Nicotinamide Mononucleotide improves oocyte maturation of mice with type 1 diabetes.

BACKGROUND: The number of patients with type 1 diabetes rises rapidly around the world in recent years. Maternal diabetes has a detrimental effect on reproductive outcomes due to decreased oocyte quality. However, the strategies to improve the oocyte quality and artificial reproductive technology (ART) efficiency of infertile females suffering from diabetes have not been fully studied. In this study, we aimed to examine the effects of nicotinamide mononucleotide (NMN) on oocyte maturation of mouse with type 1 diabetes mouse and explore the underlying mechanisms of NMN's effect. METHODS: Streptozotocin (STZ) was used to establish the mouse models with type 1 diabetes. The successful establishment of the models was confirmed by the results of body weight test, fasting blood glucose test and haematoxylin and eosin (H&E) staining. The in vitro maturation (IVM) rate of oocytes from diabetic mice was examined. Immunofluorescence staining (IF) was performed to examine the reactive oxygen species (ROS) level, spindle/chromosome structure, mitochondrial function, actin dynamics, DNA damage and histone modification of oocytes, which are potential factors affecting the oocyte quality. The quantitative reverse transcription PCR (RT-qPCR) was used to detect the mRNA levels of Sod1, Opa1, Mfn2, Drp1, Sirt1 and Sirt3 in oocytes. RESULTS: The NMN supplementation increased the oocyte maturation rate of the mice with diabetes. Furthermore, NMN supplementation improved the oocyte quality by rescuing the actin dynamics, reversing meiotic defects, improving the mitochondrial function, reducing ROS level, suppressing DNA damage and restoring changes in histone modifications of oocytes collected from the mice with diabetes. CONCLUSION: NMN could improve the maturation rate and quality of oocytes in STZ-induced diabetic mice, which provides a significant clue for the treatment of infertility of the patients with diabetes.

Metagenomics reveals the variations in functional metabolism associated with greenhouse gas emissions during legume-vegetable rotation process.

Legume-based rotation is commonly recognized for its mitigation efficiency of greenhouse gas (GHG) emissions. However, variations in GHG emission-associated metabolic functions during the legume-vegetable rotation process remain largely uncharacterized. Accordingly, a soybean-radish rotation field experiment was designed to clarify the responses of microbial communities and their GHG emission-associated functional metabolism through metagenomics. The results showed that the contents of soil organic carbon and total phosphorus significantly decreased during the soybean-radish process (P < 0.05), while soil total potassium content and bacterial richness and diversity significantly increased (P < 0.05). Moreover, the predominant bacterial phyla varied, with a decrease in the relative abundance of Proteobacteria and an increase in the relative abundance of Acidobacteria, Gemmatimonadetes, and Chloroflexi. Metagenomics clarified that bacterial carbohydrate metabolism substantially increased during the rotation process, whereas formaldehyde assimilation, methanogenesis, nitrification, and dissimilatory nitrate reduction decreased (P < 0.05). Specifically, the expression of phosphate acetyltransferase (functional methanogenesis gene, pta) and nitrate reductase gamma subunit (functional dissimilatory nitrate reduction gene, narI) was inhibited, indicating of low methane production and nitrogen metabolism. Additionally, the partial least squares path model revealed that the Shannon diversity index was negatively correlated with methane and nitrogen metabolism (P < 0.01), further demonstrating that the response of the soil bacterial microbiome responses are closely linked with GHG-associated metabolism during the soybean-radish rotation process. Collectively, our findings shed light on the responses of soil microbial communities to functional metabolism associated with GHG emissions and provide important insights to mitigate GHG emissions during the rotational cropping of legumes and vegetables.

E-nose and colorimetric sensor array combining homologous data fusion strategy discriminating the roasting degree of large-leaf yellow tea.

Different degrees of roasting result in differences in the quality and flavor of large-leaf yellow tea. The current sensory evaluation and chemical detection methods cannot meet the requirement of online differentiation of LYT roasting degree, so an accurate and comprehensive assessment method needs to be developed urgently. First, the two aroma sensing technologies were compared. Two variable screening methods and three recognition algorithms were employed to build discriminant models. The results showed that the discrimination rate of the colorimetric sensor array (CSA) in the prediction set reached 91.89 %, outperforming that of the E-nose. Subsequently, three fusion strategies were applied to improve the discrimination accuracy. The discrimination rate of the middle fusion strategy resulted in an optimal resolution of 94.59 %. The results obtained from the homologous fusion were able to evaluate the roasting degree comprehensively and accurately, which provides a new method and idea for tea aroma quality.

Flex-DLD: Deep Low-Rank Decomposition Model With Flexible Priors for Hyperspectral Image Denoising and Restoration.

Hyperspectral images (HSIs) are composed of hundreds of contiguous waveband images, offering a wealth of spatial and spectral information. However, the practical use of HSIs is often hindered by the presence of complicated noise caused by various factors such as non-uniform sensor response and dark current. Traditional methods for denoising HSIs rely on constrained optimization approaches, where selecting appropriate prior knowledge is critical for achieving satisfactory results. Nevertheless, these traditional algorithms are limited by hand-crafted priors, leaving room for improvement in their denoising performance. Recently, the supervised deep learning technique has emerged as a promising approach for HSI denoising. However, their requirement for paired training data and poor generalization ability on untrained noise distributions pose challenges in practical applications. In this paper, we design a novel algorithm by the synergism of optimization-based methods and deep learning techniques. Specifically, we introduce a plug-and-play Deep Low-rank Decomposition (DLD) model into the optimization framework. Furthermore, we propose an effective mechanism to incorporate traditional prior knowledge into the DLD model. Finally, we provide a detailed analysis of the optimization process and convergence of the proposed method. Empirical evaluations on various tasks, including hyperspectral image denoising and spectral compressive imaging, demonstrate the superiority of our approach over state-of-the-art methods.

Molecular Basis for Inhibition of Heparanases and ß-Glucuronidases by Siastatin B.

Siastatin B is a potent and effective iminosugar inhibitor of three diverse glycosidase classes, namely, sialidases, ß-d-glucuronidases, and N-acetyl-glucosaminidases. The mode of inhibition of glucuronidases, in contrast to sialidases, has long been enigmatic as siastatin B appears too bulky and incorrectly substituted to be accommodated within a ß-d-glucuronidase active site pocket. Herein, we show through crystallographic analysis of protein-inhibitor complexes that siastatin B generates both a hemiaminal and a 3-geminal diol iminosugar (3-GDI) that are, rather than the parent compound, directly responsible for enzyme inhibition. The hemiaminal product is the first observation of a natural product that belongs to the noeuromycin class of inhibitors. Additionally, the 3-GDI represents a new and potent class of the iminosugar glycosidase inhibitor. To substantiate our findings, we synthesized both the gluco- and galacto-configured 3-GDIs and characterized their binding both structurally and kinetically to exo-ß-d-glucuronidases and the anticancer target human heparanase. This revealed submicromolar inhibition of exo-ß-d-glucuronidases and an unprecedented binding mode by this new class of inhibitor. Our results reveal the mechanism by which siastatin B acts as a broad-spectrum glycosidase inhibitor, identify a new class of glycosidase inhibitor, and suggest new functionalities that can be incorporated into future generations of glycosidase inhibitors.

Rapid and comprehensive grade evaluation of Keemun black tea using efficient multidimensional data fusion.

To develop a comprehensive evaluation method for Keemun black tea, we used micro-near-infrared spectroscopy, computer vision, and colorimetric sensor array to collect data. We used support vector machine, least-squares support vector machine (LS-SVM), extreme learning machine, and partial least squares discriminant analysis algorithms to qualitatively discriminate between different grades of tea. Our results indicated that the LS-SVM model with mid-level data fusion attained an accuracy of 98.57% in the testing set. To quantitatively determine flavour substances in black tea, we used support vector regression. The correlation coefficient for the predicted sets of gallic acid, caffeine, epigallocatechin, catechin, epigallocatechin gallate, epicatechin, gallocatechin gallate and total catechins were 0.84089, 0.94249, 0.94050, 0.83820, 0.81111, 0.82670, 0.93230, and 0.93608, respectively. Furthermore, all compounds exhibited residual predictive deviation values exceeding 2. Hence, combining spectral, shape, colour, and aroma data with mid-level data can provide a rapid and comprehensive assessment of Keemun black tea quality.

ß-nicotinamide mononucleotide rescues the quality of aged oocyte and improves subsequent embryo development in pigs.

Oocyte senescence alters the shape and function, thereby weakening the fertilization potential. Nicotinamide mononucleotide (NMN) reverses age-related dysfunctions in various organs. Studies had shown long-term administration of NMN reduced the physiological decline associated in aged mice and reversed the aging of the ovaries. However, the protective effect of NMN on aged porcine oocytes is still unclear. In this study, we investigated the effects of NMN on aging porcine oocytes and subsequent embryonic development. We established a model of senescence of porcine oocytes after ovulation by extending the culture time in vitro. NMN supplementation significantly reduced reactive oxygen species (ROS) levels in senescence oocytes and increased the mRNA levels of antioxidant genes SOD1 and Cat. The mitochondrial membrane potential of aged oocytes treated with NMN was increased compared with that of untreated oocytes. In addition, the mRNA level of apoptosis-related gene Bax was significantly decreased in senescence oocytes treated with NMN, while the mRNA level of anti-apoptosis-related gene BCL-2 was significantly increased. Furthermore, NMN supplementation enhanced the subsequent development ability of senescent oocytes during in vitro aging. Compared with untreated senescent oocytes, the blastocyst formation rate and pluripotent genes of senescent oocytes treated with NMN were significantly increased. Taken together, these results suggest that NMN is beneficial for delaying the aging process in porcine oocytes.

Tracking progress towards equitable maternal and child health in Yunnan: a systematic assessment for the Health Programme for Poverty Alleviation in China during 2015-2020.

OBJECTIVES: To inform the impacts of health programmes which aimed at preventing women and children from being trapped in or returning to poverty because of illness in Yunnan, the main battlefield against poverty in China. DESIGN: The longitudinal comparative evaluation design. DATA COLLECTION AND ANALYSIS: National and Yunnan policy documents related to maternal and child health programmes for poverty alleviation during 2015-2020 were analysed. The changes in disparities in maternal and child health system inputs, service coverage, and health outcomes between poor and non-poor areas, as well as out-of-pocket payments between poor and non-poor populations were assessed before and after 2017. RESULTS: In total 12 policies and 15 programmes related to poverty alleviation for poor women and children in Yunnan were summarised. As a result of health system strengthening in Yunnan, the densities of licensed doctors, nurses, obstetricians, midwives, township health workers and female village doctors had been increased substantially in poor areas, with the annual rates of 14.3%, 22.5%, 21.8%, 23.9%, 14.1% and 7.1% separately. Although disparities existed in some of service coverage between poor and non-poor areas, the health programmes had narrowed the gaps in utilisation of facility birth, caesarean section, prenatal screening and newborn screening across Yunnan (p<0.01). The out-of-pocket payments for inpatient care for serious illnesses among women and children with poverty registration had been considerably decreased to 10.0%. Paralleling the universal coverage, maternal deaths per 100 000 livebirths and child deaths per 1000 livebirths had further declined in both poor and non-poor areas, and the impacts of health programmes on closing the gaps in child survivals across Yunnan were significant (p<0.01). CONCLUSIONS: Remarkable progress in equitable maternal and child survival has been achieved in Yunnan. The practices in Yunnan have shown the Chinese model in ending poverty by strengthening health system and implementing universal coverage with firm commitment, determined leadership, detailed blueprint and social participation.

Source-specific probabilistic risk evaluation of potentially toxic metal(loid)s in fine dust of college campuses based on positive matrix factorization and Monte Carlo simulation.

Contamination, hazard level and source of 10 widely concerned potentially toxic metal(loid)s (PTMs) Co, As, Pb, Cr, Cu, Zn, Ni, Mn, Ba, and V in fine dust with particle size below 63 µm (FD63) were investigated to assess the environmental quality of college campuses and influencing factors. PTMs sources were qualitatively analyzed using statistical methods and quantitatively apportioned using positive matrix factorization. Probabilistic contamination degrees of PTMs were evaluated using enrichment factor and Nemerow integrated enrichment factor. Eco-health risk levels of content-oriented and source-oriented for PTMs were evaluated using Monte Carlo simulation. Mean levels of Zn (643.8 mg kg-1), Pb (146.0 mg kg-1), Cr (145.9 mg kg-1), Cu (95.5 mg kg-1), and Ba (804.2 mg kg-1) in FD63 were significantly larger than soil background values. The possible sources of the concerned PTMs in FD63 were traffic non-exhaust emissions, natural source, mixed source (auto repair waste, paints and pigments) and traffic exhaust emissions, which accounted for 45.7%, 25.4%, 14.5% and 14.4% of total PTMs contents, respectively. Comprehensive contamination levels of PTMs were very high, mainly caused by Zn pollution and non-exhaust emissions. Combined ecological risk levels of PTMs were low and moderate, chiefly caused by Pb and traffic exhaust emissions. The non-cancer risks of the PTMs in FD63 to college students fell within safety level, while the carcinogenic PTMs in FD63 had a certain cancer risks to college students. The results of source-specific health risk assessment indicated that Cr and As were the priority PTMs, and the mixed source was the priority pollution source of PTMs in FD63 from college campuses, which should be paid attention to by the local government.

Epigenetic reprogramming during the maternal-to-zygotic transition.

After fertilization, sperm and oocyte fused and gave rise to a zygote which is the beginning of a new life. Then the embryonic development is monitored and regulated precisely from the transition of oocyte to the embryo at the early stage of embryogenesis, and this process is termed maternal-to-zygotic transition (MZT). MZT involves two major events that are maternal components degradation and zygotic genome activation. The epigenetic reprogramming plays crucial roles in regulating the process of MZT and supervising the normal development of early development of embryos. In recent years, benefited from the rapid development of low-input epigenome profiling technologies, new epigenetic modifications are found to be reprogrammed dramatically and may play different roles during MZT whose dysregulation will cause an abnormal development of embryos even abortion at various stages. In this review, we summarized and discussed the important novel findings on epigenetic reprogramming and the underlying molecular mechanisms regulating MZT in mammalian embryos. Our work provided comprehensive and detailed references for the in deep understanding of epigenetic regulatory network in this key biological process and also shed light on the critical roles for epigenetic reprogramming on embryonic failure during artificial reproductive technology and nature fertilization.

Current progress of pig models for liver cancer research.

Preclinical trials play critical roles in assessing the safety and efficiency of novel therapeutic strategies for human diseases including live cancer. However, most therapeutic strategies that were proved to be effective in preclinical cancer models failed in human clinical trials due to the lack of appropriate disease animal models. Therefore, it is of importance and urgent to develop a precise animal model for preclinical cancer research. Liver cancer is one of the most frequently diagnosed cancers with low 5-year survival rate. Recently, porcine attracted increasing attentions as animal model in biomedical research. Porcine liver cancer model may provide a promising platform for biomedical research due to their similarities to human being in body size, anatomical characteristics, physiology and pathophysiology. In this review, we comprehensively summarized and discussed the advantages and disadvantages, rationale, current status and progress of pig models for liver cancer research.

Research progress of biomarkers in the prediction of anti-PD-1/PD-L1 immunotherapeutic efficiency in lung cancer.

Currently, anti-PD-1/PD-L1 immunotherapy using immune checkpoint inhibitors is widely used in the treatment of multiple cancer types including lung cancer, which is a leading cause of cancer death in the world. However, only a limited proportion of lung cancer patients will benefit from anti-PD-1/PD-L1 therapy. Therefore, it is of importance to predict the response to immunotherapy for the precision treatment of patients. Although the expression of PD-L1 and tumor mutation burden (TMB) are commonly used to predict the clinical response of anti-PD-1/PD-L1 therapy, other factors such as tumor-specific genes, dMMR/MSI, and gut microbiome are also promising predictors for immunotherapy in lung cancer. Furthermore, invasive peripheral blood biomarkers including blood DNA-related biomarkers (e.g., ctDNA and bTMB), blood cell-related biomarkers (e.g., immune cells and TCR), and other blood-related biomarkers (e.g., soluble PD-L1 and cytokines) were utilized to predict the immunotherapeutic response. In this review, the current achievements of anti-PD-1/PD-L1 therapy and the potential biomarkers for the prediction of anti-PD-1/PD-L1 immunotherapy in lung cancer treatment were summarized and discussed.

Cathepsin B S-nitrosylation promotes ADAR1-mediated editing of its own mRNA transcript via an ADD1/MATR3 regulatory axis.

Genetic information is generally transferred from RNA to protein according to the classic "Central Dogma". Here, we made a striking discovery that post-translational modification of a protein specifically regulates the editing of its own mRNA. We show that S-nitrosylation of cathepsin B (CTSB) exclusively alters the adenosine-to-inosine (A-to-I) editing of its own mRNA. Mechanistically, CTSB S-nitrosylation promotes the dephosphorylation and nuclear translocation of ADD1, leading to the recruitment of MATR3 and ADAR1 to CTSB mRNA. ADAR1-mediated A-to-I RNA editing enables the binding of HuR to CTSB mRNA, resulting in increased CTSB mRNA stability and subsequently higher steady-state levels of CTSB protein. Together, we uncovered a unique feedforward mechanism of protein expression regulation mediated by the ADD1/MATR3/ADAR1 regulatory axis. Our study demonstrates a novel reverse flow of information from the post-translational modification of a protein back to the post-transcriptional regulation of its own mRNA precursor. We coined this process as "Protein-directed EDiting of its Own mRNA by ADAR1 (PEDORA)" and suggest that this constitutes an additional layer of protein expression control. "PEDORA" could represent a currently hidden mechanism in eukaryotic gene expression regulation.

Evaluation of the post-implantation development of mouse embryos derived from round spermatid injection.

Round spermatid injection (ROSI), one of the assistant reproductive technologies, was used to treat partial infertility patients suffering from non-obstructive azoospermia. However, the development efficiency and birth rate of ROSI embryos are extremely low, and it is urgent to investigate the underlying mechanisms of low efficiency to improve the clinical application of ROSI technology. Here, we analyzed and compared the genome stability of the mouse blastocyst and the post-implantation development between ROSI and ICSI embryos. We first sequenced the genome of blastocysts from mouse ROSI embryos that can correctly form male and female pronuclei (2 PN) and found that the genomes of 7 blastocysts were normal. Furthermore, the implantation rate of ROSI 2 PN embryos on embryonic day 7.5 is similar to that of ICSI embryos, and at this time, 37.50% (9/24) of deciduas have no normal gestational sac. The proportion of embryos that survived to embryonic day 11.5 in the ROSI 2 PN group, ROSI non-2 PN group, parthenogenesis group, and ICSI 2 PN group was 51.61%, 7.14%, 0.00%, and 55.00%, respectively. And two smaller fetuses were found in the ROSI 2 PN group, which is not found in the other three groups. In addition, the physiological indexes, including fetus and placenta weight, sex ratio, growth rate, and the natural breeding ability for the offspring obtained from mouse ROSI, were evaluated; ROSI mice exhibited no obvious defect or abnormality and implied that the progeny were safe. Our results provided new evidence to promote the clinical application of ROSI technology.

Prior Image Guided Snapshot Compressive Spectral Imaging.

Spectral images with rich spatial and spectral information have wide usage, however, traditional spectral imaging techniques undeniably take a long time to capture scenes. We consider the computational imaging problem of the snapshot spectral spectrometer, i.e., the Coded Aperture Snapshot Spectral Imaging (CASSI) system. For the sake of a fast and generalized reconstruction algorithm, we propose a prior image guidance-based snapshot compressive imaging method. Typically, the prior image denotes the RGB measurement captured by the additional uncoded panchromatic camera of the dual-camera CASSI system. We argue that the RGB image as a prior image can provide valuable semantic information. More importantly, we design the Prior Image Semantic Similarity (PIDS) regularization term to enhance the reconstructed spectral image fidelity. In particular, the PIDS is formulated as the difference between the total variation of the prior image and the recovered spectral image. Then, we solve the PIDS regularized reconstruction problem by the Alternating Direction Method of Multipliers (ADMM) optimization algorithm. Comprehensive experiments on various datasets demonstrate the superior performance of our method.

The alterations in multiple neurophysiological procedures are associated with frailty phenotype in older adults.

Background: Older adults oftentimes suffer from the conditions in multiple physiologic systems, interfering with their daily function and thus contributing to physical frailty. The contributions of such multisystem conditions to physical frailty have not been well characterized. Methods: In this study, 442 (mean age = 71.4 ± 8.1 years, 235 women) participants completed the assessment of frailty syndromes, including unintentional weight loss, exhaustion, slowness, low activity, and weakness, and were categorized into frail (≥3 conditions), pre-frail (1 or 2 conditions), and robust (no condition) status. Multisystem conditions including cardiovascular diseases, vascular function, hypertension, diabetes, sleep disorders, sarcopenia, cognitive impairment, and chronic pain were assessed. Structural equation modeling examined the interrelationships between these conditions and their associations with frailty syndromes. Results: Fifty (11.3%) participants were frail, 212 (48.0%) were pre-frail, and 180 (40.7%) were robust. We observed that worse vascular function was directly associated with higher risk of slowness [standardized coefficient (SC) = -0.419, p < 0.001], weakness (SC = -0.367, p < 0.001), and exhaustion (SC = -0.347, p < 0.001). Sarcopenia was associated with both slowness (SC = 0.132, p = 0.011) and weakness (SC = 0.217, p = 0.001). Chronic pain, poor sleep quality, and cognitive impairment were associated with exhaustion (SC = 0.263, p < 0.001; SC = 0.143, p = 0.016; SC = 0.178, p = 0.004, respectively). The multinomial logistic regression showed that greater number of these conditions were associated with increased probability of being frail (odds ratio>1.23, p < 0.032). Conclusion: These findings in this pilot study provide novel insights into how multisystem conditions are associated with each other and with frailty in older adults. Future longitudinal studies are warranted to explore how the changes in these health conditions alter frailty status.

Sequential administration of virus-like particle-based nanomedicine to elicit enhanced tumor chemotherapy.

Protein cages have played a long-standing role in biomedicine applications, especially in tumor chemotherapy. Among protein cages, virus like particles (VLPs) have received attention for their potential applications in vaccine development and targeted drug delivery. However, most of the existing protein-based platform technologies are plagued with immunological problems that may limit their systemic delivery efficiency as drug carriers. Here, we show that using immune-orthogonal protein cages sequentially and modifying the dominant loop epitope can circumvent adaptive immune responses and enable effective drug delivery using repeated dosing. We genetically modified three different hepadnavirus core protein derived VLPs as delivery vectors for doxorubicin (DOX). These engineered VLPs have similar assembly characteristics, particle sizes, and immunological properties. Our results indicated that there was negligible antibody cross-reactivity in either direction between these three RGD-VLPs in mice that were previously immunized against HBc VLPs. Moreover, the sequential administration of multiple RGD-VLP-based nanomedicine (DOX@RGD-VLPs) could effectively reduce immune clearance and inhibited tumor growth. Hence, this study could provide an attractive protein cage-based platform for therapeutic drug delivery.

A new process to further remove dissolved organic matter and disinfection by-product formation potential during drinking water treatment.

Biological activated carbon (BAC) will produce soluble microbial products (SMPs), which affect effluent quality. To clarify the mechanism by which BAC affects effluent water quality, the processes of a drinking water plant in Jiangsu Province were investigated. It was found that during the O3-BAC process, although ozonation could remove dissolved organic matter (DOC) to a certain extent, the DOC increased from 4.44 to 4.47 mg/L after BAC. Dissolved organic matter (DOM) in effluent from different processes was divided into five fractions based on hydrophilicity and hydrophobicity by resin fractionation. Through fluorescence excitation-emission matrix (EEM) spectroscopy combined with DOC analysis, it was found that SMPs are mainly included in transitional hydrophilic neutral (TPIN) fraction, which was the main cause of the DOC increase. Therefore, a new combined process was designed to remove TPIN effectively by coagulation after biological treatment, and found that coagulation had a good removal rate (13.2%) on TPIN. The trihalomethane formation potential (THMFP) of TPIN could be reduced effectively by 44.9% after coagulation. Compared with the old process, the new combined process had a higher removal rate (14.2-30.0%) of DOC, as well as a greater reduction of THMFP (29.0-78.6%) and haloacetic acid formation potential (HAAFP) (46.4-75.3%). This study aims to reveal the mechanism by which SMPs affect effluent water quality and exacerbate health risks, and to propose a solution to provide theoretical support for the design and optimization of drinking water treatment processes.

D-BIN: A Generalized Disentangling Batch Instance Normalization for Domain Adaptation.

Pattern recognition is significantly challenging in real-world scenarios by the variability of visual statistics. Therefore, most existing algorithms relying on the independent identically distributed assumption of training and test data suffer from the poor generalization capability of inference on unseen testing datasets. Although numerous studies, including domain discriminator or domain-invariant feature learning, are proposed to alleviate this problem, the data-driven property and lack of interpretation of their principle throw researchers and developers off. Consequently, this dilemma incurs us to rethink the essence of networks' generalization. An observation that visual patterns cannot be discriminative after style transfer inspires us to take careful consideration of the importance of style features and content features. Does the style information related to the domain bias? How to effectively disentangle content and style features across domains? In this article, we first investigate the effect of feature normalization on domain adaptation. Based on it, we propose a novel normalization module to adaptively leverage the propagated information through each channel and batch of features called disentangling batch instance normalization (D-BIN). In this module, we explicitly explore domain-specific and domaininvariant feature disentanglement. We maneuver contrastive learning to encourage images with the same semantics from different domains to have similar content representations while having dissimilar style representations. Furthermore, we construct both self-form and dual-form regularizers for preserving the mutual information (MI) between feature representations of the normalization layer in order to compensate for the loss of discriminative information and effectively match the distributions across domains. D-BIN and the constrained term can be simply plugged into state-of-the-art (SOTA) networks to improve their performance. In the end, experiments, including domain adaptation and generalization, conducted on different datasets have proven their effectiveness.

Commercial scale genetic transformation of mature seed embryo explants in maize.

A novel, efficient maize genetic transformation system was developed using Agrobacterium-mediated transformation of embryo explants from mature seeds. Seeds from field grown plants were sterilized and crushed to isolate embryo explants consisting of the coleoptile, leaf primordia, and shoot apical meristem which were then purified from the ground seed bulk preparation. The infection of relevant tissues of seed embryo explants (SEEs) by Agrobacterium was improved by the centrifugation of the explants. Transgenic plants were obtained by multiple bud induction on high cytokinin media, followed by plant regeneration on hormone-free medium. Three different selectable markers (cp4 epsps, aadA, and nptII) were successfully used for producing transgenic plants. Stable integration of transgenes in the maize genome was demonstrated by molecular analyses and germline transmission of the inserted transgenes to the next generation was confirmed by pollen segregation and progeny analysis. Phenotypic evidence for chimeric transgenic tissue was frequently observed in initial experiments but was significantly reduced by including a second bud induction step with optimized cytokinin concentration. Additional improvements, including culturing explants at an elevated temperature during bud induction led to the development of a revolutionary system for efficient transgenic plant production and genome editing. To our knowledge, this is the first report of successful transgenic plant regeneration through Agrobacterium-mediated transformation of maize mature SEEs. This system starts with mature seed that can be produced in large volumes and the SEEs explants are storable. It has significant advantages in terms of scalability and flexibility over methods that rely on immature explants.