences development and therapeutic response of non-small cell lung cancer (NSCLC). In recent years, many Wnt regulators, including proteins, miRNAs, lncRNAs, and circRNAs, have been found to promote or inhibit signaling by acting on Wnt proteins, receptors, signal transducers and transcriptional effectors. The identification of these regulators and their underlying molecular mechanisms provides important implications for how to target this pathway therapeutically. In this review, we summarize recent studies of Wnt regulators in the development and therapeutic response of NSCLC.

Clarifying the Functional Role of Serotonin in Meloidogyne graminicola Host Plant Parasitism by Immunolocalization and RNA Interference.

Serotonin (5-hydroxytryptamine) is an essential neurotransmitter involved in regulating various behaviors in plant-parasitic nematodes, including locomotion, egg laying, feeding, and mating. However, the functional role of serotonin in root-knot nematode invasion of host plants and the molecular mechanisms underlying feeding behavior remain poorly understood. In this study, we tested the effects of exogenous serotonin and the pharmacological compounds fluoxetine and methiothepin on the feeding behaviors of Meloidogyne graminicola. Our results suggested that M. graminicola possesses an endogenous serotonin signaling pathway and that serotonin plays a crucial role in modulating feeding behaviors in M. graminicola second-stage juveniles. We also identified and cloned the serotonin synthesis enzyme tryptophan hydroxylase (Mg-tph-1) in M. graminicola and investigated the role of endogenous serotonin by generating RNA interference nematodes in Mg-tph-1. Silencing Mg-tph-1 substantially reduced nematode invasion, development, and reproduction. According to the immunostaining results, we speculated that these serotonin immunoreactive cells near the nerve ring in M. graminicola are likely homologous to Caenorhabditis elegans ADFs, NSMs, and RIH serotonergic neurons. Furthermore, we investigated the impact of phytoserotonin on nematode invasion and development in rice by overexpressing OsTDC-3 or supplementing rice plants with tryptamine and found that an increase in phytoserotonin increases nematode pathogenicity. Overall, our study provides insights into the essential role of serotonin in M. graminicola host plant parasitism and proposes that the serotonergic signaling pathway could be a potential target for controlling plant-parasitic nematodes.

The soil microbial necromass carbon and the carbon pool stability drive a stronge priming effect following vegetation restoration.

The priming effect stands as a critical factor influencing the balance of soil organic carbon (SOC). Following vegetation restoration, the carbon (C) pool stability in Platycladus orientalis forests (PO) varies, and the priming effect resulting from exogenous C addition also differs significantly. Here, we selected PO with restoration ages of 10, 15, and 30 years in the rocky mountainous area in northern China and conducted measurements of soil properties, microbial communities, microbial necromass C (MNC), SOC fractions, and the priming effect characteristics to explore the main influencing factors of the priming effect, especially the microbiological mechanisms. Our results showed that the ratio of mineral-associated organic C to particulate organic C increased. The characteristics of the priming effect showed the same pattern, and there was a significant positive correlation between the C pool stability and the priming effect. The diversity of the fungal communities increased with increasing vegetation restoration age, and the content and proportion of fungal necromass C (FNC) also increased synchronously, reaching the maximum value in the soil of PO that had been restored for 30 years. In addition, the soil water content and total nitrogen indirectly affected the priming effect by influencing the microbial communities. In summary, the results suggested that vegetation restoration can enhance the C pool stability by promoting an increase in soil FNC, thereby producing a positive priming effect. This can help deepen our understanding of the SOC mineralization changes induced by fresh C input following vegetation restoration and provides a theoretical basis for better explaining the C cycle between soil and atmosphere under the vegetation restoration models in the future.

Characterization of lodging variation of weedy rice.

Weedy rice (Oryza spp.), one of the most notorious weeds of cultivated rice, evades eradication through stem lodging and seed shattering. Many studies have focused on seed shattering, whereas variations in lodging have received less attention and the underlying mechanisms that cause the differences in lodging between weedy and cultivated rice have not been studied in detail. Here, we compared lodging variation among diverse Chinese weedy rice strains and between weedy rice and co-occurring cultivated rice. The chemical composition of basal stems was determined, and transcriptome and methylome sequencing were used to assess the variation in expression of lodging-related genes. The results showed that the degree of lodging varied between indica-derived weed strains with high lodging levels, which occurred predominantly in southern China, and japonica-derived strains with lower lodging levels, which were found primarily in the north. The more lodging-prone indica weedy rice had a smaller bending stress and lower lignin content than non-lodging accessions. In comparison to co-occurring cultivated rice, there was a lower ratio of cellulose to lignin content in the lodging-prone weedy rice. Variation in DNA methylation of lignin synthesis-related OsSWN1, OsMYBX9, OsPAL1, and Os4CL3 mediated the differences in their expression levels and affected the ratio of cellulose to lignin content. Taken together, our results show that DNA methylation in lignin-related genes regulates variations in stem strength and lodging in weedy rice, and between weed strains and co-occurring cultivated rice.

Arbitrary-Shaped Text Detection with B-Spline Curve Network.

Text regions in natural scenes have complex and variable shapes. Directly using contour coordinates to describe text regions will make the modeling inadequate and lead to low accuracy of text detection. To address the problem of irregular text regions in natural scenes, we propose an arbitrary-shaped text detection model based on Deformable DETR called BSNet. The model differs from the traditional method of directly predicting contour points by using B-Spline curve to make the text contour more accurate and reduces the number of predicted parameters simultaneously. The proposed model eliminates manually designed components and dramatically simplifies the design. The proposed model achieves F-measure of 86.8% and 87.6% on CTW1500 and Total-Text, demonstrating the model's effectiveness.

Recent advances in biotechnological production of polyunsaturated fatty acids by Yarrowia lipolytica.

Owing to the important physiological functions, polyunsaturated fatty acids (PUFAs) play a vital role in protecting human health, such as preventing cancer, cardiovascular disease, and diabetes. Specifically, Yarrowia lipolytica has been identified as the most popular non-conventional oleaginous yeast, which can accumulate the abundant intracellular lipids, indicating that has great potential as an industrial host for production of PUFAs. Notably, some novel engineering strategies have been applied to endow and improve the abilities of Y. lipolytica to synthesize PUFAs, including construction and optimization of PUFAs biosynthetic pathways, improvement of preucrsors acetyl-coA and NADPH supply, inhibition of competing pathways, knockout of ß-oxidation pathways, regulation of oxidative stress defense pathways, and regulation of genes involved in upstream lipid metabolism. Besides, some bypass approaches, such as strain mating, evolutionary engineering, and computational model based on omics, also have been proposed to improve the performance of engineering strains. Generally, in this review, we summarized the recent advances in engineering strategies and bypass approaches for improving PUFAs production by Y. lipolytica. In addition, we further summarized the latest efforts of CRISPR/Cas genome editing technology in Y. lipolytica, which is aimed to provide its potential applications in PUFAs production.

Strategies for efficient production of recombinant proteins in Escherichia coli: alleviating the host burden and enhancing protein activity.

Escherichia coli, one of the most efficient expression hosts for recombinant proteins (RPs), is widely used in chemical, medical, food and other industries. However, conventional expression strains are unable to effectively express proteins with complex structures or toxicity. The key to solving this problem is to alleviate the host burden associated with protein overproduction and to enhance the ability to accurately fold and modify RPs at high expression levels. Here, we summarize the recently developed optimization strategies for the high-level production of RPs from the two aspects of host burden and protein activity. The aim is to maximize the ability of researchers to quickly select an appropriate optimization strategy for improving the production of RPs.

Developing a dynamic equilibrium system in Escherichia coli to improve the production of recombinant proteins.

The combination of Escherichia coli BL21 (DE3) and the pET expression system is used extensively for the expression of various recombinant proteins (RPs). However, RP overexpression often introduces a growth burden for the host, especially in the case of toxic proteins. The key to solving this problem is to reduce the host burden associated with protein overproduction, which is often achieved by regulating the expression or activity of T7 RNAP or growth-decoupled systems. However, these strategies mainly relieve or interrupt the robbing of host resources, and do not eliminate other types of host burdens in the production process. In this study, we constructed a production system based on a dynamic equilibrium to precisely relieve the host burden and increase the RP production. The system is composed of three modules, including the overexpression of basic growth-related genes (rRNA, RNAP core enzyme, sigma factors), prediction and overexpression of key proteins using the enzyme-constrained model ec_iECBD_1354, and dynamic regulation of growth-related and key protein expression intensity based on a burden-driven promoter. Using this system, the production of many high-burden proteins, including autolysis protein and E. coli membrane proteins, was increased to varying degrees. Among them, the cytosine transporter protein (CodB) was most significantly improved, with a 4.02-fold higher production compared to the wild strain. This system can effectively reduce the optimizing costs, and is suitable for developing various types of RP expression hosts rapidly. KEY POINTS: • The basic growth-related resources can relieve the host burden from recombinant protein. • The enzyme-constrained model can accurately predict key genes to improve yield. • The expression intensity can be dynamically adjusted with changes in burden.

Effects of vegetation and terrain changes on spatial heterogeneity of soil C-N-P in the coastal zone protected forests at northern China.

Soil organic carbon (SOC), total nitrogen (TN) and total phosphorus (TP) are important indicators reflecting soil quality, and they can be used to effectively evaluate the effect of soil remediation. Many studies have evaluated the content of SOC, TN and TP in different ecosystems. However, after constructing protected forests for ecological restoration in the ecologically fragile coastal zone, the spatial distribution and influencing mechanism of SOC, TN and TP content is still uncertain. In this study, the spatial heterogeneity and influencing factors of SOC, TN and TP in surface (0-20 cm) soil were analyzed by traditional analysis and geostatistics. A total of 39 soil samples were collected under the coastal zone protected forest types including Quercus acutissima Carruth (QAC), Pinus thunbergii Parl (PTP), mixed PTP and QAC (QP) and Castanea mollissima BL (CMB) in the coastal zone protected forests in northern China. The results show that SOC, TN and TP content were defined as moderate variation, and they also show significant changes under different protected forest types (P < 0.05). The semivariance results indicate that SOC, TN and TP all exhibited strong spatial dependence class, with Range of 224 m, 229 m and 282 m respectively, which were more than the sampling scale of 200 m. The spatial prediction results showed that SOC, TN and TP content all appear in large areas of extremely low value in CMB, and its cross validation results showed that using vegetation and terrain factors as covariates in the spatial prediction of SOC, TN and TP can improve the prediction accuracy. The results of correlation analysis showed that the influencing factor for SOC and TN, and TP were NDVI and topographical changes, respectively. In general, vegetation and terrain factors as auxiliary factors can improved the accuracy of soil C-N-P spatial distribution prediction after afforestation in coastal zone.

Electrolyte Analysis in Blood Serum by Laser-Induced Breakdown Spectroscopy Using a Portable Laser.

The fast and reliable analysis of electrolytes such as K, Na, Ca in human blood serum has become an indispensable tool for diagnosing and preventing diseases. Laser-induced breakdown spectroscopy (LIBS) has been demonstrated as a powerful analytical technique on elements. To apply LIBS to the quantitative analysis of electrolyte elements in real time, a self-developed portable laser was used to measure blood serum samples supported by glass slides and filter paper in this work. The partial least squares regression (PLSR) method was employed for predicting the concentrations of K, Na, Ca from serum LIBS spectra. Great prediction accuracies with excellent linearity were obtained for the serum samples, both on glass slides and filter paper. For blood serum on glass slides, the prediction accuracies for K, Na, Ca were 1.45%, 0.61% and 3.80%. Moreover, for blood serum on filter paper, the corresponding prediction accuracies were 7.47%, 1.56% and 0.52%. The results show that LIBS using a portable laser with the assistance of PLSR can be used for accurate quantitative analysis of elements in blood serum in real time. This work reveals that the handheld LIBS instruments will be an excellent tool for real-time clinical practice.

Downregulation of T7 RNA polymerase transcription enhances pET-based recombinant protein production in Escherichia coli BL21 (DE3) by suppressing autolysis.

Escherichia coli BL21 (DE3) is an excellent and widely used host for recombinant protein production. Many variant hosts were developed from BL21 (DE3), but improving the expression of specific proteins remains a major challenge in biotechnology. In this study, we found that when BL21 (DE3) overexpressed glucose dehydrogenase (GDH), a significant industrial enzyme, severe cell autolysis was induced. Subsequently, we observed this phenomenon in the expression of 10 other recombinant proteins. This precludes a further increase of the produced enzyme activity by extending the fermentation time, which is not conducive to the reduction of industrial enzyme production costs. Analysis of membrane structure and messenger RNA expression analysis showed that cells could underwent a form of programmed cell death (PCD) during the autolysis period. However, blocking three known PCD pathways in BL21 (DE3) did not completely alleviate autolysis completely. Consequently, we attempted to develop a strong expression host resistant to autolysis by controlling the speed of recombinant protein expression. To find a more suitable protein expression rate, the high- and low-strength promoter lacUV5 and lac were shuffled and recombined to yield the promoter variants lacUV5-1A and lac-1G. The results showed that only one base in lac promoter needs to be changed, and the A at the +1 position was changed to a G, resulting in the improved host BL21 (DE3-lac1G), which resistant to autolysis. As a consequence, the GDH activity at 43 h was greatly increased from 37.5 to 452.0 U/ml. In scale-up fermentation, the new host was able to produce the model enzyme with a high rate of 89.55 U/ml/h at 43 h, compared to only 3 U/ml/h achieved using BL21 (DE3). Importantly, BL21 (DE3-lac1G) also successfully improved the production of 10 other enzymes. The engineered E. coli strain constructed in this study conveniently optimizes recombinant protein overexpression by suppressing cell autolysis, and shows great potential for industrial applications.

Regulating the T7 RNA polymerase expression in E. coli BL21 (DE3) to provide more host options for recombinant protein production.

Escherichia coli is the most widely used bacterium in prokaryotic expression system for the production of recombinant proteins. In BL21 (DE3), the gene encoding the T7 RNA polymerase (T7 RNAP) is under control of the strong lacUV5 promoter (PlacUV5), which is leakier and more active than wild-type lac promoter (PlacWT) under certain growth conditions. These characteristics are not advantageous for the production of those recombinant proteins with toxic or growth-burdened. On the one hand, leakage expression of T7 RNAP leads to rapid production of target proteins under non-inducing period, which sucks resources away from cellular growth. Moreover, in non-inducing or inducing period, high expression of T7 RNAP production leads to the high-production of hard-to-express proteins, which may all lead to loss of the expression plasmid or the occurrence of mutations in the expressed gene. Therefore, more BL21 (DE3)-derived variant strains with rigorous expression and different expression level of T7 RNAP should be developed. Hence, we replaced PlacUV5 with other inducible promoters respectively, including arabinose promoter (ParaBAD), rhamnose promoter (PrhaBAD), tetracycline promoter (Ptet), in order to optimize the production of recombinant protein by regulating the transcription level and the leakage level of T7 RNAP. Compared with BL21 (DE3), the constructed engineered strains had higher sensitivity to inducers, among which rhamnose and tetracycline promoters had the lowest leakage ability. In the production of glucose dehydrogenase (GDH), a protein that causes host autolysis, the engineered strain BL21 (DE3::ara) exhibited higher biomass, cell survival rate and foreign protein expression level than that of BL21 (DE3). In addition, these engineered strains had been successfully applied to improve the production of membrane proteins, including E. coli cytosine transporter protein (CodB), the E. coli membrane protein insertase/foldase (YidC), and the E. coli F-ATPase subunit b (Ecb). The engineered strains constructed in this paper provided more host choices for the production of recombinant proteins.

Recent advances in the application of multiplex genome editing in Saccharomyces cerevisiae.

Saccharomyces cerevisiae is a widely used microorganism and a greatly popular cell factory for the production of various chemicals. In order to improve the yield of target chemicals, it is often necessary to increase the copy numbers of key genes or engineer the related metabolic pathways, which traditionally required time-consuming repetitive rounds of gene editing. With the development of gene-editing technologies such as meganucleases, TALENs, and the CRISPR/Cas system, multiplex genome editing has entered a period of rapid development to speed up cell factory optimization. Multi-copy insertion and removing bottlenecks in biosynthetic pathways can be achieved through gene integration and knockout, for which multiplexing can be accomplished by targeting repetitive sequences and multiple sites, respectively. Importantly, the development of the CRISPR/Cas system has greatly increased the speed and efficiency of multiplex editing. In this review, the various multiplex genome editing technologies in S. cerevisiae were summarized, and the principles, advantages, and the disadvantages were analyzed and discussed. Finally, the practical applications and future prospects of multiplex genome editing were discussed. KEY POINTS: • The development of multiplex genome editing in S. cerevisiae was summarized. • The pros and cons of various multiplex genome editing technologies are discussed. • Further prospects on the improvement of multiplex genome editing are proposed.

Carbon isotopic measurements from coastal zone protected forests in northern China: Soil carbon decomposition assessment and its influencing factors.

Panting protected forests to increase soil carbon sequestration is an effective means of reducing carbon emissions. Soil organic carbon (SOC) decomposition is one of the main indicators of soil carbon sequestration. However, SOC decomposition and its influencing factors in protected forests have not been fully characterized, especially in coastal zones. In this paper, coastal zone protected forest stands composed of Quercus acutissima Carruth (QAC), Pinus thunbergii Parl (PTP) and mixed PTP and QAC (MF) were selected as the research objects. The trends of the SOC decomposition rate were characterized by the beta (ß) value, and the influencing factors were further explored with structural equation models. The results were as follows: The SOC content decreased from leaf to litter and then to the soil profile at all sites, while the δ13C value increased. The ß value ranged from -3.12 to -5.76, with an average of -3.81. The ß value was positively correlated with the diversity and richness of soil bacteria, supporting the hypothesis that the increase in δ13C with depth was mainly caused by isotope fractionation in the process of microbial SOC decomposition. The structural equation model showed that nitrogen and the availability of nitrogen have a strong ability to explain the value of ß, which indicates that nitrogen-based edaphic variables play an important role in affecting SOC decomposition. The SOC decomposition rate in PTP was higher than that in QAC and MF. The results of this study indicate that the prediction of SOC decomposition based on the ß value is suitable for coastal zone protected forests. The incorporation of edaphic variables into global carbon cycle models may enhance the predictions of SOC dynamics in coastal zone protected forests.

OsEDM2L mediates m6 A of EAT1 transcript for proper alternative splicing and polyadenylation regulating rice tapetal degradation.

N6 -methyladenosine (m6 A) modification affects the post-transcriptional regulation of eukaryotic gene expression, but the underlying mechanisms and their effects in plants remain largely unknown. Here, we report that the N6 -adenine methyltransferase-like domain-containing protein ENHANCED DOWNY MILDEW 2-LIKE (OsEDM2L) is essential for rice (Oryza sativa L.) anther development. The osedm2l knockout mutant showed delayed tapetal programmed cell death (PCD) and defective pollen development. OsEDM2L interacts with the transcription factors basic helix-loop-helix 142 and TAPETUM DEGENERATION RETARDATION to regulate the expression of ETERNAL TAPETUM 1 (EAT1), a positive regulator of tapetal PCD. Mutation of OsEDM2L altered the transcriptomic m6 A landscape, and caused a distinct m6 A modification of the EAT1 transcript leading to dysregulation of its alternative splicing and polyadenylation, followed by suppression of the EAT1 target genes OsAP25 and OsAP37 for tapetal PCD. Therefore, OsEDM2L is indispensable for proper messenger RNA m6 A modification in rice anther development.

Color Tunable Self-Trapped Emissions from Lead-Free All Inorganic IA-IB Bimetallic Halides Cs-Ag-X (X = Cl, Br, I).

Multi-metallic halides of group IA and IB metals are emerged as a new class of color tunable emitters. While chalcogenides and perovskites are extensively studied, these families of materials are little explored. In comparison, herein, lead and cadmium free bimetallic Cs-Ag-X (X = Cl, Br, I) halides are reported where the larger ion Ag+ helped in incorporating all the halide ions which in turn tune their emission color in spanning from 397 nm (violet) to 820 nm (near infrared) as a function of their composition. The synthesis method adopted here is the solvent free ball milling of respective halides of Cs and Ag and took the record shortest time and in bulk scale. From decay lifetimes, emissions from these bimetallic halides are found as a result of fast recombination of self-trapped excitons, which exhibited not only reasonably high quantum yield in the range of 17-68% but also excellent stability to air and moisture under ambient conditions. These also show wide Stokes shift with relatively longer decay lifetimes ranging above the exciton and below the surface trap or dopant induced emissions of inorganic semiconductors, indicating a new class of materials having unique identity of their optical behaviors.

The mitochondrial aldehyde dehydrogenase OsALDH2b negatively regulates tapetum degeneration in rice.

Timely degradation of anther tapetal cells is a prerequisite for normal pollen development in flowering plants. Although several genes involved in tapetum development have been identified, the molecular basis of tapetum degeneration regulation remains poorly understood. In this study, we identified and characterized the nucleus-encoded, conserved mitochondrial aldehyde dehydrogenase OsALDH2b as a key regulator of tapetum degeneration in rice (Oryza sativa). OsALDH2b was highly expressed in anthers from meiosis to the early microspore stage. Mutation of OsALDH2b resulted in excess malonaldehyde accumulation and earlier programmed cell death in the tapetum, leading to premature tapetum degeneration and abnormal microspore development. These results demonstrate that OsALDH2b negatively regulates tapetal programmed cell death and is required for male reproductive development, providing insights into the regulation of tapetum development in plants.

Identification and validation of reference genes for real-time RT-PCR in Aphelenchoides besseyi.

Fragments of four candidate reference genes of Aphelenchoides besseyi, including actin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), ubiquitin conjugating-3 enzyme (UBC) and alpha-tubulin (α-tubulin) were cloned from the transcriptome database of A. besseyi. The expression level of these four candidate reference genes and a commonly used reference gene of A. besseyi (18S rRNA) in three experimental conditions, including the four life stages (female, male, juvenile and egg) of two populations and the mixed-stage nematodes of four populations with different origins and hosts were analyzed by RT-qPCR. The expression stability of the five candidate reference genes under the three experimental conditions was analyzed by ΔCt, geNorm, NormFinder and RefFinder respectively. The analysis results of ΔCt, geNorm, NormFinder and RefFinder all indicated that UBC was the gene with the highest average ranking of stability. In conclusion, the expression stability of UBC was optimal under the three experimental conditions, indicating that UBC could be used as a suitable reference gene instead of 18S rRNA in the RT-qPCR analysis for A. besseyi.

MAFE-Net: retinal vessel segmentation based on a multiple attention-guided fusion mechanism and ensemble learning network.

The precise and automatic recognition of retinal vessels is of utmost importance in the prevention, diagnosis and assessment of certain eye diseases, yet it brings a nontrivial uncertainty for this challenging detection mission due to the presence of intricate factors, such as uneven and indistinct curvilinear shapes, unpredictable pathological deformations, and non-uniform contrast. Therefore, we propose a unique and practical approach based on a multiple attention-guided fusion mechanism and ensemble learning network (MAFE-Net) for retinal vessel segmentation. In conventional UNet-based models, long-distance dependencies are explicitly modeled, which may cause partial scene information loss. To compensate for the deficiency, various blood vessel features can be extracted from retinal images by using an attention-guided fusion module. In the skip connection part, a unique spatial attention module is applied to remove redundant and irrelevant information; this structure helps to better integrate low-level and high-level features. The final step involves a DropOut layer that removes some neurons randomly to prevent overfitting and improve generalization. Moreover, an ensemble learning framework is designed to detect retinal vessels by combining different deep learning models. To demonstrate the effectiveness of the proposed model, experimental results were verified in public datasets STARE, DRIVE, and CHASEDB1, which achieved F1 scores of 0.842, 0.825, and 0.814, and Accuracy values of 0.975, 0.969, and 0.975, respectively. Compared with eight state-of-the-art models, the designed model produces satisfactory results both visually and quantitatively.