Immunotherapy efficacy prediction through a feature re-calibrated 2.5D neural network.

BACKGROUND AND OBJECTIVE: Lung cancer continues to be a leading cause of cancer-related mortality worldwide, with immunotherapy emerging as a promising therapeutic strategy for advanced non-small cell lung cancer (NSCLC). Despite its potential, not all patients experience benefits from immunotherapy, and the current biomarkers used for treatment selection possess inherent limitations. As a result, the implementation of imaging-based biomarkers to predict the efficacy of lung cancer treatments offers a promising avenue for improving therapeutic outcomes. METHODS: This study presents an automatic system for immunotherapy efficacy prediction on the subjects with lung cancer, facilitating significant clinical implications. Our model employs an advanced 2.5D neural network that incorporates 2D intra-slice feature extraction and 3D inter-slice feature aggregation. We further present a lesion-focused prior to guide the re-calibration for intra-slice features, and a attention-based re-calibration for the inter-slice features. Finally, we design an accumulated back-propagation strategy to optimize network parameters in a memory-efficient fashion. RESULTS: We demonstrate that the proposed method achieves impressive performance on an in-house clinical dataset, surpassing existing state-of-the-art models. Furthermore, the proposed model exhibits increased efficiency in inference for each subject on average. To further validate the effectiveness of our model and its components, we conducted comprehensive and in-depth ablation experiments and discussions. CONCLUSION: The proposed model showcases the potential to enhance physicians' diagnostic performance due to its impressive performance in predicting immunotherapy efficacy, thereby offering significant clinical application value. Moreover, we conduct adequate comparison experiments of the proposed methods and existing advanced models. These findings contribute to our understanding of the proposed model's effectiveness and serve as motivation for future work in immunotherapy efficacy prediction.

Enhancement of PRMT6 binding to a novel germline GATA1 mutation associated with congenital anemia.

Mutations in the master hematopoietic transcription factor GATA1 are often associated with functional defects in erythropoiesis and megakaryopoiesis. In this study, we identified a novel GATA1 germline mutation (c.1162delGG, p.Leu387Leufs*62) in a patient with congenital anemia and occasional thrombocytopenia. The C-terminal GATA1, a rarely studied mutational region, undergoes frameshifting translation as a consequence of this double-base deletion mutation. To investigate the specific function and pathogenic mechanism of this mutant, in vitro mutant models of stable re-expression cells were generated. The mutation was subsequently validated to cause diminished transcriptional activity of GATA1 and defective differentiation of erythroid and megakaryocytes. Using proximity labeling and mass spectrometry, we identified selective alterations in the proximal protein networks of the mutant, revealing decreased binding to a set of normal GATA1-interaction proteins, including the essential co-factor FOG1. Notably, our findings further demonstrated enhanced recruitment of the protein arginine methyltransferase PRMT6, which mediates histone modification at H3R2me2a and represses transcription activity. We also found an enhanced binding of this mutant GATA1/PRMT6 complex to the transcriptional regulatory elements of GATA1's target genes. Moreover, treatment of the PRMT6 inhibitor MS023 could partially rescue the inhibited transcriptional and impaired erythroid differentiation caused by the GATA1 mutation. Taken together, our results provide molecular insights into erythropoiesis in which mutation leads to partial loss of GATA1 function and the broader role of PRMT6 and its inhibitor MS023 in congenital anemia, highlighting PRMT6 binding as a negative factor of GATA1 transcriptional activity in aberrant hematopoiesis.

Modifiable lifestyle, mental health status and diabetic retinopathy in U.S. adults aged 18-64 years with diabetes: a population-based cross-sectional study from NHANES 1999-2018.

BACKGROUND: The relationship between integrated lifestyles, mental status and their impact on overall well-being has attracted considerable attention. This study aimed to evaluate the association between lifestyle factors, depression and diabetic retinopathy (DR) in adults aged 18-64 years. METHODS: A cohort of 3482 participants diagnosed with diabetes was drawn from the National Health and Nutrition Examination Survey (NHANES) spanning the years 1999-2018. DR was defined based on self-reported diabetic retinopathy diagnoses by professional physicians, relying on Diabetes Interview Questionnaires. Subgroup analysis was employed to assess lifestyle and psychological factors between participants with DR and those without, both overall and stratified by diabetic duration. Continuous variables were analyzed using the student's t test, while weighted Rao-Scott χ2 test were employed for categorical variables to compare characteristics among the groups. RESULTS: Of the 3482 participants, 767 were diagnosed with diabetic retinopathy, yielding a weighted DR prevalence of 20.8%. Patients with DR exhibited a higher prevalence of heavy drinking, depression, sleep deprivation, and insufficient physical activity compared to those without DR. Furthermore, multivariable logistic regression analysis revealed that sleeping less than 5 h (OR = 3.18, 95%CI: 2.04-4.95, p < 0.001) and depression (OR = 1.35, 95%CI:1.06-1.64, p = 0.025) were associated with a higher risk of DR, while moderate drinking (OR = 0.49, 95%CI: 0.32-0.75, p = 0.001) and greater physical activity (OR = 0.64, 95%CI: 0.35-0.92, p = 0.044) were identified as protective factors. CONCLUSIONS: Adults aged 18-64 years with DR exhibited a higher prevalence of lifestyle-related risk factors and poorer mental health. These findings underscore the need for concerted efforts to promote healthy lifestyles and positive emotional well-being in this population.

Walking pace and microvascular complications among individuals with type 2 diabetes: A cohort study from the UK Biobank.

INTRODUCTION: Walking pace is associated with various health-related outcomes. The aim of this study was to investigate the association between self-reported walking pace and the incidences of diabetic microvascular complications among participants with type 2 diabetes (T2D). METHODS: Self-reported walking pace was classified as brisk, average, or slow. The outcomes were the incidences of diabetic retinopathy, diabetic neuropathy, and diabetic nephropathy. COX proportional hazards models adjusted for sociodemographic, lifestyle, and health-related factors were used to estimate hazard ratios (HRs) and 95% CIs. RESULTS: A total of 14 518 participants with T2D in the UK Biobank (mean age 59.7 ± 7.0 years, 5028 [34.6%] women) were included. During a median follow-up of 12.5 (interquartile range: 11.6-13.4) years, 2980 participants developed diabetic microvascular complications. After adjusting for confounding factors, and compared with brisk walkers, slow walkers had a multivariable-adjusted HR of 1.98 (95% CI 1.58, 2.47) for composite diabetic microvascular complications, 1.54 (95% CI 1.11, 2.14) for diabetic retinopathy, 3.26 (95% CI 2.08, 5.11) for diabetic neuropathy, and 2.32 (95% CI 1.91, 2.82) for diabetic nephropathy. Average walking pace was associated with a higher risk for diabetic nephropathy (HR 1.51, 95 CI% 1.27-1.79) compared with brisk walking. Additionally, ≥1 diabetic microvascular complication occurred in 447 (14.7%) of participants with brisk walking pace, 1702 (19.5%) with average walking pace, and 831 (30.4%) with slow walking pace. Time from study recruitment to first diagnosis was shorter in participants who reported a slow walking pace, compared with brisk or average walkers. Among participants who had diabetic nephropathy as their first diagnosis, slow walking pace was associated with subsequent risk of a second diabetic microvascular complication (HR 3.88, 95 CI% 2.27-6.60). CONCLUSIONS: Self-reported slow walking pace is associated with a higher risk of diabetic microvascular complications among participants with T2D in this population-based cohort study.

Co-MOF@MWCNTs/GCE for the sensitive detection of TBHQ in food samples.

tert-Butylhydroquinone (TBHQ) is a novel synthetic antioxidant with a higher safety profile and antioxidant effect that is more excellent than other synthetic antioxidants and is internationally recognized as one of the best food antioxidants. However, its excessive use in food can have unfavorable effects on the human body. Thus, it is critical to establish a rapid method for the detection of TBHQ in food samples. In this study, a cobalt-based metal-organic framework (Co-MOF) was fabricated by a one-pot hydrothermal method and embedded in multi-walled carbon nanotubes (MWCNTs) to construct an economical and sensitive electrochemical sensor for TBHQ. The results showed that this sensor possessed a wide linear range (0.004-20 µM and 20-300 µM), a low limit of detection (LOD = 2.5 nM, S/N = 3) as well as an ultra-high sensitivity (43.19 µA µM-1 cm-2). Moreover, the sensor also has superior selectivity, repeatability, reproducibility and anti-interference ability and can be successfully applied for the detection of TBHQ in samples of instant noodles and potato chips.

Association of BMI and waist circumference with diabetic microvascular complications: A prospective cohort study from the UK Biobank and Mendelian randomization analysis.

AIMS: To investigate the precise association between BMI and waist circumference (WC) and diabetic complications, including retinopathy (DR), nephropathy (DN) and peripheral neuropathy (DPN). METHODS: A multivariable-adjusted Cox proportional hazard model was used to evaluate the observed association from 30,541 UK Biobank participants with diabetes. A two-sample Mendelian randomization (MR) framework was applied to summary-level GWASs of BMI and WC comprising a total of 461,460 and 462,166 participants from UK Biobank to explore the potential causal association. RESULTS: Higher BMI and WC were associated with increased risks of DR, DN, and DPN (HR (95% CI), per-SD increase: BMI: DR 1.09 (1.04-1.13), DN 1.37 (1.33-1.41), DPN 1.27 (1.20-1.34); WC: DR 1.11 (1.07-1.16), DN 1.41 (1.36-1.46), DPN 1.38 (1.30-1.45)) in the UK Biobank cohort. Univariate MR indicated that increased BMI and WC were causal risk factors for these complications (OR (95% CI), per-SD increase: BMI: DR 1.33 (1.22-1.45), DN 1.74 (1.47-2.07), DPN 2.20 (1.67-2.90); WC: DR 1.43 (1.27-1.61), DN 2.03 (1.62-2.55), DPN 2.80 (1.99-3.92)), and the effect sizes remained significant after adjustment for glycated hemoglobin. CONCLUSIONS: Prospective observational and MR analyses provided evidence that high BMI and WC may represent potential causal risk factors for diabetic microvascular complications. Weight control might modify the risks of these complications independently of glycemic control and should be considered as a therapeutic recommendation.

Photolysis of Phosphaketenyltetrylenes with a Carbazolyl Substituent.

Phosphaketenes of divalent group 14 compounds can potentially serve as precursors for the synthesis of heavy multiple-bond systems. We have employed the dtbp Cbz substituent (dtbp Cbz=1,8-bis(3,5-ditertbutylphenyl)-3,6-ditertbutylcarbazolyl) to prepare such phosphaketenyltetrylenes [(dtbp Cbz)EPCO] (E=Ge, Sn, Pb). While the phosphaketenyltetrylenes are stable at ambient conditions, they can be readily decarbonylated photolytically. For the germylene and stannylene derivatives, dimeric diphosphene-type products [(dtbp Cbz)EP]2 (E=Ge, Sn) were obtained. In contrast, photolysis of the phosphaketenylplumbylene, via isomerisation of the [(dtbp Cbz)PbP] intermediate to [(dtbp Cbz)PPb], afforded an unsymmetric and incompletely decarbonylated product [(dtbp Cbz)2 Pb2 P2 CO] formally comprising a [(dtbp Cbz)PPb] and a [(dtbp Cbz)PbPCO] moiety.

Environmentally relevant concentrations of fipronil selectively disrupt venous vessel development in zebrafish embryos/larvae.

The pesticide fipronil is widely dispersed in aquatic environments and frequently detected in the general population. Although the adverse effects on embryonic growth by fipronil exposure have been extensively documented, the early responses for its developmental toxicity are largely unknown. In the present study, we explored the sensitive targets of fipronil, focusing on vascular injury using zebrafish embryos/larvae and cultured human endothelial cells. Exposure to 5-500 µg/L fipronil at the early stage impeded the growth of sub-intestinal venous plexus (SIVP), caudal vein plexus (CVP), and common cardinal veins (CCV). The damages on venous vessels occurred at exposure to the environmentally relevant concentration as low as 5 µg/L fipronil, whereas no significant change was observed in general toxicity indexes. In contrast, vascular development of the dorsal aorta (DA) or intersegmental artery (ISA) was not affected. In addition, the mRNA levels of vascular markers and vessel type-specific function genes exhibited significant decreases in venous genes, including nr2f2, ephb4a, and flt4, but no appreciable change in arterial genes. Likewise, the more pronounced changes in cell death and cytoskeleton disruption were shown in human umbilical vein endothelial cells as compared with human aortic endothelial cells. Furthermore, molecular docking supported a stronger affinity of fipronil and its metabolites to the proteins correlated with venous development, such as BMPR2 and SMARCA4. These results reveal the heterogeneity in developing vasculature responsive to fipronil's exposure. The preferential impacts on the veins confer higher sensitivity, allowing them to be appropriate targets for monitoring fipronil's developmental toxicity.

Single-cell multi-omics in the medicinal plant Catharanthus roseus.

Advances in omics technologies now permit the generation of highly contiguous genome assemblies, detection of transcripts and metabolites at the level of single cells and high-resolution determination of gene regulatory features. Here, using a complementary, multi-omics approach, we interrogated the monoterpene indole alkaloid (MIA) biosynthetic pathway in Catharanthus roseus, a source of leading anticancer drugs. We identified clusters of genes involved in MIA biosynthesis on the eight C. roseus chromosomes and extensive gene duplication of MIA pathway genes. Clustering was not limited to the linear genome, and through chromatin interaction data, MIA pathway genes were present within the same topologically associated domain, permitting the identification of a secologanin transporter. Single-cell RNA-sequencing revealed sequential cell-type-specific partitioning of the leaf MIA biosynthetic pathway that, when coupled with a single-cell metabolomics approach, permitted the identification of a reductase that yields the bis-indole alkaloid anhydrovinblastine. We also revealed cell-type-specific expression in the root MIA pathway.

Land carbon metabolism in Xiamen-Zhangzhou-Quanzhou region based on ecological network analysis.

Discovering the underlying mechanisms between carbon metabolism and carbon balance of human-natural system is of important theoretical and practical significance for reducing regional carbon emissions and promoting low-carbon development. Taking Xiamen-Zhangzhou-Quanzhou region from 2000 to 2020 as an example, we constructed the framework of spatial network model of land carbon metabolism based on carbon flow, and inquired into the spatial and temporal heterogeneity in the carbon metabolic structure, function and ecological relationships by using the ecological network analysis. The results showed that the dominant negative carbon transitions related with land use changes came from the conversion of cultivated land to industrial and transportation land, and that the high value areas of negative carbon flow were mainly distributed in the areas with relatively developed industries in the middle and east parts of Xiamen-Zhangzhou-Quanzhou region. Competition relationships were the dominant type and the spatial expansion was obvious, which led to the decreases of the integral ecological utility index and affected the regional carbon metabolic balance. The ecological network hierarchy of driving weight changed from a pyramid structure to a relatively more regular structure, with the producer contributing the most. The ecological network hie-rarchy of pull weight changed from a pyramid structure to an inverted pyramid structure, mainly due to the excessive increase of industrial and transportation land weights. Low-carbon development should focus on the sources of negative carbon transitions caused by land use conversion and its comprehensive impacts on carbon metabolic balance, to formulate differentiated low-carbon land use patterns and carbon emission reduction policies.

Different associations of anthropometric indices with diabetic retinopathy and diabetic kidney disease in chinese patients with type 2 diabetes mellitus.

AIMS: To investigate the associations of anthropometric indices, including body mass index (BMI), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), waist circumference (WC) and hip circumference (HC), with diabetic retinopathy (DR) and diabetic kidney disease (DKD) in Chinese patients with type 2 diabetes mellitus (T2DM). MATERIALS AND METHODS: This cross-sectional study evaluated 5226 Chinese participants with T2DM at three hospitals between 2005 and 2016. Logistic regression models and restricted cubic spline analysis were used to assess the associations of anthropometric indices with DR and DKD. RESULTS: A BMI of around 25 kg/m2 was related to a low risk of DR (OR based on the third fifth: 0.752, 95%CI: 0.615-0.920). Besides, HC had an inverse association with DR in men independently of BMI (OR based on the highest fifth: 0.495, 95%CI: 0.350-0.697). In the restricted cubic spline models, BMI, WHtR, WC, and HC showed J-shaped associations with DKD, while WHR showed an S-shaped association with DKD. Compared to the lowest fifth, the odds ratios (OR) based on the highest fifth of BMI, WHR, WHtR, WC and HC for DKD were 1.927 (1.572-2.366), 1.566 (1.277-1.923), 1.910 (1.554-2.351), 1.624 (1.312-2.012) and 1.585 (1.300-1.937) respectively in multivariable models. CONCLUSIONS: A median BMI and a large hip might be related to a low risk of DR, while lower levels of all the anthropometric indices were associated with a lower risk of DKD. Our findings suggested maintain a median BMI, a low WHR, a low WHtR and a large hip for prevention of DR and DKD.

"Simple Tidy GeneCoEx": A gene co-expression analysis workflow powered by tidyverse and graph-based clustering in R.

Gene co-expression analysis is an effective method to detect groups (or modules) of co-expressed genes that display similar expression patterns, which may function in the same biological processes. Here, we present "Simple Tidy GeneCoEx", a gene co-expression analysis workflow written in the R programming language. The workflow is highly customizable across multiple stages of the pipeline including gene selection, edge selection, clustering resolution, and data visualization. Powered by the tidyverse package ecosystem and network analysis functions provided by the igraph package, the workflow detects gene co-expression modules whose members are highly interconnected. Step-by-step instructions with two use case examples as well as source code are available at https://github.com/cxli233/SimpleTidy_GeneCoEx.

Transient receptor potential channel 6 knockdown prevents high glucose-induced Müller cell pyroptosis.

BACKGROUND: Transient receptor potential channel 6 (TRPC6) is reported to be involved in the pathogenesis of diabetic complications, but its role in diabetic retinopathy (DR) remains unknown. The aim of our study was to determine the role and mechanism of TRPC6 in DR. METHODS: High glucose was used to construct a DR cell model using rat retinal Müller cells (rMC-1). Intracellular Ca2+, reactive oxygen species (ROS) and cell pyroptosis were evaluated by flow cytometry. Protein levels of NLRP3, pro-caspase-1, active caspase-1, gasdermin D (GSDMD), GSDMD-N, TRPC6 and H3K27ac were detected by Western blot. mRNA levels of EP300 and TRPC6 were analyzed by reverse transcriptase-polymerase chain reaction (RT-PCR). Levels of IL-1ß and IL-18 were estimated by enzyme linked immunosorbent assay (ELISA). The interaction between EP300 and TRPC6 was validated by a chromatin immunoprecipitation assay. RESULTS: The knockdown of TRPC6 reduced inflammation and cell pyroptosis in HG induced rMC-1 cells, whereas overexpression of TRPC6 had the opposite effects. The inhibition of ROS and NLRP3 reversed TRPC6-mediated cell pyroptosis in the DR cell model. In addition, EP300 increased the expression of H3K27ac and TRPC6 to promote cell pyroptosis, which was suppressed by the knockdown of TRPC6. CONCLUSIONS: Our study revealed a novel EP300/H3K27ac/TRPC6 signaling pathway that may contribute to HG induced Müller cell pyroptosis. TRPC6 played a novel role in Müller cell pyroptosis triggered by HG, and may be a potential target for DR treatment in the future.

Research Progress on Therapeutic Targeting of Cancer-Associated Fibroblasts to Tackle Treatment-Resistant NSCLC.

Non-small cell lung cancer (NSCLC) accounts for most lung cancer cases and is the leading cause of cancer-related deaths worldwide. Treatment options for lung cancer are no longer limited to surgery, radiotherapy, and chemotherapy, as targeted therapy and immunotherapy offer a new hope for patients. However, drug resistance in chemotherapy and targeted therapy, and the low response rates to immunotherapy remain important challenges. Similar to tumor development, drug resistance occurs because of significant effects exerted by the tumor microenvironment (TME) along with cancer cell mutations. Cancer-associated fibroblasts (CAFs) are a key component of the TME and possess multiple functions, including cross-talking with cancer cells, remodeling of the extracellular matrix (ECM), secretion of various cytokines, and promotion of epithelial-mesenchymal transition, which in turn provide support for the growth, invasion, metastasis, and drug resistance of cancer cells. Therefore, CAFs represent valuable therapeutic targets for lung cancer. Herein, we review the latest progress in the use of CAFs as potential targets and mediators of drug resistance for NSCLC treatment. We explored the role of CAFs on the regulation of the TME and surrounding ECM, with particular emphasis on treatment strategies involving combined CAF targeting within the current framework of cancer treatment.

Analysis of hazard perception characteristics based on driving behavior considering overt and covert hazard scenarios.

The drivers' hazard perception plays an important role in preventing and reducing the occurrence of traffic accidents. In order to explore the drivers' hazard perception and their behavioral characteristics in overt and covert hazards, hazardous events of three traffic conflict types (vehicle to vehicle, vehicle to cyclist and vehicle to pedestrian) were designed for overt and covert hazards based on the UC-win/Road driving simulation software, respectively. 35 drivers were organized to conduct the driving simulation tests. The data of driving behavior was collected when they were driving. A comparative analysis of drivers' hazard perception ability and driving behavior characteristics was carried out for hazardous scenarios and traffic conflict types. The result has shown that drivers are more likely to take slowing measures or brake earlier in overt hazard scenarios to ensure safe driving. And drivers are more likely to be involved in collisions in covert hazard scenarios. The types of traffic conflict have a significant effect on the hazard perception ability of drivers (F = 5.92, p < 0.01). Drivers have the strongest hazard perception for cyclists and the weakest hazard perception for pedestrians. Traffic conflict types has a significant effect on drivers' average braking depth (F = 32.31, p < 0.01), average speed (F = 13.78, p < 0.01), and average acceleration (F = 9.26, p < 0.01).

Impact of Notch3 Activation on Aortic Aneurysm Development in Marfan Syndrome.

BACKGROUND: The leading cause of mortality in patients with Marfan syndrome (MFS) is thoracic aortic aneurysm and dissection. Notch signaling is essential for vessel morphogenesis and function. However, the role of Notch signaling in aortic pathology and aortic smooth muscle cell (SMC) differentiation in Marfan syndrome (MFS) is not completely understood. METHODS: RNA-sequencing on ascending aortic tissue from a mouse model of MFS, Fbn1mgR/mgR , and wild-type controls was performed. Notch 3 expression and activation in aortic tissue were confirmed with real-time RT-PCR, immunohistochemistry, and Western blot. Fbn1mgR/mgR and wild-type mice were treated with a γ-secretase inhibitor, DAPT, to block Notch activation. Aortic aneurysms and rupture were evaluated with connective tissue staining, ultrasound, and life table analysis. RESULTS: The murine RNA-sequencing data were validated with mouse and human MFS aortic tissue, demonstrating elevated Notch3 activation in MFS. Data further revealed that upregulation and activation of Notch3 were concomitant with increased expression of SMC contractile markers. Inhibiting Notch3 activation with DAPT attenuated aortic enlargement and improved survival of Fbn1mgR/mgR mice. DAPT treatment reduced elastin fiber fragmentation in the aorta and reversed the differentiation of SMCs. CONCLUSIONS: Our data demonstrated that matrix abnormalities in the aorta of MFS are associated with increased Notch3 activation. Enhanced Notch3 activation in MFS contributed to aortic aneurysm formation in MFS. This might be mediated by inducing a contractile phenotypic change of SMC. Our results suggest that inhibiting Notch3 activation may provide a strategy to prevent and treat aortic aneurysms in MFS.

Distinct Co-occurrence Relationships and Assembly Processes of Active Methane-Oxidizing Bacterial Communities Between Paddy and Natural Wetlands of Northeast China.

Studies of methane-oxidizing bacteria are updating our views of their composition and function in paddy and natural wetlands. However, few studies have characterized differences in the methane-oxidizing bacterial communities between paddy and natural wetlands. Here, we conducted a 13C stable isotope-probing experiment and high-throughput sequencing to determine the structure profiling, co-occurrence relationships, and assembly processes of methanotrophic communities in four wetlands of Northeast China. There was a clear difference in community structure between paddy and natural wetlands. LEfSe analyses revealed that Methylobacter, FWs, and Methylosinus were enriched in natural wetlands, while Methylosarcina were prevailing in paddy, all identified as indicative methanotrophs. We observed distinct co-occurrence relationships between paddy and natural wetlands: more robust and complex connections in natural wetlands than paddy wetlands. Furthermore, the relative importance of stochastic processes was greater than that of deterministic processes, as stochastic processes explained >50% of the variation in communities. These results demonstrated that the co-occurrence relationships and assembly processes of active methanotrophic communities in paddy and natural wetlands were distinct. Overall, the results of this study enhance our understanding of the communities of methane-oxidizing bacteria in paddy and natural wetlands of Northeast China.

Resetting of the 24-nt siRNA landscape in rice zygotes.

The zygote, a totipotent stem cell, is crucial to the life cycle of sexually reproducing organisms. It is produced by the fusion of two differentiated cells-the egg and sperm, which in plants have radically different siRNA transcriptomes from each other and from multicellular embryos. Owing to technical challenges, the epigenetic changes that accompany the transition from differentiated gametes to totipotent zygote are poorly understood. Because siRNAs serve as both regulators and outputs of the epigenome, we characterized small RNA transcriptomes of zygotes from rice. Zygote small RNAs exhibit extensive maternal carryover and an apparent lack of paternal contribution, indicated by absence of sperm signature siRNAs. Zygote formation is accompanied by widespread redistribution of 24-nt siRNAs relative to gametes, such that ∼70% of the zygote siRNA loci do not overlap any egg cell siRNA loci. Newly detected siRNA loci in zygote are gene-proximal and not associated with centromeric heterochromatin, similar to canonical siRNAs, in sharp contrast to gametic siRNA loci that are gene-distal and heterochromatic. In addition, zygote but not egg siRNA loci are associated with high DNA methylation in the mature embryo. Thus, the zygote begins transitioning before the first embryonic division to an siRNA profile that is associated with future RdDM in embryogenesis. These findings indicate that, in addition to changes in gene expression, the transition to totipotency in the plant zygote is accompanied by resetting of the epigenetic reprogramming that occurred during gamete formation.

Domain generalization on medical imaging classification using episodic training with task augmentation.

Medical imaging datasets usually exhibit domain shift due to the variations of scanner vendors, imaging protocols, etc. This raises the concern about the generalization capacity of machine learning models. Domain generalization (DG), which aims to learn a model from multiple source domains such that it can be directly generalized to unseen test domains, seems particularly promising to medical imaging community. To address DG, recent model-agnostic meta-learning (MAML) has been introduced, which transfers the knowledge from previous training tasks to facilitate the learning of novel testing tasks. However, in clinical practice, there are usually only a few annotated source domains available, which decreases the capacity of training task generation and thus increases the risk of overfitting to training tasks in the paradigm. In this paper, we propose a novel DG scheme of episodic training with task augmentation on medical imaging classification. Based on meta-learning, we develop the paradigm of episodic training to construct the knowledge transfer from episodic training-task simulation to the real testing task of DG. Motivated by the limited number of source domains in real-world medical deployment, we consider the unique task-level overfitting and we propose task augmentation to enhance the variety during training task generation to alleviate it. With the established learning framework, we further exploit a novel meta-objective to regularize the deep embedding of training domains. To validate the effectiveness of the proposed method, we perform experiments on histopathological images and abdominal CT images.

Few-shot medical image segmentation using a global correlation network with discriminative embedding.

Despite impressive developments in deep convolutional neural networks for medical imaging, the paradigm of supervised learning requires numerous annotations in training to avoid overfitting. In clinical cases, massive semantic annotations are difficult to acquire where biomedical expert knowledge is required. Moreover, it is common when only a few annotated classes are available. In this study, we proposed a new approach to few-shot medical image segmentation, which enables a segmentation model to quickly generalize to an unseen class with few training images. We constructed a few-shot image segmentation mechanism using a deep convolutional network trained episodically. Motivated by the spatial consistency and regularity in medical images, we developed an efficient global correlation module to model the correlation between a support and query image and incorporate it into the deep network. We enhanced the discrimination ability of the deep embedding scheme to encourage clustering of feature domains belonging to the same class while keeping feature domains of different organs far apart. We experimented using anatomical abdomen images from both CT and MRI modalities.