Search

[Display omitted] • Janus nanosheets composed of AuNPs/PdNPs and rGO were prepared by a simple method. • The obtained Janus nanosheets were useful in emulsion stabilization and catalysis. • Largely enhanced hydrogenation and photocatalytic degradation were demonstrated. Interface/emulsion catalysis is an effective approach for many organic and inorganic chemical reactions. Positioning catalysts at the interface is essential for highly efficient interface/emulsion catalysis. The sheet-like Janus nanostructures may have great advantages in stabilizing emulsions and improving reaction efficiency. In this work, by using a simple emulsion reaction and photodeposition, we successfully prepare the Janus nanosheets with opposite sides composed of AuNPs/PdNPs and reduced graphene oxide (rGO) respectively, referred to as AuNP-rGO/PdNP-rGO Janus nanosheets. Due to the distinct hydrophilicities on opposite sides, the prepared Janus nanosheets demonstrate high efficiency in emulsion stabilization and emulsion catalysis. By introducing the prepared PdNP-rGO and AuNP-rGO Janus nanosheets in emulsion reactions, highly efficient hydrogenation of p -nitroanisole and photocatalytic degradation of Nile red are demonstrated. This work provides a simple and cost-effective method to prepare Janus nanosheets and realize highly efficient emulsion reactions, and this strategy can be extended to promote many other interface/emulsion reactions. [ABSTRACT FROM AUTHOR]

Nasopharyngeal carcinoma (NPC) is one of the Epstein-Barr virus (EBV)-associated malignancies and has a distinct geographical distribution. The high mortality rates of NPC patients with advanced and recurrent disease highlight the urgent need for biomarkers for early diagnosis and effective treatments. In this study, we developed DNA aptamers that specifically bind to EBV positive NPC cells by the Cell-SELEX procedure. We further identified the EphA2 (ephrin type-A receptor 2)/CD98hc (CD98 heavy chain) complex as the potential target of the aptamer EA-3 by combining aptamer-based separation and mass spectrometry analysis. Our results revealed for the first time that EphA2 colocalized with CD98hc at the plasma membrane and EphA2 coimmunoprecipitated with CD98hc, which may serve as a starting point for exploring the potential functions of the complex of EphA2 and CD98hc in NPCs. Here, we demonstrated that aptamers can be useful for the identification of protein complexes on the surface of cancer cells. • We report DNA aptamers that specifically bind to EBV positive nasopharyngeal carcinoma. • EphA2 colocalizes and coimmunoprecipitate with CD98hc. • The DNA aptamer binds with the complex of EphA2 and CD98hc. [ABSTRACT FROM AUTHOR]

Multilingual meta learning has emerged as a promising paradigm for transferring knowledge from source languages to facilitate the learning of low-resource target languages. Loss functions are a type of meta-knowledge that is crucial to the effective training of neural networks. However, the misalignment between the loss functions and the learning paradigms of meta learning degrades the network's performance. To address this challenge, we propose a new method called Task-based Meta PolyLoss (TMPL) for meta learning. By regarding speech recognition tasks as normal samples and applying PolyLoss to the meta loss function, TMPL can be denoted as a linear combination of polynomial functions based on task query loss. Theoretical analysis shows that TMPL improves meta learning by enabling attention adjustment across different tasks, which can be tailored for different datasets. Experiments on three datasets demonstrated that gradient-based meta learning methods achieve superior performance with TMPL. Furthermore, our experiments validate that the task-based loss function effectively mitigates the misalignment issue. • Multilingual meta-learning ignore the misalignment problem between the loss functions and the learning paradigms of meta learning. • Task-based Meta PolyLoss is a more general meta-loss function, which can be denoted as a linear combination of polynomial functions. • Theoretical analysis shows that TMPL adjust attention across different tasks, which can be tailored for different datasets. • Experiments on three datasets show that verify the effectiveness of TMPL, especially in very low-resource setting. • Experiments analysis shows the task-based loss function effectively mitigates the misalignment issue and is more appropriate for meta-learning. [ABSTRACT FROM AUTHOR]

Low-resource automatic speech recognition is a challenging task. To resolve this issue, multilingual meta-learning learns a better model initialization from many source languages, allowing for rapid adaption to target languages. However, differences in data scales and learning difficulties vary greatly from one language to another. As a result, the model favors large-scale and simple source languages. Moreover, the shared semantic space of various languages is difficult to learn due to a lack of restrictions on multilingual pre-training. In this paper, we propose a meta adversarial learning approach to address this problem. The meta-learner will be guided to learn language-independent information by using an adversarial auxiliary objective of language identification, which makes the shared semantic space more compact and improves model generalization. Additionally, we optimize adversarial training using Wasserstein distance and temporal normalization, enabling more stable and simple training. Experiment results on IARPA BABEL and OpenSLR show a significant performance improvement. It also outperforms state-of-the-art results by a large margin in all target languages, and especially in few-shot settings. Finally, we demonstrate how our method is superior by using t-SNE visualization. • Multilingual meta-learning ignore the imbalance problem across source languages. • Meta adversarial learning can build a more compact semantic space and improves the generalization capability of the model. • Optimized adversarial learning will be more stable by using Wasserstein distance and temporal normalization. • Experiments on target languages verify the effectiveness of meta adversarial learning, especially in very low-resource setting. • Experiments analysis shows the principle and superiority of meta adversarial learning. [ABSTRACT FROM AUTHOR]

Enhancing aminopeptidase fermentation using cheap substrate is very necessary for industrial production and application of aminopeptidase. The suitability of using raw soybean meal (SBM) as substrate for aminopeptidase production by Bacillus licheniformis SWJS33 in submerged fermentation was first evaluated. SBM alone provided both nitrogen and carbon sources for SWJS33 to produce aminopeptidase. Supplementation of 3 g/L glucose and 15 g/L yeast extract further improved the aminopeptidase yield by 0.63 and 1.3 folds, respectively. Addition of yeast extract altered the metabolic process of SWJS33, including shortening the lag phase, increasing the maximum specific cell growth rate and maximum specific aminopeptidase formation rate, and changing the utilization process of nutrients during the fermentation. What's more, SWJS33 secreted aminopeptidase prior to endoproteases when quick-release nitrogen (e.g. yeast extract) was supplied. High dissolved oxygen was demanded during aminopeptidase fermentation. The aminopeptidase production was successfully up scaled to 20 L fermenter. • SBM alone satisfied the basic nutritional demand for aminopeptidase production. • Supplying glucose/yeast extract (YE) significantly improved the enzyme yield. • Glucose/YE addition altered the metabolic process of SWJS33. • SWJS33 secreted more aminopeptidase prior to endoproteases when YE was supplied. • High dissolved oxygen was demanded during aminopeptidase fermentation. [ABSTRACT FROM AUTHOR]