Genome-wide molecular evolution analysis of the GRF and GIF gene families in Plantae (Archaeplastida).

BACKGROUND: Plant growth-regulating factors (GRFs) and GRF-interacting factors (GIFs) interact with each other and collectively have important regulatory roles in plant growth, development, and stress responses. Therefore, it is of great significance to explore the systematic evolution of GRF and GIF gene families. However, our knowledge and understanding of the role of GRF and GIF genes during plant evolution has been fragmentary. RESULTS: In this study, a large number of genomic and transcriptomic datasets of algae, mosses, ferns, gymnosperms and angiosperms were used to systematically analyze the evolution of GRF and GIF genes during the evolution of plants. The results showed that GRF gene first appeared in the charophyte Klebsormidium nitens, whereas the GIF genes originated relatively early, and these two gene families were mainly expanded by segmental duplication events after plant terrestrialization. During the process of evolution, the protein sequences and functions of GRF and GIF family genes are relatively conservative. As cooperative partner, GRF and GIF genes contain the similar types of cis-acting elements in their promoter regions, which enables them to have similar transcriptional response patterns, and both show higher levels of expression in reproductive organs and tissues and organs with strong capacity for cell division. Based on protein-protein interaction analysis and verification, we found that the GRF-GIF protein partnership began to be established in pteridophytes and is highly conserved across different terrestrial plants. CONCLUSIONS: These results provide a foundation for further exploration of the molecular evolution and biological functions of GRF and GIF genes.

Improving Lightweight AdderNet via Distillation From ℓ2 to ℓ1-norm.

To achieve efficient inference with a hardware-friendly design, Adder Neural Networks (ANNs) are proposed to replace expensive multiplication operations in Convolutional Neural Networks (CNNs) with cheap additions through utilizing l1 -norm for similarity measurement instead of cosine distance. However, we observe that there exists an increasing gap between CNNs and ANNs with reducing parameters, which cannot be eliminated by existing algorithms. In this paper, we present a simple yet effective Norm-Guided Distillation (NGD) method for l1 -norm ANNs to learn superior performance from l2 -norm ANNs. Although CNNs achieve similar accuracy with l2 -norm ANNs, the clustering performance based on l2 -distance can be easily learned by l1 -norm ANNs compared with cross correlation in CNNs. The features in l2 -norm ANNs are encouraged to achieve intra-class centralization and inter-class decentralization to amplify this advantage. Furthermore, the roughly estimated gradients in vanilla ANNs are modified to a progressive approximation from l2 -norm to l1 -norm so that a more accurate optimization can be achieved. Extensive evaluations on several benchmarks demonstrate the effectiveness of NGD on lightweight networks. For example, our method improves ANN by 10.43% with 0.25× GhostNet on CIFAR-100 and 3.1% with 1.0× GhostNet on ImageNet.

A study on puerarin in situ gel eye drops: Formulation optimization and pharmacokinetics on rabbits by microdialysis.

Puerarin (PUE), an isoflavonoid isolated from Pueraria lobata (Willd) Ohwi root, is a ß-adrenergic receptor inhibitor used in treating glaucoma. The concentration range of gellan gum was determined based on the formulation viscosity and gelling capacity. PVP-K30 and gellan gum were used as variables, with the viscosity of formulation: STF = 40: 21, the 4 h permeation rate of rabbit isolated sclera, and 2 h in vitro release rate as response values. The JMP software was used to optimize the results, presenting that gellan gum was the main factor influencing viscosity. The in vitro release and permeation rate were primarily influenced by PVP-K30. The optimal prescription was 0.45% gellan gum and 6.0% PVP-K30. The in vitro release and permeation characteristics of puerarin in situ gel (PUE-ISG) were investigated using PUE solution as a control. The dialysis bag method results indicated that the release of the solution group leveled off after 4 h, while the PUE-ISG group had been continuously releasing. However, the cumulative release rates of the two were no longer significantly different at 10 h. The cumulative permeation rates of the ISG and solution groups were not significantly different (P > 0.05) in the rabbit isolated sclera. The apparent permeability Papp and steady-state flux Jss of PUE-ISG were 0.950 ± 0.059 cm/h and 9.504 ± 0.587 mg·(cm·h), respectively. A sensitive and stable HPLC-MS/MS analytical method for quantifying aqueous humor concentrations of PUE was validated. A microdialysis technique was successfully used in the aqueous humor pharmacokinetics study to sample aqueous humor from rabbit eye continuously. The results revealed that PUE-ISG significantly increased the drug concentration in the aqueous humor, with Cmax and AUC(0-t) 3.77 and 4.40 times higher than those of the solution group, respectively. Tmax was also significantly prolonged, indicating good prospects for clinical application. The developed PUE-ISG preparation has the characteristics of rapid drug release and sustained permeation, and increase the drug concentration in aqueous humor, with all inactive ingredients remaining within the maximum allowable limits recommended by the FDA guideline.

Preparation and Characterization of Ion-Sensitive Brimonidine Tartrate In Situ Gel for Ocular Delivery.

Brimonidine tartrate (BRT) is a highly selective α2 adrenergic receptor agonist as treatment for patients with open angle glaucoma and high intraocular pressure. The objective of this study was to formulate an ophthalmic ion-sensitive in situ gel (ISG) of BRT to increase the retention time of the drug and its bioavailability. The optimum formulation of 2 mg/mL BRT-ISG was obtained with 0.45% gellan gum as the gel matrix. In vitro release results showed that the water-soluble drug bromonidine tartrate in ocular in situ gels exhibited a high burst effect and fast release in solution. The results of dialysis membrane permeation showed that there was a significant difference between the commercially available and BRT-ISG groups after 45 min. The results of the pre-corneal retention study indicated that gellan gum can effectively prolong ocular surface retention. Preliminary stability results showed that it should be stored in a cool and dark place, and the formulation under long-term preservation can be basically stable. The pharmacokinetic study of the BRT-ISG in the anterior chamber of the rabbit eye was studied by microdialysis technique, and microdialysis samples were analyzed by LC-MS/MS. The pharmacokinetic study showed that the BRT-ISG reached Cmax (8.16 mg/L) at 93 min after administration, which was 2.7 times that of the BRT eye drops, and the AUC(0-t) (1397.08 mg·min/L) was 3.4 times that of the BRT eye drops. The optimal prescription can prolong the retention time of BRT in front of the cornea and significantly improve the bioavailability of BRT in the eye. Combined with the results of in vitro release, permeation and pre-corneal retention studies, the improvement of BRT-ISG bioavailability in rabbit eyes was found to be mainly due to the retention effect after the mixture of ISG and tears.

A Survey on Vision Transformer.

Transformer, first applied to the field of natural language processing, is a type of deep neural network mainly based on the self-attention mechanism. Thanks to its strong representation capabilities, researchers are looking at ways to apply transformer to computer vision tasks. In a variety of visual benchmarks, transformer-based models perform similar to or better than other types of networks such as convolutional and recurrent neural networks. Given its high performance and less need for vision-specific inductive bias, transformer is receiving more and more attention from the computer vision community. In this paper, we review these vision transformer models by categorizing them in different tasks and analyzing their advantages and disadvantages. The main categories we explore include the backbone network, high/mid-level vision, low-level vision, and video processing. We also include efficient transformer methods for pushing transformer into real device-based applications. Furthermore, we also take a brief look at the self-attention mechanism in computer vision, as it is the base component in transformer. Toward the end of this paper, we discuss the challenges and provide several further research directions for vision transformers.

Correlation Analysis of the Bacterial Community and Wood Properties of Populus × euramericana cv. "74/76" Wet Heartwood.

Wetwood disease of poplar limits the processing and manufacturing of poplar, and the pathogenic bacteria of wet heartwood are poorly known. We used high-throughput sequencing methods to analyze the bacterial community of the heartwood, sapwood, root tissue, and rhizosphere soil of Populus × euramericana cv. "74/76" (poplar 107) in wetwood trees and healthy trees to explore the cause of poplar wetwood disease. Bacterial diversity and community structure were analyzed, and the correlation between wood properties and bacterial relative abundance was analyzed to explore their relationship. Two alpha-diversity indices of endophytic bacteria in the heartwood of wetwood trees were significantly (p < 0.05) lower than that in the heartwood of healthy trees, and the community structure between the two types of trees was significantly different. No significant differences in the alpha-diversity indices nor community structure were observed in the sapwood, root tissue, or rhizosphere bacterial community of diseased and healthy trees. The distribution of dominant bacteria genus in the heartwood of diseased and healthy trees differed. Proteiniphilum, Actinotalea, and Methanobacterium were the dominant genera in diseased trees' heartwood. Proteiniphilum, Dysgonomonas, and Bacteroides were the dominant genera in healthy trees' heartwood. The relative abundance of Proteiniphilum, Actinotalea, and Methanobacterium was significantly higher in the heartwood of wetwood trees than those of healthy trees. A db-RDA analysis found that these three bacterial genera were positively correlated with the rate of wet heartwood. These three bacterial genera may be the main pathogens causing poplar wetwood disease.

A Reactive Superhydrophobic Platform for Living Photolithography.

Superhydrophobic surfaces with regional functions have widespread applications in biotechnology, diagnostic applications, and micro-chemical synthesis and analysis. However, owing to their chemical inertness, superhydrophobic surfaces with chemical reactivity are difficult to achieve. Superhydrophobic surfaces that can be further modified with varied densities and expanded species of the functional moieties are not readily available. In this study, a single-step approach to achieve a reactive superhydrophobic surface is reported, on which chemical grafting of a library of molecules can be carried out through surface-initiated atom-transfer radical addition or surface-initiated atom-transfer radical polymerization. The excellent spatial and temporal controllability of these chemical processes under visible light enables us to take advantage of programmed liquid-crystal-display (LCD) or Digital Light Processing (DLP) photolithography systems to effortlessly regulate the location, density, and species of the functional molecules on the reactive superhydrophobic surface. The distinctive properties of this surface will provide new insight into intelligent superhydrophobic material development and practical applications, such as aqueous/oil microdroplets array, multi-anti-counterfeiting labels and integrated microfluidic reactors with enzymes for chemical logic learning.

Probing 1D convolutional neural network adapted to near-infrared spectroscopy for efficient classification of mixed fish.

Salmon and Cod are economically significant world-class fish that have high economic value. It is difficult to accurately sort and process them by appearance during harvest and transportation. Conventional chemical detection means are time-consuming and costly, which greatly affects the cost and efficiency of Fishery production. Therefore, there is an urgent need for smart Fisheries methods which use for the classification of mixed fish. In this paper, near-infrared spectroscopy (NIRS) was used to assess salmon and cod samples. This study aims to evaluate feasibility of a back-propagation neural network (BPNN) and a convolutional neural network (CNN) for identifying different species of fishes by the corresponding spectra in comparison to traditional chemometrics Partial Least Squares. After comparing the effects of different batch sizes, number of convolutional kernels, number of convolutional layers, and number of pooling layers on the classification of NIRS spectra comparing different structures of one-dimensional (1D)-CNN, we propose the 1D-CNN-8 model that is most suitable for the classification of mixed fish. Compared with the results of traditional chemometrics methods and BPNN, the prediction model of the 1D-CNN model can reach 98.00% Accuracy and the parameters are significantly better than others. Meanwhile, the parameters and floating-point operations of the optimal model are both small. Therefore, the improved CNN model based on the NIRS can effectively and quickly identify different kinds of fish samples and contribute to realizing edge computing at the same time.

Exploration of compressive sensing in the classification of frozen fish based on two-dimensional correlation spectrum.

In order to classify imported frozen fish, effectively a spectral data compression method was presented based on two-dimensional correlation spectroscopy. In the experiment, the near-infrared spectral data of Oncorhynchus keta, Oncorhynchus nerka and Oncorhynchus gorbuscha of Salmonidae were collected. And two-dimensional correlation spectroscopy among the three fish samples was constructed. The study found that the auto-correlation peaks intensities at 650 nm, 1724 nm and 1908 nm were almost zero, which were taken as the separation point of the spectra. Therefore, each spectral data is divided into 4 segments and the integral of each segment is obtained. The original spectra of 201 points in each group were compressed into 4 points. Then, the compressed spectral data were input into the support vector machine to establish the discriminant model of three kinds of frozen fish. At the same time, the Competitive Adaptive Reweighted Sampling and the Successive Projections Algorithm were used to screen the original spectra. The classification results were compared with the result of the spectral data compression method of two-dimensional correlation spectroscopy. The result shows: the compression rate of the proposed method is 98.01%; the accuracy rate of support vector machine training set is 100%; the accuracy rate of validation set is up to 100%. The results shows that the proposed spectral data compression method based on two-dimensional correlation spectral technology has high compression rate and accurate classification.

Combined transcriptome and proteome analysis provides insights into anthocyanin accumulation in the leaves of red-leaved poplars.

KEY MESSAGE: Anthocyanin was highly accumulated in the leaves of red-leaved poplars; Many structural genes involved in anthocyanin synthesis were significantly up-regulated in 'Quanhong' and 'Xuanhong'; TTG2, HYH, and HY5 may be directly involved in the regulation of anthocyanin synthesis in both red-leaved poplars. The red-leaved poplar cultivars 'Quanhong' and 'Xuanhong' are bud mutations of Populus deltoides cv. 'Zhonglin 2025'. These cultivars are valued for their beautiful shape, lack of flying catkins, and ornamental leaf colors. However, the understanding of the molecular mechanism of anthocyanin accumulation in the leaves of red-leaved poplars is still unclear. Here, we profiled the changes of pigment content, transcriptome and proteome expression in the leaves of three poplar cultivars and the results showed that the ratios of anthocyanin to total chlorophyll in both red-leaved poplars were higher than that in 'Zhonglin 2025', indicating that the anthocyanin was highly accumulated in the leaves of red-leaved poplars. Based on the results of combined transcriptome and proteome analysis, 15 and 11 differentially expressed genes/proteins involved in anthocyanin synthesis were screened in 'Quanhong' and 'Xuanhong', respectively, indicating that the two red-leaved poplar cultivars have slightly different patterns of regulating anthocyanin biosynthesis. Among the 120 transcription factors, 3 (HY5, HYH, and TTG2), may be directly involved in the regulation of anthocyanin synthesis in both red-leaved poplars. This study screens the candidate genes involved in anthocyanin accumulation in the leaves of red-leaved poplars and lays a foundation for further exploring the molecular mechanism of leaf red coloration in red-leaved poplars.

Whole-genome resequencing using next-generation and Nanopore sequencing for molecular characterization of T-DNA integration in transgenic poplar 741.

BACKGROUND: The molecular characterization information of T-DNA integration is not only required by public risk assessors and regulators, but is also closely related to the expression of exogenous and endogenous genes. At present, with the development of sequencing technology, whole-genome resequencing has become an attractive approach to identify unknown genetically modified events and characterise T-DNA integration events. RESULTS: In this study, we performed genome resequencing of Pb29, a transgenic high-resistance poplar 741 line that has been commercialized, using next-generation and Nanopore sequencing. The results revealed that there are two T-DNA insertion sites, located at 9,283,905-9,283,937 bp on chromosome 3 (Chr03) and 10,868,777-10,868,803 bp on Chr10. The accuracy of the T-DNA insertion locations and directions was verified using polymerase chain reaction amplification. Through sequence alignment, different degrees of base deletions were detected on the T-DNA left and right border sequences, and in the flanking sequences of the insertion sites. An unknown fragment was inserted between the Chr03 insertion site and the right flanking sequence, but the Pb29 genome did not undergo chromosomal rearrangement. It is worth noting that we did not detect the API gene in the Pb29 genome, indicating that Pb29 is a transgenic line containing only the BtCry1AC gene. On Chr03, the insertion of T-DNA disrupted a gene encoding TAF12 protein, but the transcriptional abundance of this gene did not change significantly in the leaves of Pb29. Additionally, except for the gene located closest to the T-DNA integration site, the expression levels of four other neighboring genes did not change significantly in the leaves of Pb29. CONCLUSIONS: This study provides molecular characterization information of T-DNA integration in transgenic poplar 741 line Pb29, which contribute to safety supervision and further extensive commercial planting of transgenic poplar 741.

Effects of different types and doses of whey protein on the physiological and intestinal flora in D-galactose induced aging mice.

The elderly usually suffer from many diseases. Improving the quality of life of the elderly is an urgent social issue. In this present study, D-galactose treated aging mice models were used to reveal the effects of different animal sources and different doses of whey protein (WP) on the immune indexes organs and intestinal flora. A total of 9 groups were set up, including normal control (NC), negative control (NS), positive control (Vc), low-, medium- and high-doses of cow WP intervention groups (CL, CM and CH for short, correspondingly) and low-, medium- and high-doses of goat WP intervention groups (GL, GM and GH for short, correspondingly). The body weight gain, thymus/body weight ratio, superoxide dismutase (SOD) activity, malondialdehyde (MDA) content, spleen immunoglobulins G (IgG), spleen interleukin-2 (IL-2) and spleen interleukin-2 (IL-6) were measured. Then, the intestinal contents were collected, and 16s genes of intestinal bacteria were sequenced to reveal the changes in bacterial flora structure. WP intervention significantly increased the weight gain, thymus/body ratio and SOD activity, but decrease the content of MDA. WP intervention increased some immune indicators. All the WP treated aging mice showed similar values of physiological indexes to that of the Vc group, even better. The relative abundance of Lactobacillus and Stenotrophomonas was increased and decreased, respectively, by both cow and goat WP. Lactobacillus may be involved in regulating the functional repair of organisms. In contrast, Stenotrophomonas might play a negative role in the immune and antioxidant capacity of the body. Combining physiological indicators and intestinal flora structure, low-concentration WP for cow and goat might be optimal for aging models.

Bi-Stream Pose-Guided Region Ensemble Network for Fingertip Localization From Stereo Images.

In human-computer interaction, it is important to accurately estimate the hand pose, especially fingertips. However, traditional approaches to fingertip localization mainly rely on depth images and thus suffer considerably from noise and missing values. Instead of depth images, stereo images can also provide 3-D information of hands. There are nevertheless limitations on the dataset size, global viewpoints, hand articulations, and hand shapes in publicly available stereo-based hand pose datasets. To mitigate these limitations and promote further research on hand pose estimation from stereo images, we build a new large-scale binocular hand pose dataset called THU-Bi-Hand, offering a new perspective for fingertip localization. In the THU-Bi-Hand dataset, there are 447k pairs of stereo images of different hand shapes from ten subjects with accurate 3-D location annotations of the wrist and five fingertips. Captured with minimal restriction on the range of hand motion, the dataset covers a large global viewpoint space and hand articulation space. To better present the performance of fingertip localization on THU-Bi-Hand, we propose a novel scheme termed bi-stream pose-guided region ensemble network (Bi-Pose-REN). It extracts more representative feature regions around joints in the feature maps under the guidance of the previously estimated pose. The feature regions are integrated hierarchically according to the topology of hand joints to regress a refined hand pose. Bi-Pose-REN and several existing methods are evaluated on THU-Bi-Hand so that benchmarks are provided for further research. Experimental results show that our Bi-Pose-REN has achieved the best performance on THU-Bi-Hand.

MFA-Net: Motion Feature Augmented Network for Dynamic Hand Gesture Recognition from Skeletal Data.

Dynamic hand gesture recognition has attracted increasing attention because of its importance for human⁻computer interaction. In this paper, we propose a novel motion feature augmented network (MFA-Net) for dynamic hand gesture recognition from skeletal data. MFA-Net exploits motion features of finger and global movements to augment features of deep network for gesture recognition. To describe finger articulated movements, finger motion features are extracted from the hand skeleton sequence via a variational autoencoder. Global motion features are utilized to represent the global movements of hand skeleton. These motion features along with the skeleton sequence are then fed into three branches of a recurrent neural network (RNN), which augment the motion features for RNN and improve the classification performance. The proposed MFA-Net is evaluated on two challenging skeleton-based dynamic hand gesture datasets, including DHG-14/28 dataset and SHREC'17 dataset. Experimental results demonstrate that our proposed method achieves comparable performance on DHG-14/28 dataset and better performance on SHREC'17 dataset when compared with start-of-the-art methods.