Search

Abstract Deep learning-based inpainting models have achieved success in restoring natural images, yet their application to ancient paintings encounters challenges due to the loss of texture, lines, and color. To address these issues, we introduce an ancient painting inpainting model based on dual encoders and contextual information to overcome the lack of feature extraction and detail texture recovery when restoring ancient paintings. Specifically, the proposed model employs a gated encoding branch that aims to minimize information loss and effectively capture semantic information from ancient paintings. A dense multi-scale feature fusion module is designed to extract texture and detail information at various scales, while dilated depthwise separable convolutions are utilized to reduce parameters and enhance computational efficiency. Furthermore, a contextual feature aggregation module is incorporated to extract contextual features, enhancing the overall consistency of the inpainting results. Finally, a color loss function is introduced to ensure color consistency in the restored area, harmonizing it with the surrounding region. The experimental results indicate that the proposed model effectively restores the texture details of ancient paintings, outperforming other methods both qualitatively and quantitatively. Additionally, the model is tested on real damaged ancient paintings to validate its practicality and efficacy.

Abstract Background Polyketide synthases (PKSs) are classified into three types based on their enzyme structures. Among them, type III PKSs, catalyzing the iterative condensation of malonyl-coenzyme A (CoA) with a CoA-linked starter molecule, are important synthases of valuable natural products. However, low efficiency and byproducts formation often limit their applications in recombinant overproduction. Results Herein, a rapid growth selection system is designed based on the accumulation and derepression of toxic acyl-CoA starter molecule intermediate products, which could be potentially applicable to most type III polyketides biosynthesis. This approach is validated by engineering both chalcone synthases (CHS) and host cell genome, to improve naringenin productions in Escherichia coli. From directed evolution of key enzyme CHS, beneficial mutant with ~ threefold improvement in capability of naringenin biosynthesis was selected and characterized. From directed genome evolution, effect of thioesterases on CHS catalysis is first discovered, expanding our understanding of byproduct formation mechanism in type III PKSs. Taken together, a whole-cell catalyst producing 1082 mg L−1 naringenin in flask with E value (evaluating product specificity) improved from 50.1% to 96.7% is obtained. Conclusions The growth selection system has greatly contributed to both enhanced activity and discovery of byproduct formation mechanism in CHS. This research provides new insights in the catalytic mechanisms of CHS and sheds light on engineering highly efficient heterologous bio-factories to produce naringenin, and potentially more high-value type III polyketides, with minimized byproducts formation.

Exploring the common complexities and challenges of translating poetry from English into Chinese is one of the most pressing topics within contemporary Chinese and Western translation studies. This study aims to present the translation history of works by Emily Dickinson, examine the related research in China, and to analyse the translation potential of transferring three characteristics of the original work: hermeticism, intertextuality and prosody. The study is based on a comparative-literary analysis of Emily Dickinson's translations of four verses, namely 1725, 31, 258, 269 and the translations of the contemporary Chinese translator Jiang Feng. The study found that Jiang Feng's Chinese translation exhibits eight main features: 1) syntactic modification; 2) addition; 3) replacement of specific words with those that are broader in semantic sense; 4) compliance with punctuation similarity rules; 5) omission; 6) presence of translator's notes; 7) transmission of the key idea, but partial neglect of form and rhyme; 8) replacement of words with cultural indicators that are inherent to the Chinese language. This study provides sinologist translators with further potential for research. [ABSTRACT FROM AUTHOR]

Edible insects have become a recognized alternative and sustainable source of high-quality proteins and fats for livestock or human consumption. In the production process of black soldier fly (BSF), (Hermetia illucens L. [Diptera: Stratiomyidae]), initial substrate pH is a critical parameter to ensure the best value of insect biomass, life history traits, and quality bio-fertilizer. This study examined the impact of initial pH values on BSF larvae production, development time, and adult longevity. The BSF were reared on artificial diet with initial pH of 2.0, 4.0, 6.0, 8.0, and 10.0; the control was set at 7.0. Final BSF larval weight was significantly greater in substrates having initial pH 6.0 (0.21 g), control 7.0 (0.20 g), and 10.0 (0.20 g) with no significant difference among them, whereas larval weight reared with initial pH 2.0 and 4.0 were lowest at 0.16 g (-23%). Prepupal weight was greatest when larvae were reared on substrates with initial pH 6.0 (0.18 g), control 7.0 (0.19 g), 8.0 (0.18 g), and 10.0 (0.18 g). In contrast, the prepupal weight of larvae reared on diets with initial pH 2.0 was lowest at 0.15 g (-22%). Larval development time was 21.19 d at pH 8.0, about 3 d (12.5%) shorter than that of those reared on diets with initial pH 6.0, 7.0 control, and 10.0. In all treatments, pH shifted to 5.7 after 3-4 d and 8.5 after 16-17 d except for two groups (2.0 and 4.0) where the pH remained slightly acidic 5.0 and 6.5, respectively. [ABSTRACT FROM AUTHOR]

Support vector machine (SVM) is one of the most effective classification methods for cancer detection. The efficiency and quality of a SVM classifier depends strongly on several important features and a set of proper parameters. Here, a series of classification analyses, with one set of photoacoustic data from ovarian tissues ex vivo and a widely used breast cancer dataset- the Wisconsin Diagnostic Breast Cancer (WDBC), revealed the different accuracy of a SVM classification in terms of the number of features used and the parameters selected. A pattern recognition system is proposed by means of SVM-Recursive Feature Elimination (RFE) with the Radial Basis Function (RBF) kernel. To improve the effectiveness and robustness of the system, an optimized tuning ensemble algorithm called as SVM-RFE(C) with correlation filter was implemented to quantify feature and parameter information based on cross validation. The proposed algorithm is first demonstrated outperforming SVM-RFE on WDBC. Then the best accuracy of 94.643% and sensitivity of 94.595% were achieved when using SVM-RFE(C) to test 57 new PAT data from 19 patients. The experiment results show that the classifier constructed with SVM-RFE(C) algorithm is able to learn additional information from new data and has significant potential in ovarian cancer diagnosis. [ABSTRACT FROM AUTHOR]