Effects of S-adenosylmethionine on AfsKRS regulation in pristinamycin biosynthesis in Streptomyces pristinaespiralis.

In Streptomyces pristinaespiralis, AfsKRS system has differential regulation for PI and PII component biosynthesis of pristinamycin, but it is unknown whether S-adenosylmethionine (SAM) plays an important role in the AfsK-AfsR-AfsS signal transduction cascade during pristinamycin production. The possible target of exogenous SAM in the AfsKRS system and the biological role of SAM during the production of PI and PII were investigated using three mutantsΔafsK,ΔafsR andΔafsS defective in signal cascade pathway of AfsKRS. It was found that external SAM had a significant activation of PI production (1.85-fold increase) but had no obvious effect on PII production in the original strain F618 with the normal response of AfsKRS regulation. Addition of SAM resulted in a similar increase in pristinamycin yield in the mutant with defective afsK or afsR, but induced more crucial activation of PI biosynthesis than PII biosynthesis both in ΔafsK (1.65-fold and 1.15-fold increase respectively) and ΔafsR (1.27-fold and 1.09-fold increase respectively). Exogenous SAM only significantly enhanced PII production in ΔafsS (1.1-fold increase). These results could provide valuable insights into the regulatory function of the AfsKRS system in S. pristinaespiralis.

CommonsenseVIS: Visualizing and Understanding Commonsense Reasoning Capabilities of Natural Language Models.

Recently, large pretrained language models have achieved compelling performance on commonsense benchmarks. Nevertheless, it is unclear what commonsense knowledge the models learn and whether they solely exploit spurious patterns. Feature attributions are popular explainability techniques that identify important input concepts for model outputs. However, commonsense knowledge tends to be implicit and rarely explicitly presented in inputs. These methods cannot infer models' implicit reasoning over mentioned concepts. We present CommonsenseVIS, a visual explanatory system that utilizes external commonsense knowledge bases to contextualize model behavior for commonsense question-answering. Specifically, we extract relevant commonsense knowledge in inputs as references to align model behavior with human knowledge. Our system features multi-level visualization and interactive model probing and editing for different concepts and their underlying relations. Through a user study, we show that CommonsenseVIS helps NLP experts conduct a systematic and scalable visual analysis of models' relational reasoning over concepts in different situations.

ShortcutLens: A Visual Analytics Approach for Exploring Shortcuts in Natural Language Understanding Dataset.

Benchmark datasets play an important role in evaluating Natural Language Understanding (NLU) models. However, shortcuts-unwanted biases in the benchmark datasets-can damage the effectiveness of benchmark datasets in revealing models' real capabilities. Since shortcuts vary in coverage, productivity, and semantic meaning, it is challenging for NLU experts to systematically understand and avoid them when creating benchmark datasets. In this paper, we develop a visual analytics system, ShortcutLens, to help NLU experts explore shortcuts in NLU benchmark datasets. The system allows users to conduct multi-level exploration of shortcuts. Specifically, Statistics View helps users grasp the statistics such as coverage and productivity of shortcuts in the benchmark dataset. Template View employs hierarchical and interpretable templates to summarize different types of shortcuts. Instance View allows users to check the corresponding instances covered by the shortcuts. We conduct case studies and expert interviews to evaluate the effectiveness and usability of the system. The results demonstrate that ShortcutLens supports users in gaining a better understanding of benchmark dataset issues through shortcuts, inspiring them to create challenging and pertinent benchmark datasets.

Novel insights into the predominant factors affecting the bioavailability of polycyclic aromatic hydrocarbons in industrial contaminated areas using PLS-developed model.

Bioavailability is recognized as a useful technical standard for risk assessment and pollution rehabilitation. However, knowledge on the bioavailability of polycyclic aromatic hydrocarbons (PAHs) in contaminated site soils is still limited, especially concerning the influential mechanism. With an abundance of soil collections from nine industrial areas in China, the bioavailabilities, as conceptually defined as bioconcentration factors (BCFs) of PAHs were analyzed using biomimetic extraction of hydroxypropyl-ß-cyclodextrin (HPCD). Apart from the total content of PAHs varying with the different pyrogenic sources, the BCFs were greatly dependent on the soil physicochemical properties from the spatial scale and inversely proportional to the number of rings. Pearson correlation analysis indicated a weak relationship between bioavailability and the soil dissolved organic matter (DOM), pH and particle size. To incorporate the soil physicochemical properties and structural characteristics of PAHs determined by density functional theory (DFT), the optimum model for bioavailability was developed for BCFs by partial least square (PLS) analysis. The PLS-derived model was shown to be predictive within the applicability domain (AD). The structural characteristics, e.g., molecular polarizability and frontier orbital energy level that favor the soil adsorption of PAH isomers via dispersion interactions, and electron exchanges were indicated to be more impactful on bioavailability than soil environmental factors. However, soil factors should not be neglected, because the pH, DOM, etc. were significantly influential. It makes sense that the higher DOM causes greater bioavailability via increasing the free-dissolved fractions of PAHs. Interestingly, the effect of pH on bioavailability was spectrally validated by excitation-emission matrix (EEM) fluorescence, showing that the interaction between DOM and pyrene strengthened the fluorescence quenching of chromophores with the decline in pH.

Comparative evaluation and prioritization of key influences on biodegradation of 2,2',4,4'-tetrabrominated diphenyl ether by bacterial isolate B. xenovorans LB400.

Polybrominated diphenyl ethers (PBDEs) are a class of persistent organic pollutants being widely distributed and harmful to human health and wildlife, and the development of sustainable rehabilitation strategies including microbial degradation is of great concern. Although the increasing number of bacteria, especially the broad-spectrum and potent aerobes have been isolated for the efficient removal of PBDEs, the external influences and the corresponding influential mechanism on biodegradation are not fully understood yet. Given the wide-spectrum biodegradability of aerobic bacterial isolate, B. xenovorans LB400 for PBDEs, the dual impacts of many pivotal factors including pH, temperature, presence of dissolved organic matter (DOM) and cadmium ion etc. were comprehensively revealed on biodegradation of 2,2',4,4'-tetrabromodiphenyl ether (BDE-47). Due to the structural resemblance and stimulation of specific enzyme activity in bacteria, the biphenyl as substrates showed the greater capacity than non-aromatic compounds in improving biodegradation. The individual adaptation to neutrality and cultivation at about 30 °C was beneficial for biodegradation since the bacterial cellular viability and enzyme activity was mostly preserved. Although it was possibly good for the induction of hormesis and favorable to enhance the permeability or bioavailability of pollutant, the exceeding increase of Cd2+ or DOM may not give the profitable increase of biodegradation yet for the detrimental effect. For biodegradation, the mechanistic relationship that took account of the integrative correlation with the influential factors was artfully developed using partial least square (PLS) regression technique. Relative to the most significant influence of culture time and initial concentration of BDE-47, the larger relevance of other factors primarily marked as pH and DOM was consecutively shown after the quantitative prioritization. This may not only help understand the influential mechanism but provide a prioritizing regulation strategy for biodegradation of BDE-47. The PLS-derived relationship was validated with the certain predictability in biodegradation, and could be used as an alternative to accelerate a priori evaluation of suitability or improve the feasibility of such bacteria in remediation of PBDEs in the environment.

GNNLens: A Visual Analytics Approach for Prediction Error Diagnosis of Graph Neural Networks.

Graph Neural Networks (GNNs) aim to extend deep learning techniques to graph data and have achieved significant progress in graph analysis tasks (e.g., node classification) in recent years. However, similar to other deep neural networks like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), GNNs behave like a black box with their details hidden from model developers and users. It is therefore difficult to diagnose possible errors of GNNs. Despite many visual analytics studies being done on CNNs and RNNs, little research has addressed the challenges for GNNs. This paper fills the research gap with an interactive visual analysis tool, GNNLens, to assist model developers and users in understanding and analyzing GNNs. Specifically, Parallel Sets View and Projection View enable users to quickly identify and validate error patterns in the set of wrong predictions; Graph View and Feature Matrix View offer a detailed analysis of individual nodes to assist users in forming hypotheses about the error patterns. Since GNNs jointly model the graph structure and the node features, we reveal the relative influences of the two types of information by comparing the predictions of three models: GNN, Multi-Layer Perceptron (MLP), and GNN Without Using Features (GNNWUF). Two case studies and interviews with domain experts demonstrate the effectiveness of GNNLens in facilitating the understanding of GNN models and their errors.

Development of the degradation bacteria in household food waste and analysis of the microbial community in aerobic composting.

By screening the strains and testing different combinations of diverse bacteria, we developed a compound bacterial agent composed of 5 g Bacillus amyloliquefaciens (B2), 10 g Pseudomonas aeruginosa (F4), 5 g Paenibacillus lautus (303), and 10 ml composite strains (DOD) for the degradation of household food waste (HFW). The final mass loss of HFW in aerobic composting with the compound bacteria agent B2+F4+303+DOD (group C) was 84.52%, increased by 20.83% over that loss in natural composting (group A). Analysis of 16S rRNA high-throughput sequencing showed that the phyla in group A and group C mainly included Firmicutes, Proteobacteria, and Cyanobacteria. At the genus level, Pediococcus was the dominant genus in group A, of which the microbial community performed better in maintaining a stable microbial system in the later stage of composting, while Weissella accounted for a larger proportion of group C, which acted well in reducing the final mass of composting. Ochrobactrum was closely related to the removal of odors in the early stage of group C. The relative abundance of compound bacterial agents was always at a rather low level, suggesting that it affected the composting process by changing the proportion of dominant bacteria in the compost.

Congener-specificity, dioxygenation dependency and association with enzyme binding for biodegradation of polybrominated diphenyl ethers by typical aerobic bacteria: Experimental and theoretical studies.

Polybrominated diphenyl ethers (PBDEs) are a group of organic pollutants that have attracted much concerns of scientific community over the ubiquitous distribution, chemical persistence and toxicological risks in the environment. Though a great number of aerobic bacteria have been isolated for the rapid removal of PBDEs, the knowledge about biodegradation characteristics and mechanism is less provided yet. Herein, the congener-specificity of aerobic biodegradation of PBDEs by typical bacteria, i.e. B. xenovorans LB400 was identified with the different biodegradation kinetics, of which the changes were largely hinged on the bromination pattern. The more bromination isomerically at ortho-sites other than meta-sites or the single bromination at one of aromatic rings might always exert the positive effect. The biodegradation of PBDEs should be thermodynamically constrained to some extent because the calculated Gibbs free energy changes of initial dioxygenation by quantum chemical method increased with the increase of bromination. Within the transition state theory, the high correlativity between the apparent biodegradation rates and Gibbs free energy changes implied the predominance and rate-limiting character of initial dioxygenation, while the regioselectivity of dioxygenation at the ortho/meta-sites was also manifested for the more negative charge population. The molecular binding with the active domain of dioxygenase BphA1 in aerobe was firstly investigated using docking approach. As significantly illustrated with the positive relationship, the higher binding affinity with BphA1 should probably signify the more rapid biodegradation. Besides the edge-on π-π stacking of PBDEs with F227 or Y277 and π-cation formulation with histidines (H233, H239) in BphA1, the reticular hydrophobic contacts appeared as the major force to underpin the high binding affinity and rapid biodegradation of PBDEs. Overall, the experimental and theoretical results would not only help understand the aerobic biodegradation mechanism, but facilitate enhancing applicability or strategy development of engineering bacteria for bioremediation of PBDEs in the environment.

Major biotransformation of phthalic acid esters in Eisenia fetida: Mechanistic insights and association with catalytic enzymes and intestinal symbionts.

Phthalic acid esters (PAEs) are an important group of organic pollutants that are widely used as plasticizers in the environment. The PAEs in soil organisms are likely to be biotransformed into a variety of metabolites, and the combined toxicity of PAEs and their metabolites might be more serious than PAEs alone. However, there are only a few studies on PAE biotransformation by terrestrial animals, e.g. earthworms. Herein, the key biotransformation pathways of PAEs and their association with catalytic enzymes and intestinal symbionts in earthworms were studied using in vivo and in vitro incubation approaches. The widely distributed PAE in soil, dibutyl phthalate (DBP), was proven to be biotransformed rapidly together with apparent bioaccumulation in earthworms. The biotransformation of PAE congeners with medium or long side chains appeared to be faster compared with those with short side chains. DBP was biotransformed into butyl methyl phthalate (BMP), monobutyl phthalate (MBP), and phthalic acid (PA) through esterolysis and transesterification. Besides, the generation of small quantities of low-molecular weight metabolites via ß-oxidation, decarboxylation or ring-cleavage, was also observed, especially when the appropriate proportion of NADPH coenzyme was applied to transfer electrons for oxidases. Interestingly, the esterolysis of PAEs was mainly regulated by the cytoplasmic carboxylesterase (CarE) in earthworms, with a Michaelis constant (Km) of 0.416 mM in the catalysis of DBP. The stronger esterolysis in non-intestinal tissues indicated that the CarE was primarily secreted by non-intestinal tissues of earthworms. Additionally, the intestinal symbiotic bacteria of earthworms could respond to PAE stress, leading to the changes in their diversity and composition. The enrichment of some genera e.g. Bacillus and Paracoccus, and the enhancement of metabolism function, e.g. amino acids, energy, lipids biosynthesis and oxidase secretion, indicated their important role in the degradation of PAEs.

M2Lens: Visualizing and Explaining Multimodal Models for Sentiment Analysis.

Multimodal sentiment analysis aims to recognize people's attitudes from multiple communication channels such as verbal content (i.e., text), voice, and facial expressions. It has become a vibrant and important research topic in natural language processing. Much research focuses on modeling the complex intra- and inter-modal interactions between different communication channels. However, current multimodal models with strong performance are often deep-learning-based techniques and work like black boxes. It is not clear how models utilize multimodal information for sentiment predictions. Despite recent advances in techniques for enhancing the explainability of machine learning models, they often target unimodal scenarios (e.g., images, sentences), and little research has been done on explaining multimodal models. In this paper, we present an interactive visual analytics system, M2 Lens, to visualize and explain multimodal models for sentiment analysis. M2 Lens provides explanations on intra- and inter-modal interactions at the global, subset, and local levels. Specifically, it summarizes the influence of three typical interaction types (i.e., dominance, complement, and conflict) on the model predictions. Moreover, M2 Lens identifies frequent and influential multimodal features and supports the multi-faceted exploration of model behaviors from language, acoustic, and visual modalities. Through two case studies and expert interviews, we demonstrate our system can help users gain deep insights into the multimodal models for sentiment analysis.

Functional investigation of AfsKRS regulatory system for pristinamycin biosynthesis in Streptomyces pristinaespiralis.

Three genes encoding AfsK, AfsR, AfsS homologues in Streptomyces pristinaespiralis were studied, respectively, to investigate regulatory role of AfsKRS system for pristinamycin biosynthesis. Transcription change and gene inactivation analysis indicated that these genes had active transcription and positive regulation for the improvement of pristinamycin production in S. pristinaespiralis. The analysis of AfsKRS-defective mutagenesis indicated that there might be a positive correlation between the product of afsK and pristinamycin I biosynthesis, and a negative correlation to pristinamycin II biosynthesis. However, both afsR and afsS might have negative correlation to pristinamycin I production and positive correlation to pristinamycin II production. The effects on pristinamycin production of AfsKRS disruptants by protein kinase inhibitor K252a indicated that AfsR, both not AfsK and AfsS, was the inhibition target of K252a in S. pristinaespiralis, and AfsR should serve as a pleiotropic regulator to have differential regulation on biosynthesis of pristinamycin I and II components. Based on above study, it might be deduced that different signal transduction patterns via AfsK, AfsR, AfsS of AfsKRS system should be involved in respective regulation for biosynthesis of pristinamycin I and II in S. pristinaespiralis. In conclusion, the investigation could give some valuable clues for exploring furtherly regulatory function of AfsKRS system in S. pristinaespiralis.

DeepDrawing: A Deep Learning Approach to Graph Drawing.

Node-link diagrams are widely used to facilitate network explorations. However, when using a graph drawing technique to visualize networks, users often need to tune different algorithm-specific parameters iteratively by comparing the corresponding drawing results in order to achieve a desired visual effect. This trial and error process is often tedious and time-consuming, especially for non-expert users. Inspired by the powerful data modelling and prediction capabilities of deep learning techniques, we explore the possibility of applying deep learning techniques to graph drawing. Specifically, we propose using a graph-LSTM-based approach to directly map network structures to graph drawings. Given a set of layout examples as the training dataset, we train the proposed graph-LSTM-based model to capture their layout characteristics. Then, the trained model is used to generate graph drawings in a similar style for new networks. We evaluated the proposed approach on two special types of layouts (i.e., grid layouts and star layouts) and two general types of layouts (i.e., ForceAtlas2 and PivotMDS) in both qualitative and quantitative ways. The results provide support for the effectiveness of our approach. We also conducted a time cost assessment on the drawings of small graphs with 20 to 50 nodes. We further report the lessons we learned and discuss the limitations and future work.

A novel carbonyl reductase with anti-Prelog stereospecificity for the production of t-butyl 6-cyano-(3R, 5R)-dihydroxyhexanoate.

A novel gene (crc1) from Candida boidinii was cloned and then overexpressed in a recombinant strain BL21(DE3)/pET30a-crc1 of Escherichia coli. The resulting carbonyl reductase was prepared through fermentations using the recombinant strain. The purified enzyme showed an NADPH-dependent activity and specific activity was 4.65 U/mg using t-butyl 6-cyano-(5R)-hydroxy-3-oxohexanoate (ATS-6) as substrate. The enzyme was optimally active at 35 °C and pH 7, respectively. The apparent K m and V max of the enzyme for ATS-6 are 1.5 mM and 21.1 µmol/min mg, respectively, indicating excellent anti-Prelog stereospecificity. Under the optimum condition, t-butyl 6-cyano-(3R,5R)-dihydroxyhexanoate (ATS-7) was prepared with the enzyme with high d.e. value (99.9%) and good conversion (94%) in 4 h, indicating high stereoselectivity and conversion efficiency in biotransformation of ATS-6 to ATS-7.

Correction to: An efficient biocatalytic synthesis of imidazole-4-acetic acid.

In the original publication of the article, the affiliations of authors Jun Huang, Changjiang Lv and Jiaqi Mei were misplaced. The correct information for author affiliations is provided in this correction.

An efficient biocatalytic synthesis of imidazole-4-acetic acid.

OBJECTIVE: To develop a new and efficient biocatalytic synthesis method of imidazole-4-acetic acid (IAA) from L-histidine (L-His). RESULTS: L-His was converted to imidazole-4-pyruvic acid (IPA) by an Escherichia coli whole-cell biocatalyst expressing membrane-bound L-amino acid deaminase (mL-AAD) from Proteus vulgaris firstly. The obtained IPA was subsequently decarboxylated to IAA under the action of H2O2. Under optimum conditions, 34.97 mM IAA can be produced from 50 mM L-His, with a yield of 69.9%. CONCLUSIONS: Compared to the traditional chemical synthesis, this biocatalytic method for IAA production is not only environmentally friendly, but also more cost effective, thus being promising for industrial IAA production.

Physiology-Oriented Engineering Strategy to Improve Gamma-Aminobutyrate Production in Lactobacillus brevis.

Gamma-aminobutyrate (GABA) is an important chemical in the pharmaceutical field. GABA-producing lactic acid bacteria (LAB) offer the opportunity of developing this health-oriented product. In this study, the gadA, gadB, gadC, gadCB, and gadCA gene segments of Lactobacillus brevis were cloned into pMG36e, and strain Lb. brevis/pMG36e-gadA was selected for thorough characterization in terms of GABA production after analysis of GAD activities. Subsequently, a physiology-oriented engineering strategy was adopted to construct an FoF1-ATPase deficient strain NRA6 with higher GAD activity. As expected, strain NRA6 could produce GABA at a concentration of 43.65 g/L with a 98.42% GABA conversion rate in GYP fermentation medium, which is 1.22-fold higher than that obtained by the wild-type strain in the same condition. This work demonstrates how the acid stress response mechanisms of LAB can be employed to develop cell factories with improved production efficiency and contributes to research into the development of the physiology-oriented engineering.

Nup62, associated with spindle microtubule rather than spindle matrix, is involved in chromosome alignment and spindle assembly during mitosis.

An increasing number of the active mitotic functions of nucleoporins in the distinct steps of mitosis have been assigned over the past few years. As one of FG-repeats containing nucleoporins, Nup62 has been found to be involved in nuclear transport, cell migration, virus infection, and cell cycle regulation. However, the role and mechanism of Nup62 in mitotic regulation have not been fully revealed. In this paper, it was revealed that a fraction of Nup62 was associated with mitotic spindle microtubule instead of spindle matrix, and the localization of Nup62 in the mitotic spindle depended on its three coiled-coil domains rather than Crm1, although Nup62 strongly interacted with Crm1 during mitosis. Moreover, depletion of Nup62 by small interference of RNA seriously induced the defects of chromosome alignment and spindle assembly although the bipolar spindle was observed in most of the Nup62 knock-down cells. Notably, congression of polar chromosome defect was observed in more than 30% of Nup62 knock-down cells. These findings revealed that Nup62 was a novel mitotic spindle associated nucleoporin and involved in chromosome alignment and spindle assembly.

Rapamycin preserves gut homeostasis during Drosophila aging.

Gut homeostasis plays an important role in maintaining the overall body health during aging. Rapamycin, a specific inhibitor of mTOR, exerts prolongevity effects in evolutionarily diverse species. However, its impact on the intestinal homeostasis remains poorly understood. Here, we demonstrate that rapamycin can slow down the proliferation rate of intestinal stem cells (ISCs) in the aging guts and induce autophagy in the intestinal epithelium in Drosophila. Rapamycin can also significantly affect the FOXO associated genes in intestine and up-regulate the negative regulators of IMD/Rel pathway, consequently delaying the microbial expansion in the aging guts. Collectively, these findings reveal that rapamycin can delay the intestinal aging by inhibiting mTOR and thus keeping stem cell proliferation in check. These results will further explain the mechanism of healthspan and lifespan extension by rapamycin in Drosophila.

[Peripheral blood cell factors of Graves ophthalmopathy and effect of intervention with tripterygium glycosides].

To explore the effect of tripterygium glycosides on the level of peripheral blood cell factors of Graves ophthalmopathy (GO). In the study, 64 patients of GO in moderate-severe acute stage were selected, and randomly divided into the treatment group (32 cases) and the control group (32 cases). Both of the two groups were provided with basic treatment. The control group was added with prednisone(0. 75 mg kg-1 d-1 ), which gradually reduced (by 5-10 mg week-1 )to the minimum dose of 5 mg d-1. The treatment group was treated with 20 mg tripterygium glycosides, three times a day. One therapy course is three months. The levels of peripheral blood cells(TNF-alpha , IL-2, IL-10, IFN-gamma)of the two groups before and after the treatment and the clinical efficacy were observed. The study indicated that, before the treatment, TNF-alpha, IL-2, IFN-gamma in both groups were significantly higher than that in the health group, but with IL-10 notably lower than the healthy group. After the treatment, TNF-a, IL-2, IFN-gamma in the treatment group significantly decreased, but with IL-10 significantly increasing (P <0. 01). After the treatment, the two groups showed significant difference (P <0. 01). The total clinical efficacy in the treatment group was 88. 10% , and that in the control group was 57. 14% (P <0. 01). After the treatment, the two groups showed significant changes in the exophthalmos degree (P < 0. 01). The results showed that the level of peripheral blood cells (TNF-alpha,IL-2, IL-10, IFN-gamma)of GO patients was positively correlated with the severity of ocular disease. The combined therapy of tripterygium glycosides and methimazole show such advantages as low side effect and high clin-

Probing the molecular mechanisms for pristinamycin yield enhancement in Streptomyces pristinaespiralis.

The mechanisms for the enhancement of pristinamycin production in the high-yielding recombinants of Streptomyces pristinaespiralis obtained by genome shuffling were investigated by quantitative real-time PCR (Q-PCR) and amplified fragment length polymorphism (AFLP) techniques. Q-PCR analysis showed that snaB and snbA involved, respectively, in the biosynthesis of pristinamycins II and I component had more extended high expression in the recombinant than that in the ancestor during fermentation process, indicating their expression changes might be key factors during the biosynthesis of the antibiotic. In addition, the antecedent establishment of the high self-resistance to pristinamycin, because ptr resistance gene started high-level expression ahead of the onset of the antibiotic production in the recombinant, might also lead to the increase of the antibiotics yield. AFLP analysis of these recombinants revealed genome variation of two novel genes, the homologs of AfsR regulatory gene and transposase gene, indicating these two gene variations were probably responsible for yield improvement of pristinamycin. This study provided several potential molecular clues for pristinamycin yield enhancement.