Search

Zero-Shot Sketch-Based Image Retrieval (ZS-SBIR) is a novel cross-modal retrieval task, where abstract sketches are used as queries to retrieve natural images under zero-shot scenario. Most existing methods regard ZS-SBIR as a traditional classification problem and employ a cross-entropy or triplet-based loss to achieve retrieval, which neglect the problems of the domain gap between sketches and natural images and the large intra-class diversity in sketches. Toward this end, we propose a novel Domain-Smoothing Network (DSN) for ZS-SBIR. Specifically, a cross-modal contrastive method is proposed to learn generalized representations to smooth the domain gap by mining relations with additional augmented samples. Furthermore, a category-specific memory bank with sketch features is explored to reduce intra-class diversity in the sketch domain. Extensive experiments demonstrate that our approach notably outperforms the state-of-the-art methods in both Sketchy and TU-Berlin datasets. Our source code is publicly available at https://github.com/haowang1992/DSN., Comment: Accepted to IJCAI 2021

Learning feature embedding directly from images without any human supervision is a very challenging and essential task in the field of computer vision and machine learning. Following the paradigm in supervised manner, most existing unsupervised metric learning approaches mainly focus on binary similarity in Euclidean space. However, these methods cannot achieve promising performance in many practical applications, where the manual information is lacking and data exhibits non-Euclidean latent anatomy. To address this limitation, we propose an Unsupervised Hyperbolic Metric Learning method with Hierarchical Similarity. It considers the natural hierarchies of data by taking advantage of Hyperbolic metric learning and hierarchical clustering, which can effectively excavate richer similarity information beyond binary in modeling. More importantly, we design a new loss function to capture the hierarchical similarity among samples to enhance the stability of the proposed method. Extensive experimental results on benchmark datasets demonstrate that our method achieves state-of-the-art performance compared with current unsupervised deep metric learning approaches.

Xiaolei Liang,1,* Jiexi Yan,2,* Yingwei Lu,3 Shan Liu,4 Xiaojing Chai5 1Key Laboratory for Gynecologic Oncology Gansu Province, Department of Obstetrics and Gynecology, The First Hospital of Lanzhou University, Lanzhou, People’s Republic of China; 2The Precision Medicine Laboratory, The First Hospital of Lanzhou University, Lanzhou, People’s Republic of China; 3Gansu Provincial Center for Disease Control and Prevention, Lanzhou, People’s Republic of China; 4The First Clinical Medicine School, Lanzhou University, Lanzhou, People’s Republic of China; 5The Key Laboratory, The First Hospital of Lanzhou University, Lanzhou, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xiaojing ChaiThe Key Laboratory, The First Hospital of Lanzhou University, No. 1, Donggang Road, Lanzhou, People’s Republic of ChinaEmail chaixiaoj@163.comPurpose: Increasingly complex diseases require novel drugs for their treatment. Antimicrobial peptides (AMPs) are promising candidate treatments due to their broad existence and special characteristics. However, the current understanding of AMPs is not sufficient to allow them to be produced commercially for clinical use.Materials and Methods: Melectin, from the venom of the cleptoparasitic bee Melecta albifrons, does not exhibit sequence homology with other wasp venom peptides. To investigate this more deeply, we explored the antibacterial and antitumor activities of Melectin and related mechanisms.Results: Our results demonstrate that Melectin possesses antimicrobial properties against standard sensitive/clinical drug-resistant bacteria strains as well as antitumor activity. It has an α-helix form and exhibits moderate cytotoxicity. Its action mechanisms are involved with membrane interfering and DNA binding. The membrane interfering effect was distinct between different phospholipid compositions.Conclusion: We found that Melectin may serve as a new potential template in the battle against multidrug resistance, and our study indicated that there are promising prospects for medically applicable drugs based on AMPs.Keywords: antimicrobial peptide, melectin, antibacterial, antitumor, action mechanism

Alzheimer's disease (AD) is a severe type of neurodegeneration which worsens human memory, thinking and cognition along a temporal continuum. How to identify the informative phenotypic neuroimaging markers and accurately predict cognitive assessment are crucial for early detection and diagnosis Alzheimer's disease. Regression models are widely used to predict the relationship between imaging biomarkers and cognitive assessment, and identify discriminative neuroimaging markers. Most existing methods use different matrix norms as the similarity measures of the empirical loss or regularization to improve the prediction performance, but ignore the inherent geometry of the cognitive data. To tackle this issue, in this paper we propose a novel robust matrix regression model with imposing Wasserstein distances on both loss function and regularization. It successfully integrate Wasserstein distance into the regression model, which can excavate the latent geometry of cognitive data. We introduce an efficient algorithm to solve the proposed new model with convergence analysis. Empirical results on cognitive data of the ADNI cohort demonstrate the great effectiveness of the proposed method for clinical cognitive predication.

Research question The aim of this study was to investigate how female age affects the predictive effect of sperm DNA fragmentation index (DFI) on clinical outcomes with assisted reproductive technology. Design A total of 2371 patients, comprising 2115 men with a normal DFI (≤30), 256 men with a high DFI (>30) and women of different ages, were recruited and investigated. All patients had normal chromosome karyotypes and were undergoing their first fresh IVF or intracytoplasmic sperm injection (ICSI) cycles. Clinical outcomes were analysed according to the two DFI groups and female age ≤30 and >30 years. Binary logistic regression analysis was performed to identify factors associated with clinical outcome. Results The proportion of couples with at least one good-quality embryo in the DFI ≤30 group was higher than that in the DFI >30 group. When female age exceeded 30 years, clinical pregnancy rate and the proportion of couples with good-quality embryos in the DFI >30 group were lower compared with DFI ≤30; however, there were no differences in outcomes for female age ≤30 years according to DFI. When DFI >30, the cut-off value of female age was 30.5 for detecting clinical pregnancy; the sensitivity was 62.0%, and the specificity was 63.6%. Clinical pregnancy rate and proportion of couples with good-quality embryos were lower in the DFI >30 versus DFI ≤30 group with a female age above 30 years for IVF but not for ICSI. Conclusion Female age has a negative effect and should be considered in predicting the effects of sperm DNA fragmentation on pregnancy outcomes.

Multidrug-resistant bacteria are emerging as a global threat, making the search for alternative compounds urgent. Antimicrobial peptides (AMPs) became a promising hotspot due to their distinct action mechanism and possibility to be used as an alternative or complement to traditional antibiotics. However, gaining a better understanding about the relationship between antimicrobial peptides structure and its bioactivity is crucial for the development of next generation of antimicrobial agents. NK-2, derived from mammalian protein NK-lysin, has potent antitumor and bactericidal abilities. As proline was considered to be an effective α-helix breaker due to its restricted conformation, to better comprehend the effects of proline in the structure-activity relationship of NK-2, we constructed two NK-2 analogs. We examined the biological activities of NK-2 and its proline substitution analogs and analyzed the resulting conformational changes. Our results showed that introducing proline into the primary sequence of NK-2 significantly decreased the antitumor, antibacterial, and cytotoxic effects, as well as DNA binding activity by changing the α-helix content. However, α-helical content was not the only determining factor, the position of proline inserted was also critical. This study will allow for clearer insight into the role of proline in structure and bioactivity of NK-2 and provide a foundation for future studies.

Quercetin is widely known as potent natural antioxidant and scavenger of reactive oxygen species (ROS) and nitric oxide both in vitro and in vivo. Quercetin has a wide range of biological functions and health-promoting effects. There are more and more interests in the addition of this flavonol to various traditional food products. However, the in vitro toxicity of quercetin to mature human sperm remains unknown. In this study, we investigated the in vitro effects of quercetin on human sperm functions. The results showed that the total sperm motility were significantly inhibited compared to the controls following exposure to 100, 200 and 400µM quercetin for 6 and 12h; quercetin did not affect human sperm viability. The acrosome reaction and capacitation induced by progesterone were dose-dependently inhibited by quercetin. Furthermore, quercetin induced a significantly decrease of human sperm [Ca

The incidence of life-threatening invasive fungal infections increased significantly in recent years. However, the antifungal therapeutic options are very limited. Antimicrobial peptides are a class of potential lead chemical for the development of novel antifungal agents. Antimicrobial peptide polybia-CP was purified from the venom of the social wasp Polybia paulista. In this study, we synthesized polybia-CP and determined its antifungal effects against a series of Candidian species. Our results showed that polybia-CP has potent antifungal activity and fungicidal activity against the tested fungal cells with a proposed membrane-active action mode. In addition, polybia-CP could induce the increase of cellular reactive oxygen species production, which would attribute to its antifungal activity. In conclusion, the present study suggests that polybia-CP has potential as an antifungal agent or may offer a new strategy for antifungal therapeutic option.