Revisiting Computer-Aided Tuberculosis Diagnosis.

Tuberculosis (TB) is a major global health threat, causing millions of deaths annually. Although early diagnosis and treatment can greatly improve the chances of survival, it remains a major challenge, especially in developing countries. Recently, computer-aided tuberculosis diagnosis (CTD) using deep learning has shown promise, but progress is hindered by limited training data. To address this, we establish a large-scale dataset, namely the Tuberculosis X-ray (TBX11 K) dataset, which contains 11 200 chest X-ray (CXR) images with corresponding bounding box annotations for TB areas. This dataset enables the training of sophisticated detectors for high-quality CTD. Furthermore, we propose a strong baseline, SymFormer, for simultaneous CXR image classification and TB infection area detection. SymFormer incorporates Symmetric Search Attention (SymAttention) to tackle the bilateral symmetry property of CXR images for learning discriminative features. Since CXR images may not strictly adhere to the bilateral symmetry property, we also propose Symmetric Positional Encoding (SPE) to facilitate SymAttention through feature recalibration. To promote future research on CTD, we build a benchmark by introducing evaluation metrics, evaluating baseline models reformed from existing detectors, and running an online challenge. Experiments show that SymFormer achieves state-of-the-art performance on the TBX11 K dataset.

The transcription factor NAC102 confers cadmium tolerance by regulating WAKL11 expression and cell wall pectin metabolism in Arabidopsis.

Cadmium (Cd) toxicity severely limits plant growth and development. Moreover, Cd accumulation in vegetables, fruits, and food crops poses health risks to animals and humans. Although the root cell wall has been implicated in Cd stress in plants, whether Cd binding by cell wall polysaccharides contributes to tolerance remains controversial, and the mechanism underlying transcriptional regulation of cell wall polysaccharide biosynthesis in response to Cd stress is unknown. Here, we functionally characterized an Arabidopsis thaliana NAC-type transcription factor, NAC102, revealing its role in Cd stress responses. Cd stress rapidly induced accumulation of NAC102.1, the major transcript encoding functional NAC102, especially in the root apex. Compared to wild type (WT) plants, a nac102 mutant exhibited enhanced Cd sensitivity, whereas NAC102.1-overexpressing plants displayed the opposite phenotype. Furthermore, NAC102 localizes to the nucleus, binds directly to the promoter of WALL-ASSOCIATED KINASE-LIKE PROTEIN11 (WAKL11), and induces transcription, thereby facilitating pectin degradation and decreasing Cd binding by pectin. Moreover, WAKL11 overexpression restored Cd tolerance in nac102 mutants to the WT levels, which was correlated with a lower pectin content and lower levels of pectin-bound Cd. Taken together, our work shows that the NAC102-WAKL11 module regulates cell wall pectin metabolism and Cd binding, thus conferring Cd tolerance in Arabidopsis.

P2T: Pyramid Pooling Transformer for Scene Understanding.

Recently, the vision transformer has achieved great success by pushing the state-of-the-art of various vision tasks. One of the most challenging problems in the vision transformer is that the large sequence length of image tokens leads to high computational cost (quadratic complexity). A popular solution to this problem is to use a single pooling operation to reduce the sequence length. This paper considers how to improve existing vision transformers, where the pooled feature extracted by a single pooling operation seems less powerful. To this end, we note that pyramid pooling has been demonstrated to be effective in various vision tasks owing to its powerful ability in context abstraction. However, pyramid pooling has not been explored in backbone network design. To bridge this gap, we propose to adapt pyramid pooling to Multi-Head Self-Attention (MHSA) in the vision transformer, simultaneously reducing the sequence length and capturing powerful contextual features. Plugged with our pooling-based MHSA, we build a universal vision transformer backbone, dubbed Pyramid Pooling Transformer (P2T). Extensive experiments demonstrate that, when applied P2T as the backbone network, it shows substantial superiority in various vision tasks such as image classification, semantic segmentation, object detection, and instance segmentation, compared to previous CNN- and transformer-based networks. The code will be released at https://github.com/yuhuan-wu/P2T.

EDN: Salient Object Detection via Extremely-Downsampled Network.

Recent progress on salient object detection (SOD) mainly benefits from multi-scale learning, where the high-level and low-level features collaborate in locating salient objects and discovering fine details, respectively. However, most efforts are devoted to low-level feature learning by fusing multi-scale features or enhancing boundary representations. High-level features, which although have long proven effective for many other tasks, yet have been barely studied for SOD. In this paper, we tap into this gap and show that enhancing high-level features is essential for SOD as well. To this end, we introduce an Extremely-Downsampled Network (EDN), which employs an extreme downsampling technique to effectively learn a global view of the whole image, leading to accurate salient object localization. To accomplish better multi-level feature fusion, we construct the Scale-Correlated Pyramid Convolution (SCPC) to build an elegant decoder for recovering object details from the above extreme downsampling. Extensive experiments demonstrate that EDN achieves state-of-the-art performance with real-time speed. Our efficient EDN-Lite also achieves competitive performance with a speed of 316fps. Hence, this work is expected to spark some new thinking in SOD. Code is available at https://github.com/yuhuan-wu/EDN.

Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation.

Weakly supervised semantic instance segmentation with only image-level supervision, instead of relying on expensive pixel-wise masks or bounding box annotations, is an important problem to alleviate the data-hungry nature of deep learning. In this article, we tackle this challenging problem by aggregating the image-level information of all training images into a large knowledge graph and exploiting semantic relationships from this graph. Specifically, our effort starts with some generic segment-based object proposals (SOP) without category priors. We propose a multiple instance learning (MIL) framework, which can be trained in an end-to-end manner using training images with image-level labels. For each proposal, this MIL framework can simultaneously compute probability distributions and category-aware semantic features, with which we can formulate a large undirected graph. The category of background is also included in this graph to remove the massive noisy object proposals. An optimal multi-way cut of this graph can thus assign a reliable category label to each proposal. The denoised SOP with assigned category labels can be viewed as pseudo instance segmentation of training images, which are used to train fully supervised models. The proposed approach achieves state-of-the-art performance for both weakly supervised instance segmentation and semantic segmentation. The code is available at https://github.com/yun-liu/LIID.

MobileSal: Extremely Efficient RGB-D Salient Object Detection.

The high computational cost of neural networks has prevented recent successes in RGB-D salient object detection (SOD) from benefiting real-world applications. Hence, this article introduces a novel network, MobileSal, which focuses on efficient RGB-D SOD using mobile networks for deep feature extraction. However, mobile networks are less powerful in feature representation than cumbersome networks. To this end, we observe that the depth information of color images can strengthen the feature representation related to SOD if leveraged properly. Therefore, we propose an implicit depth restoration (IDR) technique to strengthen the mobile networks' feature representation capability for RGB-D SOD. IDR is only adopted in the training phase and is omitted during testing, so it is computationally free. Besides, we propose compact pyramid refinement (CPR) for efficient multi-level feature aggregation to derive salient objects with clear boundaries. With IDR and CPR incorporated, MobileSal performs favorably against state-of-the-art methods on six challenging RGB-D SOD datasets with much faster speed (450fps for the input size of 320×320) and fewer parameters (6.5M). The code is released at https://mmcheng.net/mobilesal.

H2A.B is a cancer/testis factor involved in the activation of ribosome biogenesis in Hodgkin lymphoma.

Testis-specific regulators of chromatin function are commonly ectopically expressed in human cancers, but their roles are poorly understood. Examination of 81 primary Hodgkin lymphoma (HL) samples showed that the ectopic expression of the eutherian testis-specific histone variant H2A.B is an inherent feature of HL. In experiments using two HL cell lines derived from different subtypes of HL, H2A.B knockdown inhibited cell proliferation. H2A.B was enriched in both nucleoli of these HL cell lines and primary HL samples. We found that H2A.B enhanced ribosomal DNA (rDNA) transcription, was enriched at the rDNA promoter and transcribed regions, and interacted with RNA Pol I. Depletion of H2A.B caused the loss of RNA Pol I from rDNA chromatin. Remarkably, H2A.B was also required for high levels of ribosomal protein gene expression being located at the transcriptional start site and within the gene body. H2A.B knockdown reduced gene body chromatin accessibility of active RNA Pol II genes concurrent with a decrease in transcription. Taken together, our data show that in HL H2A.B has acquired a new function, the ability to increase ribosome biogenesis.

Regularized Densely-Connected Pyramid Network for Salient Instance Segmentation.

Much of the recent efforts on salient object detection (SOD) have been devoted to producing accurate saliency maps without being aware of their instance labels. To this end, we propose a new pipeline for end-to-end salient instance segmentation (SIS) that predicts a class-agnostic mask for each detected salient instance. To better use the rich feature hierarchies in deep networks and enhance the side predictions, we propose the regularized dense connections, which attentively promote informative features and suppress non-informative ones from all feature pyramids. A novel multi-level RoIAlign based decoder is introduced to adaptively aggregate multi-level features for better mask predictions. Such strategies can be well-encapsulated into the Mask R-CNN pipeline. Extensive experiments on popular benchmarks demonstrate that our design significantly outperforms existing state-of-the-art competitors by 6.3% (58.6% vs. 52.3%) in terms of the AP metric. The code is available at https://github.com/yuhuan-wu/RDPNet.

JCS: An Explainable COVID-19 Diagnosis System by Joint Classification and Segmentation.

Recently, the coronavirus disease 2019 (COVID-19) has caused a pandemic disease in over 200 countries, influencing billions of humans. To control the infection, identifying and separating the infected people is the most crucial step. The main diagnostic tool is the Reverse Transcription Polymerase Chain Reaction (RT-PCR) test. Still, the sensitivity of the RT-PCR test is not high enough to effectively prevent the pandemic. The chest CT scan test provides a valuable complementary tool to the RT-PCR test, and it can identify the patients in the early-stage with high sensitivity. However, the chest CT scan test is usually time-consuming, requiring about 21.5 minutes per case. This paper develops a novel Joint Classification and Segmentation (JCS) system to perform real-time and explainable COVID- 19 chest CT diagnosis. To train our JCS system, we construct a large scale COVID- 19 Classification and Segmentation (COVID-CS) dataset, with 144,167 chest CT images of 400 COVID- 19 patients and 350 uninfected cases. 3,855 chest CT images of 200 patients are annotated with fine-grained pixel-level labels of opacifications, which are increased attenuation of the lung parenchyma. We also have annotated lesion counts, opacification areas, and locations and thus benefit various diagnosis aspects. Extensive experiments demonstrate that the proposed JCS diagnosis system is very efficient for COVID-19 classification and segmentation. It obtains an average sensitivity of 95.0% and a specificity of 93.0% on the classification test set, and 78.5% Dice score on the segmentation test set of our COVID-CS dataset. The COVID-CS dataset and code are available at https://github.com/yuhuan-wu/JCS.

Correlation Analysis Reveals an Important Role of GC Content in Accumulation of Deletion Mutations in the Coding Region of Angiosperm Plastomes.

Variation in GC content is assumed to correlate with various processes, including mutation biases, recombination, and environmental parameters. To date, most genomic studies exploring the evolution of GC content have focused on nuclear genomes, but relatively few have concentrated on organelle genomes. We explored the mechanisms maintaining the GC content in angiosperm plastomes, with a particular focus on the hypothesis of phylogenetic dependence and the correlation with deletion mutations. We measured three genetic traits, namely, GC content, A/T tracts, and G/C tracts, in the coding region of plastid genomes for 1382 angiosperm species representing 350 families and 64 orders, and tested the phylogenetic signal. Then, we performed correlation analyses and revealed the variation in evolutionary rate of selected traits using RRphylo. The plastid GC content in the coding region varied from 28.10% to 43.20% across angiosperms, with a few non-photosynthetic species showing highly reduced values, highlighting the significance of functional constraints. We found strong phylogenetic signal in A/T tracts, but weak ones in GC content and G/C tracts, indicating adaptive potential. GC content was positively and negatively correlated with G/C and A/T tracts, respectively, suggesting a trade-off between these two deletion events. GC content evolved at various rates across the phylogeny, with significant increases in monocots and Lamiids, and a decrease in Fabids, implying the effects of some other factors. We hypothesize that variation in plastid GC content might be a mixed strategy of species to optimize fitness in fluctuating climates, partly through influencing the trade-off between AT → GC and GC → AT mutations.

Comparison of eight complete plastid genomes from three moss families Amblystegiaceae, Calliergonaceae and Pylaisiaceae.

We sequenced and assembled eight complete plastid genomes from three closely related pleurocarpous moss families: Amblystegium serpens, Campyliadelphus stellatus, Cratoneuron filicinum, Drepanocladus aduncus, and Leptodictyum humile (Amblystegiaceae), Calliergon sarmentosum and Warnstorfia exannulata (Calliergonaceae), and Calliergonella cuspidata (Pylaisiaceae). The newly generated plastid genomes range from 124,256 to 124,819 bp, with two inverted repeat regions (9,624-9,696 bp) separated by a large single-copy region (86,422-86,924 bp) and a small single-copy region (18,430-18,514 bp). All these plastid genomes encode 116 unique genes including 82 protein-coding genes, 30 tRNA genes and four rRNAgenes. The overall GC content is between 28.6%-29.3%. Phylogenetic analysis showed that all Amblystegiaceae species Amblystegium serpens, Campyliadelphus stellatus, Cratoneuron filicinum, Drepanocladus aduncus, Leptodictyum humile, and Sanionia uncinata clustered in one clade, which is sister to the Pylaisiaceae species Calliergonella cuspidata. The two Calliergonaceae species Calliergon sarmentosum and Warnstorfia exannulata form a clade and is sister to Amblystegiaceae and Pylaisiaceae.

[Effects of IAA on physiological response to aluminum stress and DNA damage of Trichosanthes kirilowii.] / å²å“šä¹™é ¸å¯¹é“èƒè¿«ä¸‹æ æ¥¼ç”Ÿç†å“åº”åŠDNA损伤的缓解作用.

To investigate the physiological response of IAA (indoleacetic acid) to Trichosanthes kirilowii under aluminum stress and the mitigation of DNA damage, the effects of IAA (0, 10, 25, 50, 75 µmol·L-1 denoted by I0, I10, I25, I50 and I75, respectively) on antioxidant enzyme activity, malondialdehyde (MDA), photosynthetic characteristics, root activity, chlorophyll content and DNA damage of two varieties of T. kirilowii under 300 and 800 µmol·L-1 aluminum environment (denoted by Al300 and Al800, respectively) were examined in hydroponic culture experiments with Hebei Anguo (aluminium tolerant variety) and Zhejiang Puyang Trichosanthes kirilowii (aluminum sensitive variety). The results showed that the activities of superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT), photosynthesis ability, and root activity of both varieties were inhibited to different degrees by aluminum stress, MDA content was significantly increased, and DNA damage was aggravated. The maximum increase of SOD, CAT and POD activities in Anguo and Puyang T. kirilowii under aluminum stress by IAA application were 15.0%, 31.2%, 72.3% and 13.8%, 26.9%, 26.4%, respectively, chlorophyll content increased by 49.9% and 17.9%, MDA accumulation decreased by 39.2% and 22.4% and fluorescence parameters were significantly improved. The treatment of 25 µmol·L-1 IAA significantly increased root activity by 159.1% and 90.9%, while 50 µmol·L-1 IAA effectively slowed the DNA tailing phenomenon in roots, with the product (OTM) value of tail DNA percentage content and tail moment being decreased by 27.6%. 10-50 µmol·L-1 IAA could effectively improve the physiological activity of T. kirilowii under aluminum stress and slow the degree of DNA damage. The tolerance of Anguo variety to aluminum stress was stronger than that of Puyang variety.

Chloroplast phylogenomics of liverworts: a reappraisal of the backbone phylogeny of liverworts with emphasis on Ptilidiales.

As one of the four main lineages diverging from the early diversification of land plants, the phylogeny of liverworts holds the information about nearly 500 Myr of independent adaptation to changing environments. Thus, resolving the phylogenetic history of liverworts will provide unique insights into the successful diversification of early land plants in terrestrial ecosystems. However, the deep diverging events of this group remain incompletely resolved, such as the definite position of Ptilidiales. Here, we aimed to reconstruct the backbone relationships of liverworts using 84 protein-coding chloroplast genes, a dataset comprising 35 representatives from all major lineages of liverworts, and three phylogenetic analyses, namely maximum parsimony, maximum likelihood and Bayesian inference. To test the impact of composition biases, the phylogenetic analyses were carried out using three alignments representing the same dataset either as: (i) nucleotides, (ii) amino acids, or (iii) recoded nucleotides applying ambiguity base code. Chloroplast genome data consistently supported the monophyletic origin of three major lineages in liverworts, as well as the majority of backbone relationships. Ptilidiales were found to be sister to Jungermanniales. The rapid accumulation of G/C tracks as a consequence of increased GC content is an important cause for the long branches inferred in this group. Our study not only provides empirical evidence to support the significance of plastid genome sequencing to reconstruct the phylogeny of this important plant lineage, but also suggests that the GC content has played a critical role in the evolutionary dynamics of plastid genomes in land plants.

Isolation and characterization of Aspergillus niger NBC001 underlying suppression against Heterodera glycines.

Heterodera glycines is the most pervasive soybean pests worldwide. Biocontrol provides a strategy to sustainably control nematodes. In this study, 22 fungal isolates were obtained and identified from cysts of Heterodera spp. Among them, Aspergillus niger NBC001 showed high nematicidal activity against H. glycines. The 2-fold dilution of NBC001 culture filtrate caused 89% mortality of second-stage juveniles and inhibited more than 98% of egg hatching in vitro. In both pot and field experiments, the numbers of H. glycines cysts in soybean seedlings dressed with the the 5-fold concentrated culture filtrate of NBC001 were significantly reduced by 43% and 28%, respectively. In addition, application of NBC001 remarkably reduced the penetration of nematodes into the roots. Histochemical and fluorometric staining analyses indicate that application of NBC001 stimulated hydrogen peroxide activity in the roots and triggered callose deposition in the leaves and roots. Transcription of the PR1a and EREBP genes in the salicylic acid and ethylene signaling pathways was upregulated in soybean plants treated with NBC001. However, the application of concentrated culture filtrate of NBC001 had no significant impacts on the soil microbial community based on next generation DNA sequencing technology. In summary, NBC001 may be a good biocontrol agent against H. glycines via stimulation of the immunity/defense of the plant host.

Dual Sequence Control of Uniform Macromolecules through Consecutive Single Addition by Selective Passerini Reaction.

The selective Passerini reactions of 4-formylbenzoic acid and 4-isocyanobenzoic acid with aliphatic isocyanides and aldehydes were utilized to synthesize sequence-defined uniform macromolecules. Our strategy does not involve any protecting groups or reactive group transformation steps and allows direct and consecutive propagation of sequence in each step. Introduction of diverse side groups by using different aliphatic components provided a range of sequence-defined uniform macromolecules in high yield and gram scale. The strategy also allows further Passerini self-coupling or cross-coupling of the formed sequences with other small molecules, affording polymers with up to 5098.3 Da and 20 side groups. Thus, this strategy will show promise for more efficient synthesis of new sequence-defined macromolecules.

Evaluation of the inhibitory effects of heavy metals on anammox activity: A batch test study.

This study evaluated the interactive effect of Cu(II) and Zn(II) on anaerobic ammonium oxidation (anammox) activity using response surface methodology with a central composite design. A regression model equation was developed and validated to predict the normalized anammox activity (NAA) of anammox granules exposed to various heavy metal concentrations. The joint inhibitory effect tended to exacerbate initially and reversed as the concentrations increased and then moderated again. The most severe inhibition, resulting in a NAA of 20.1%, occurred at Cu(II) and Zn(II) concentrations of 16.3 and 20.0mgL(-1), respectively. Notably, the cumulative toxicity was mitigated with the aid of intermittent exposure acclimatization. Additionally, pre-exposure to Cu(II) in the absence of substrates strongly inhibited anammox activity. However, the presence of NO2(-) significantly enhanced Cu(II) inhibition. Therefore, such conditions should be avoided to minimize the disturbance of the anammox process.

Persistence and Viability of Lecanicillium lecanii in Chinese Agricultural Soil.

The entomopathogenic fungus L. lecanii has been developed as biopesticides and used widely for biological control of several insects in agricultural practice. Due to the lack of isolation/count methods for L. lecanii in soil, the persistence of this fungus in soil appears to have attracted no attention. A selective medium and count method for L. lecanii in soil based on cetyl trimethyl ammonium bromide (CTAB) was developed, and then the persistence and viability of this fungus in soil were investigated under field conditions between 2012 and 2014. The results showed that the rate of recovery for L. lecanii in soil on the selective CTAB medium was satisfactory. The minimum CFUs for L. lecanii on the selective medium (0.5 g/L CTAB) was about 102 conidia/g soil. The L. lecanii density in soil declined quickly in the first month after inoculation with fungal conidia, kept stable for 6 to 10 months, and then decreased gradually until undetectable. L. lecanii could persist for at least 14 months in the agricultural soil of northern China. The colony growth, conidia yield and germination rate on plates, as well as the median lethal concentration or times (LC50 or LT50) to aphids, mycelium growth in aphids and sporulation on aphids of L. lecanii did not change significantly during the persistence in soil. In general, the count method developed here was a very useful tool for monitoring the dynamics of natural or introduced L. lecanii populations in soil, and the data on the persistence of L. lecanii in soil reported here were helpful for biological control and environmental risk assessment.

[Joint effects of apoptosis induced by microcystins and bacterial lipopolysaccharides on grass carp (Ctenopharyngodon idellus) lymphocytes].

In this study, grass carp (Ctenopharyngodon idellus) lymphocytes were used as the vitro test object to demonstrate the joint effects of microcystins (MC-LR) and bacterial lipopolysaccharides (LPS) on fish immune system. The results showed that MC-LR and LPS in the single and combined exposure groups could both induce grass carp lymphocytes apoptosis with typical ladder-like DNA electrophoresis characteristics. However, comparing the apoptosis rate of the combined and single exposure groups, it was suggested that bacterial LPS could cooperate with MC-LR causing a higher rate of fish lymphocytes apoptosis (2.1 and 3.3-fold of that for the single exposure group I (MC-LR) and II (LPS), respectively), and there existed a significant dose-response relationship. The MC-LR cooperating with bacterial LPS decreased the activity of glutathione S-transferase (GST), increased the levels of intracellular reactive oxygen species (ROS) and malondialdehyde (MDA), resulted in DNA damage and cell arrest in G0 phase, which inhibited cell proliferation and accelerated apoptosis. It was proved that MC-LR exacerbated fish immunotoxicity by collaborating with LPS, which had a serious adverse effect on aquaculture industry.

[Toxic effect of nodularin on the apoptosis of grass carp (Ctenopharyngodon idellus) lymphocytes and related mechanisms].

Nodularin is a new kind of cyanobacterial toxins found in water blooms worldwide, and fish is very easily to suffer from the negative effects induced by this toxin. This study found that nodularin could induce the apoptosis of Ctenopharyngodon idellus lymphocytes in vitro. The observation of transmission electron microscopy showed that under the impacts of nodularin, the lymphocytes presented typical apoptosis features, such as obvious cytoplasm condensation and nuclear chromatin agglutination and marginalization. The DNA ladder fragmentation further confirmed the occurrence of the apoptosis of the lymphocytes. After incubated with 1, 10, and 100 microg x L(-1) of nodularin for 12 h, the percentages of apoptotic lymphocytes reached 19.4%, 31.6%, and 71.6%, respectively, suggesting a dose-dependent manner. The nodularin-induced apoptosis was related to the increase of intracellular reactive oxygen species (ROS), decline of mitochondrial transmembrane electric potential, up-regulation of intracellular Ca2+, down-regulation of Bcl-2, and up-regulation of Bax. Meanwhile, the activation of Caspase-3 and Caspase-9 were involved in the process of apoptosis. These results strongly indicated that nodularin could induce the apoptosis of fish lymphocytes via mitochondria pathway.

[Effects of gypsum on the firing of pyrogen-treated thermosensitive neurons in PO/AH of cats].

AIM: To investigate the possible central mechanism of antipyretic effects of Chinese medicine gypsum. METHODS: Gypsum was injected after the fever model was established. The firing rate of thermosensitive neurons in preoptic-anterior hypothalamus(PO/AH) region was recorded by using extracellular microelectrode technique. RESULTS: The injection of pyrogen evoked decrease in firing rate of the warm-sensitive neurons and increase in the cold-sensitive neurons in the region of PO/AH; the changes of the firing rate of pyrogen- treated warm-sensitive and cold-sensitive neurons could be reversed by the injection of gypsum. CONCLUSION: The result may suggest that antipyretic action of gypsum is mediated by its influences on the thermosensitivity neurons in the region of PO/AH.