Comparative genomic analysis between newly sequenced Brucella suis Vaccine Strain S2 and the Virulent Brucella suis Strain 1330.

BACKGROUND: Brucellosis is a bacterial disease caused by Brucella infection. In the late fifties, Brucella suis vaccine strain S2 with reduced virulence was obtained by serial transfer of a virulent B. suis biovar 1 strain in China. It has been widely used for vaccination in China since 1971. Until now, the mechanisms underlie virulence attenuation of S2 are still unknown. RESULTS: In this paper, the whole genome sequencing of S2 was carried out by Illumina Hiseq2000 sequencing method. We further performed the comparative genomic analysis to find out the differences between S2 and the virulent Brucella suis strain 1330. We found premature stops in outer membrane autotransporter omaA and eryD genes. Single mutations were found in phosphatidylcholine synthase, phosphorglucosamine mutase, pyruvate kinase and FliF, which have been reported to be related to the virulence of Brucella or other bacteria. Of the other different proteins between S2 and 1330, such as Omp2b, periplasmic sugar-binding protein, and oligopeptide ABC transporter, no definitive implications related to bacterial virulence were found, which await further investigation. CONCLUSIONS: The data presented here provided the rational basis for designing Brucella vaccines that could be used in other strains.

ES-GNN: Generalizing Graph Neural Networks Beyond Homophily With Edge Splitting.

While Graph Neural Networks (GNNs) have achieved enormous success in multiple graph analytical tasks, modern variants mostly rely on the strong inductive bias of homophily. However, real-world networks typically exhibit both homophilic and heterophilic linking patterns, wherein adjacent nodes may share dissimilar attributes and distinct labels. Therefore, GNNs smoothing node proximity holistically may aggregate both task-relevant and irrelevant (even harmful) information, limiting their ability to generalize to heterophilic graphs and potentially causing non-robustness. In this work, we propose a novel Edge Splitting GNN (ES-GNN) framework to adaptively distinguish between graph edges either relevant or irrelevant to learning tasks. This essentially transfers the original graph into two subgraphs with the same node set but complementary edge sets dynamically. Given that, information propagation separately on these subgraphs and edge splitting are alternatively conducted, thus disentangling the task-relevant and irrelevant features. Theoretically, we show that our ES-GNN can be regarded as a solution to a disentangled graph denoising problem, which further illustrates our motivations and interprets the improved generalization beyond homophily. Extensive experiments over 11 benchmark and 1 synthetic datasets not only demonstrate the effective performance of ES-GNN but also highlight its robustness to adversarial graphs and mitigation of the over-smoothing problem.

The optimal management of large upper ureteral stones (> 1.5 cm) with infection in elderly patients: a comparative analysis of three minimally invasive surgical approaches.

PURPOSE: The aim of this study was to investigate the efficacy and safety of three minimally invasive surgical approaches for treating large upper ureteral stones complicated by infection in elderly (> 60 years) patients. METHODS: Clinical data from 95 elderly patients with large upper ureteral stones and infection, treated at our hospital between January 2018 and April 2023, were retrospectively analyzed. The surgical approaches included FURL (flexible ureteroscopic lithotripsy) 33 cases, mPCNL (minimally percutaneous nephrolithotomy) 29 cases, and RLUL (retroperitoneal laparoscopic ureterolithotomy) 33 cases. Surgical time, intraoperative blood loss, postoperative hospital stay, reoperation rate, incidence of postoperative complications, and hospitalization costs were observed and compared among the three groups. RESULTS: No statistically significant difference was found in stone clearance rates among the three groups (P > 0.05). The FURL group exhibited advantages over the mPCNL and RLUL groups in surgical time, intraoperative blood loss, and postoperative hospital stay (P < 0.05). However, it also had the highest reoperation rate and hospitalization costs (P < 0.05). RLUL demonstrated superiority over the mPCNL and RLUL groups in terms of reoperation rate, incidence of complications, and hospitalization costs (P < 0.05). Notably, mPCNL exhibited the highest complication rate at 37.9% (P < 0.05). CONCLUSION: For elderly patients with large upper ureteral stones complicated by infection, FURL, mPCNL, and RLUL represent effective surgical approaches. Further attention is needed regarding the perioperative safety of mPCNL. RLUL, which offers higher safety, efficacy, and cost-effectiveness, can be considered a primary surgical option for these patients.

Perturbation diversity certificates robust generalization.

Whilst adversarial training has been proven to be one most effective defending method against adversarial attacks for deep neural networks, it suffers from over-fitting on training adversarial data and thus may not guarantee the robust generalization. This may result from the fact that the conventional adversarial training methods generate adversarial perturbations usually in a supervised way so that the resulting adversarial examples are highly biased towards the decision boundary, leading to an inhomogeneous data distribution. To mitigate this limitation, we propose to generate adversarial examples from a perturbation diversity perspective. Specifically, the generated perturbed samples are not only adversarial but also diverse so as to certify robust generalization and significant robustness improvement through a homogeneous data distribution. We provide theoretical and empirical analysis, establishing a foundation to support the proposed method. As a major contribution, we prove that promoting perturbations diversity can lead to a better robust generalization bound. To verify our methods' effectiveness, we conduct extensive experiments over different datasets (e.g., CIFAR-10, CIFAR-100, SVHN) with different adversarial attacks (e.g., PGD, CW). Experimental results show that our method outperforms other state-of-the-art (e.g., PGD and Feature Scattering) in robust generalization performance.

[Construction of Brucella abortus A19-delta VirB12 mutant and evaluation of its protective efficacy against 2308 strain challenge in BALB/c mice].

OBJECTIVE: A19-deltaVirB12 deletion mutant of Brucella abortus was constructed by using homologous recombination technology. BALB/c mice were vaccinated intraperitoneally with the mutant to evaluate protective efficacy against Brucella abortus 2308 challenge. METHOD: A SacB gene was amplified by PCR from pIB279 plasmid. The sequences upstream and downstream of the VirB12 gene were amplified by PCR from Brucella abortus A19. These three PCR products were subsequently inserted into pBK-CMV vector, namely pBK-CMV-SacB-VirB12. This construct was transformed into Brucella abortus A19. The A19-delta VirB12 mutants were obtained by Kan(r) and 5% sucrose selection. Six-week-old female BALB/c mice were distributed into three treatment groups, including A19-delta VirB12 group, A19 group and PBS control group. BALB/c mice were vaccinated intraperitoneally at a dose of 5.0 x 10(4) CFU. At the 45-days post-immunization, all of mice were challenged with 2308 strain. Fifteen days after thechallenge, the levels of infection were expressed as means of the log10 CFU/spleen values. The histological changes were assessed among the groups. RESULTS: Compared with PBS control group, the A19-delta VirB12 deleted mutant had astatistically significant protection against 2308 challengesimilar to A19 strain. Western blotting showed that A19-delta VirB12 mutant did not express VirB12 protein. CONCLUSION: The A19-delta VirB12 deleted mutantelicits a strong protective immunity, and may becomea promising vaccine candidate.

The comparison between contrast-enhanced ultrasound and contrast-enhanced magnetic resonance imaging in diagnosing bladder urothelial carcinoma.

PURPOSE: The surgical treatment of bladder urothelial carcinoma depends on whether the tumor has invaded the bladder muscular layer. Normal ultrasound and contrast-enhanced magnetic resonance imaging (contrast-enhanced MRI) are widely used in patients bear bladder tumors; the latter is also widely used in estimating the muscularis invasion of bladder cancer (BC). However, contrast-enhanced ultrasound (CEUS) is rarely used in this aspect. As the gold standard in diagnosing muscularis invasion remains being pathological examination, this study was set to find out whether there are differences between CEUS and contrast-enhanced MRI in diagnosing bladder malignant tumors and in diagnosing the muscularis invasion of the bladder urothelial carcinoma under the guide of the pathological results. METHODS: 160 patients from Yongchuan Hospital of Chongqing Medical University and The Second Affiliated Hospital of Chongqing Medical University were recruited from July 1st, 2021, to June 30th, 2022. Patients are arranged to undergo CEUS, contrast-enhanced MRI and then take a surgery. After surgery, we compare the results of CEUS, MRI and the pathological results, using software to run the statistical examinations. RESULTS: The accuracies of CEUS and contrast-enhanced MRI in diagnosing malignant bladder tumors were 85.90 and 84.62%, and they had no differences (P > 0.05). While the accuracies of CEUS and contrast-enhanced MRI in diagnosing the muscularis invasion were 84.62 and 76.92%, in which CEUS had a better sensitivity (P < 0.05). CONCLUSIONS: We found that CEUS and contrast-enhanced MRI had no differences in diagnosing the different pathological types (benign or malignant) of BC, but CEUS holds a better sensitivity in diagnosing muscularis invasions of bladder urothelial carcinoma than the contrast-enhanced MRI.

The relationship between inflammatory response markers and the prognosis of non-muscle invasive bladder cancer and the development of a nomogram model.

Purpose: Patients with non-muscle invasive bladder cancer (NMIBC) have a high possibility of recurrence after surgery. We aimed to assess the factors associated with tumor recurrence and to construct a nomogram model that can contribute to personalized treatment plans of each patient. Methods: 496 patients with primary bladder cancer (BC) from 2 centers were retrospectively analyzed. Preoperative neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR), systemic immune-inflammation index (SII), and traditional clinical parameters were collected, then using univariate and multivariate Cox regression analysis to find out the independent risk factors associated with tumor recurrence among them, and then these independent factors were incorporated into the nomogram model. The internal calibration curves and the external calibration curves were used to verify their usefulness. Results: In the training cohort, 150 patients (43.1%) experienced recurrence. After Cox regression analysis, the independent risk factors affecting recurrence-free survival (RFS) were tumor grade, immediate postoperative instillation therapy (IPPIT), NLR, and SII. These factors were used to construct a model to predict RFS 1, 2, 3, and 5 years of NMIBC patients after surgery. And then, we found that the constructed model outperforms the conventional model in terms of accuracy and predictability, the results were verified by statistical tests. Conclusion: Preoperative inflammatory response markers have a predictive value for postoperative recurrence in patients with NMIBC. The constructed nomogram model can be helpful in guiding personalized clinical evaluation and subsequent treatment.

A Little Energy Goes a Long Way: Build an Energy-Efficient, Accurate Spiking Neural Network From Convolutional Neural Network.

This article conforms to a recent trend of developing an energy-efficient Spiking Neural Network (SNN), which takes advantage of the sophisticated training regime of Convolutional Neural Network (CNN) and converts a well-trained CNN to an SNN. We observe that the existing CNN-to-SNN conversion algorithms may keep a certain amount of residual current in the spiking neurons in SNN, and the residual current may cause significant accuracy loss when inference time is short. To deal with this, we propose a unified framework to equalize the output of the convolutional or dense layer in CNN and the accumulated current in SNN, and maximally align the spiking rate of a neuron with its corresponding charge. This framework enables us to design a novel explicit current control (ECC) method for the CNN-to-SNN conversion which considers multiple objectives at the same time during the conversion, including accuracy, latency, and energy efficiency. We conduct an extensive set of experiments on different neural network architectures, e.g., VGG, ResNet, and DenseNet, to evaluate the resulting SNNs. The benchmark datasets include not only the image datasets such as CIFAR-10/100 and ImageNet but also the Dynamic Vision Sensor (DVS) image datasets such as DVS-CIFAR-10. The experimental results show the superior performance of our ECC method over the state-of-the-art.

Learning Disentangled Graph Convolutional Networks Locally and Globally.

Graph convolutional networks (GCNs) emerge as the most successful learning models for graph-structured data. Despite their success, existing GCNs usually ignore the entangled latent factors typically arising in real-world graphs, which results in nonexplainable node representations. Even worse, while the emphasis has been placed on local graph information, the global knowledge of the entire graph is lost to a certain extent. In this work, to address these issues, we propose a novel framework for GCNs, termed LGD-GCN, taking advantage of both local and global information for disentangling node representations in the latent space. Specifically, we propose to represent a disentangled latent continuous space with a statistical mixture model, by leveraging neighborhood routing mechanism locally. From the latent space, various new graphs can then be disentangled and learned, to overall reflect the hidden structures with respect to different factors. On the one hand, a novel regularizer is designed to encourage interfactor diversity for model expressivity in the latent space. On the other hand, the factor-specific information is encoded globally via employing a message passing along these new graphs, in order to strengthen intrafactor consistency. Extensive evaluations on both synthetic and five benchmark datasets show that LGD-GCN brings significant performance gains over the recent competitive models in both disentangling and node classification. Particularly, LGD-GCN is able to outperform averagely the disentangled state-of-the-arts by 7.4% on social network datasets.

[Determination of elements of hypericum perforatum L. in Xinjiang by microwave digestion-ICP-AES].

Sample digestion by microwave is of some merits: simple, rapid, saving agents, nonpollution and being easy digested. HNO3-HClO4 (5:1) was used as a microwave digestion agent. A direct method is reported for the determination of Fe, Ca, Mg, Zn, Mn, Cu, Na, K, Ba, Al, Pb and Cr in Hypericum perforatum L. in Xinjiang by inductively coupled plasma atomic emission spectrometry(ICP-AES). The selection of digestion conditions of the technique is described. The recoveries for these elements were 93.2%-103.0%, and the relative standard deviation (RSD) of the twelve samples were 0.4%-2.9%. According to standard sample GBW07602 bush twigs and leaves, the authors checked the method for accuracy and precision. The results were found to be basically consistent with the reference values. It was proved that the method features satisfactory precision and accuracy. The method was applied to the determination of plant materials with good agreement.

[Determination of twelve elements in ephedrine extraction by microwave digestion-AAS].

The sample digested by microwave are of some merits: simple, rapid, saving agents, nonpollution and being easily digested. Microwave-digested sample with HNO3-H2O2 and AAS method were established for the determination of Fe, Ca, Mg, Zn, Mn, Cu, Na, K, Ni, Cd, Pb and Cr in ephedrine extraction. In this paper, the selection of digestion conditions, including pressure, power, acids and time of digestion was evaluated. According to standard sample GBW08501 Tea-leaves, we checked the method accuracy and precision. The accuracy values are in the range of 91.1%-105%. The RSDs of flame atomic spectrometry and graphite furnace atomic spectrometry are in the range of 0.2%-2.3% and less than 7%, respectively. The results are very consistent with the the reference values. It proved that the method has satisfactory precision and accuracy. The proposed method is simple, rapid, of nonpollution and safe. It was has been applied to the determination of trace elements in ephedrine extraction and provided useful data for production in factory.