Merge-and-Split Graph Convolutional Network for Skeleton-Based Interaction Recognition.

We introduce an innovative approach to address a significant challenge in interaction recognition, specifically the capture of correlation features between different interaction body parts. These features are often overlooked by traditional graph convolution networks commonly used in interaction recognition tasks. Our solution, the Merge-and-Split Graph Convolutional Network, takes a unique perspective, treating interaction recognition as a global problem. It leverages a Merge-and-Split Graph structure to effectively capture dependencies between interaction body parts. To extract the essential interaction features, we introduce the Merge-and-Split Graph Convolution module, which seamlessly combines the Merge-and-Split Graph with Graph Convolutional Networks. This fusion enables the extraction of rich semantic information between adjacent joint points. In addition, we introduce a Short-term Dependence module designed to extract joint and motion characteristics specific to each type of interaction. Furthermore, to extract correlation features between different hierarchical sets, we present the Hierarchical Guided Attention Module. This module plays a crucial role in highlighting the relevant hierarchical sets that contain essential interaction information. The effectiveness of our proposed model is demonstrated by achieving state-of-the-art performance on 2 widely recognized datasets, namely, the NTU60 and NTU120 interaction datasets. Our model's efficacy is rigorously validated through extensive experiments, and we have made the code available for the research community at https://github.com/wanghq05/MS-GCN/.

In Silico Prediction of Chemical Acute Dermal Toxicity Using Explainable Machine Learning Methods.

The research on acute dermal toxicity has consistently been a crucial component in assessing the potential risks of human exposure to active ingredients in pesticides and related plant protection products. However, it is difficult to directly identify the acute dermal toxicity of potential compounds through animal experiments alone. In our study, we separately integrated 1735 experimental data based on rabbits and 1679 experimental data based on rats to construct acute dermal toxicity prediction models using machine learning and deep learning algorithms. The best models for the two animal species achieved AUC values of 78.0 and 82.0%, respectively, on 10-fold cross-validation. Additionally, we employed SARpy to extract structural alerts, and in conjunction with Shapley additive explanation and attentive FP heatmap, we identified important features and structural fragments associated with acute dermal toxicity. This approach offers valuable insights for the detection of positive compounds. Moreover, a standalone software tool was developed to make acute dermal toxicity prediction easier. In summary, our research would provide an effective tool for acute dermal toxicity evaluation of pesticides, cosmetics, and drug safety assessment.

AttentiveSkin: To Predict Skin Corrosion/Irritation Potentials of Chemicals via Explainable Machine Learning Methods.

Skin Corrosion/Irritation (Corr./Irrit.) has long been a health hazard in the Globally Harmonized System (GHS). Several in silico models have been built to predict Skin Corr./Irrit. as an alternative to the increasingly restricted animal testing. However, current studies are limited by data amount/quality and model availability. To address these issues, we compiled a traceable consensus GHS data set comprising 731 Corr., 1283 Irrit., and 1205 negative (Neg.) samples from 6 governmental databases and 2 external data sets. Then, a series of binary classifiers were developed with five machine learning (ML) algorithms and six molecular representations. For 10-fold cross-validation, the best Corr. vs Neg. classifier achieved an Area Under the Receiver Operating Characteristic Curve (AUC) of 97.1%, while the best Irrit. vs Neg. classifier achieved an AUC of 84.7%. Compared with existing in silico tools on external validation, our Attentive FP classifiers showed the highest metrics on Corr. vs Neg. and the second highest accuracy on Irrit. vs Neg. The SHapley Additive exPlanation approach was further applied to figure out important molecular features, and the attention weights were visualized to perform interpretable prediction. Structural alerts associated with Skin Corr./Irrit. were also identified. The interpretable Attentive FP classifiers were integrated into the software AttentiveSkin at https://github.com/BeeBeeWong/AttentiveSkin. The conventional ML classifiers are also provided on our platform admetSAR at http://lmmd.ecust.edu.cn/admetsar2/. Considering the data deficiency and the limited model availability of Skin Corr./Irrit., we believe that our data set and models could facilitate chemical safety assessment and relevant studies.

Extended ray-mapping method based on differentiable ray-tracing for non-paraxial and off-axis freeform illumination lens design.

The ray-mapping method has been widely used for designing freeform illumination lenses. However, in non-paraxial or off-axis situations, it remains challenging to obtain an integrable ray-mapping, often requiring a complex iterative correction process for the initial mapping. To address this challenge, we propose an extended ray-mapping method that incorporates differentiable ray-tracing into the design pipeline of the ray-mapping method. This enables accurate surface construction according to ray-mapping and efficient shape correction based on irradiance distribution. The proposed method involves two optimization stages. In the first stage, the freeform surface is preliminarily optimized to closely match the optimal transport mapping. The obtained freeform surface is then further optimized in the second stage to minimize the divergence between the target and simulated irradiance distributions. Additionally, the mean curvature of the freeform surface is also constrained in the second stage to facilitate the fabrication of the final freeform surface. Non-paraxial illumination lenses and off-axis illumination lenses have been designed using the proposed method within ten minutes, and simulations demonstrate that the approach is effective and robust.

Ultrarapid Nanomanufacturing of High-Quality Bimetallic Anode Library toward Stable Potassium-Ion Storage.

Bimetallic alloy nanomaterials are promising anode materials for potassium-ion batteries (KIBs) due to their high electrochemical performance. The most well-adopted fabrication method for bimetallic alloy nanomaterials is tube furnace annealing (TFA) synthesis, which can hardly satisfy the trade-off among granularity, dispersity and grain coarsening due to mutual constraints. Herein, we report a facile, scalable and ultrafast high-temperature radiation (HTR) method for the fabrication of a library of ultrafine bimetallic alloys with narrow size distribution (≈10-20 nm), uniform dispersion and high loading. The metal-anchor containing heteroatoms (i.e., O and N), ultrarapid heating/cooling rate (≈103  K s-1 ) and super-short heating duration (several seconds) synergistically contribute to the successful synthesis of small-sized alloy anodes. As a proof-of-concept demonstration, the as-prepared BiSb-HTR anode shows ultrahigh stability indicated by negligible degradation after 800 cycles. The in situ X-ray diffraction reveals the K+ storage mechanism of BiSb-HTR. This study can shed light on the new, rapid and scalable nanomanufacturing of high-quality bimetallic alloys toward extended applications of energy storage, energy conversion and electrocatalysis.

Digestion and utilization of plant-based diets by transgenic pigs secreting ß-glucanase, xylanase, and phytase in their salivary glands.

Novel transgenic (TG) pigs co-expressing three microbial enzymes, ß-glucanase, xylanase, and phytase, in their salivary glands were previously generated, which exhibited reduced phosphorus and nitrogen emissions and improved growth performances. In the present study, we attempted to explore the age-related change of the TG enzymic activity, the residual activity of the enzymes in the simulated gastrointestinal tract, and the effect of the transgenes on the digestion of nitrogen and phosphorus content in the fiber-rich, plant-based diets. Results showed that all the three enzymes were stably expressed over the growing and finishing periods in the F2 generation TG pigs. In simulated gastric juice, all the three enzymes exhibited excellent gastrointestinal environment adaptability. The apparent total tract digestibility of phosphorus was increased by 69.05% and 499.64%, while fecal phosphate outputs were decreased by 56.66% and 37.32%, in the TG pigs compared with the wild-type littermates fed with low non-starch polysaccharides diets and high fiber diets, respectively. Over half of available phosphorus and water-soluble phosphorus in fecal phosphorus were reduced. We also found the performance of phosphorus, calcium, and nitrogen retention rates were significantly improved, resulting in faster growth performance in TG pigs. The results indicate that TG pigs can effectively digest the high-fiber diets and exhibit good growth performance compared with wild type pigs.

Mini-dCas13X-mediated RNA editing restores dystrophin expression in a humanized mouse model of Duchenne muscular dystrophy.

Approximately 10% of monogenic diseases are caused by nonsense point mutations that generate premature termination codons (PTCs), resulting in a truncated protein and nonsense-mediated decay of the mutant mRNAs. Here, we demonstrate a mini-dCas13X-mediated RNA adenine base editing (mxABE) strategy to treat nonsense mutation-related monogenic diseases via A-to-G editing in a genetically humanized mouse model of Duchenne muscular dystrophy (DMD). Initially, we identified a nonsense point mutation (c.4174C>T, p.Gln1392*) in the DMD gene of a patient and validated its pathogenicity in humanized mice. In this model, mxABE packaged in a single adeno-associated virus (AAV) reached A-to-G editing rates up to 84% in vivo, at least 20-fold greater than rates reported in previous studies using other RNA editing modalities. Furthermore, mxABE restored robust expression of dystrophin protein to over 50% of WT levels by enabling PTC read-through in multiple muscle tissues. Importantly, systemic delivery of mxABE by AAV also rescued dystrophin expression to averages of 37%, 6%, and 54% of WT levels in the diaphragm, tibialis anterior, and heart muscle, respectively, as well as rescued muscle function. Our data strongly suggest that mxABE-based strategies may be a viable new treatment modality for DMD and other monogenic diseases.

A Novel Hierarchical Structure of SnCu2Se4/d-Ti3C2Tx/NPC for a Lithium/Sodium Ion Battery and Hybrid Capacitor with Long-Term Cycling Stabilities.

To alleviate kinetics imbalance and capacity insufficiency simultaneously, a novel hierarchical structure (SnCu2Se4/d-Ti3C2Tx/NPC) composed of delaminated Ti3C2Tx, SnCu2Se4 nanoparticles, and N-doped porous carbon layers is designed as a battery-type anode for lithium/sodium ion hybrid capacitor (LIC/SIC). The combination of SnCu2Se4 nanoparticles with high specific capacity, d-Ti3C2Tx with accelerated ion diffusion path, and NPC with enhanced electronic conductivity makes the SnCu2Se4/d-Ti3C2Tx/NPC composite possess excellent cycling stabilities in half-cell lithium-ion and sodium-ion batteries (LIB and SIB), with capacities of 114 mAh g-1 after 6000 cycles at 10 A g-1 for LIB and 296 mAh g-1 after 900 cycles at 1.0 A g-1 for SIB. The rate performance is also outstanding, with recovered capacity of 738 mAh g-1 at 0.1 A g-1 after cycles at current densities up to 50 A g-1 for LIB. Subsequently, LIC and SIC based on the SnCu2Se4/d-Ti3C2Tx/NPC anode and activated carbon cathode exhibit high energy densities of 147.9 and 158.6 Wh kg-1 at a power density of 100 W kg-1, respectively. They also possess distinctive long lifespans with capacity retentions of 78 and 81% after 10,000 cycles at 1.0 A g-1, respectively, demonstrating the feasibility of SnCu2Se4/d-Ti3C2Tx/NPC toward energy devices requiring high energy density, power density, and long-term stability.

A novel strategy to investigate the factors regulating the Treg to Tfr transition during acute viral infection.

Follicular regulatory T cells (Tfrs), a specialized subset of regulatory T cells (Tregs), have a particular role in the control of follicular helper T cell-driven germinal center (GC) responses. Following differentiation signals similar to those received by follicular helper T cells (Tfhs), Tfrs gain expression of characteristic chemokine receptors and transcription factors, such as CXCR5 and Bcl-6, allowing them to migrate into the B-cell follicle and perform in situ suppression. Thus, together with Tfhs, Tfrs help maintaining an optimized GC-reaction. However, the mechanism underlying the Treg-to-Tfr transition remains obscure. Here, we established a highly reproducible protocol for investigating the differentiation of Tregs into Tfrs by constructing spleen-chimeric mice combined with retrovirus transduction. We demonstrated that using this strategy, over 4 folds of Tregs could differentiate into Tfrs in Bcl-6 overexpression group compared to control counterparts (Migr1), and Bcl-6 could efficiently promote Tfr differentiation during acute viral infection. Hence, this method provides us an easy access to investigate the factors that regulate the differentiation program that converts Tregs into Tfrs, which will enhance our understanding of the networks regulating GC-reaction and shed new light on the molecular basis of immune homeostasis.

Prevents kudzu starch from agglomeration during rapid pasting with hot water by a non-destructive superheated steam treatment.

Superheated steam (SST) at different moisture contents (10% ∼ 30%) was used to prevent the agglomeration of kudzu starch during rapid pasting with hot water. Changes in pasting-related properties and multi-scale structures were investigated. At moisture content of 20%, SST dramatically reduced the agglomeration rate from 42.20% to 2.97% without destroying the microstructure of kudzu starch or deteriorating the rheological properties of kudzu starch paste, which was superior to the conventional pre-gelatinization treatment. The agglomeration was prevented mainly by decreasing the swelling power and increasing the pasting temperature of kudzu starch. The slight disruption of multi-scale structures may facilitate faster water absorption by kudzu starch, but it was not the primary prevention mechanism. Moreover, the solubility of kudzu starch was not related to the agglomeration, since it was significantly decreased by SST. Our findings could provide new insights into the rapid pasting of starchy powders or flours with hot water.

Adaptation of Gut Microbiome to Transgenic Pigs Secreting ß-Glucanase, Xylanase, and Phytase.

We previously generated transgenic pigs with enhanced growth rate and reduced nutrient loss. However, the composition of their gut microbiome is unknown. In this study, we successfully generated EGFP marker-free transgenic (MF-TG) pigs with high expression levels of microbial ß-glucanase, xylanase, and phytase in the parotid gland. We collected intestinal contents from the ileum, cecum and colon of five MF-TG and five wild-type (WT) sows and investigated the gut microbiome of the transgenic pigs via metagenomic analysis. Results showed that the levels of probiotics, such as Lactobacillus reuteri and Streptococcus, were more abundant in the cecum of the MF-TG pigs and higher than those of WT pigs. By contrast, the levels of harmful microorganisms, such as Campylobacter, Chlamydia trachomatis, and Campylobacter fetus, and various unidentified viruses, were higher in the cecum of the WT pigs than those of the MF-TG pigs. By comparing unigenes and the eggNOG database, we found that the microorganisms in the colon of the MF-TG pigs had high fractional abundance in DNA (cytosine-5)-methyltransferase 1 and serine-type D-Ala-D-Ala carboxypeptidase, whereas the aspartate carbamoyltransferase regulatory subunit and outer membrane protein pathways were enriched in the WT pigs. Moreover, the microorganisms in the cecum of the MF-TG pigs were active in GlycosylTransferase Family 8 (GT8), Glycoside Hydrolase Family 13 (GH13), and Glycoside Hydrolase Family 32 (GH32). Furthermore, the levels of numerous carbohydrases, such as glucan 1,3-beta-glucosidase, xylan 1,4-beta-xylosidase and exo-1,3-1,4-glucanase, were higher in the cecum of the MF-TG pigs than those of the WT pigs. The results indicated that intestinal microbes can change adaptively to the secretion of transgenic enzymes, thereby forming a benign cooperation with their host. This cooperation could be beneficial for improving feed efficiency.

A Cas9-transcription factor fusion protein enhances homology-directed repair efficiency.

Precise gene insertion or replacement in cells and animals that requires incorporation of a foreign DNA template into the genome target site by homology-directed repair (HDR) remains an inefficient process. One of the limiting factors for the inefficiency of HDR lies in the limited chance for colocalization of the donor template and target in the huge genome space. We here present a strategy to enhance HDR efficiency in animal cells by spatial and temporal colocalization of the donor and Cas9 by coupling the CRISPR system with a transcription factor (TF). We first identified that THAP domain-containing 11 (THAP11) can coordinate with CRISPR/Cas9 to increase HDR stably through screening multiple TFs from different species. We next designed donor structures with different fusion patterns with TF-specific DNA-binding motifs and found that appending two copies of THAP11-specific DNA binding motifs to both ends of the double-stranded donor DNA has an optimal effect to promote HDR. The THAP11-fused CRISPR system achieved more than twofold increase in HDR-mediated knock-in efficiency for enhanced green fluorescent protein (EGFP) tagging of endogenous genes in 293T cells. We also demonstrated up to 6-fold increases of knock-in through the combinational use of the TF-fused CRISPR and valnemulin, a recently discovered small-molecule HDR enhancer. This modified CRISPR system provides a simple but highly efficient platform to facilitate CRISPR-mediated KI manipulations.

Generation of Multi-Transgenic Pigs Using PiggyBac Transposons Co-expressing Pectinase, Xylanase, Cellulase, ß-1.3-1.4-Glucanase and Phytase.

The current challenges facing the pork industry are to maximize feed efficiency and minimize fecal emissions. Unlike ruminants, pigs lack several digestive enzymes such as pectinase, xylanase, cellulase, ß-1.3-1.4-glucanase, and phytase which are essential to hydrolyze the cell walls of grains to release endocellular nutrients into their digestive tracts. Herein, we synthesized multiple cellulase and pectinase genes derived from lower organisms and then codon-optimized these genes to be expressed in pigs. These genes were then cloned into our previously optimized XynB (xylanase)- EsAPPA (phytase) bicistronic construct. We then successfully generated transgenic pigs that expressed the four enzymes [Pg7fn (pectinase), XynB (xylanase), EsAPPA (phytase), and TeEGI (cellulase and ß-glucanase)] using somatic cell cloning. The expression of these genes was parotid gland specific. Enzymatic assays using the saliva of these founders demonstrated high levels of phytase (2.0∼3.4 U/mL) and xylanase (0.25∼0.42 U/mL) activities, but low levels of pectinase (0.06∼0.08 U/mL) activity. These multi-transgenic pigs are expected to contribute to enhance feed utilization and reduce environmental impact.

Annealing treatment of amylose and amylopectin extracted from rice starch.

Annealing behavior of amylose and amylopectin was unclear. In this work, high purity amylose and amylopectin were extracted from rice starch, and structural properties of the retrograded rice starch, amylose, and amylopectin before and after annealing treatment were explored. It was found that the purity of the amylose and amylopectin was 95.64% ±â€¯2.69% and 94.98% ±â€¯0.97%, respectively. Their molecular weight was (2.93 ±â€¯0.21) × 106 Da and (5.90 ±â€¯0.13) × 107 Da, respectively. Besides, the relative crystallinities and ratios of 1047 cm-1/1022 cm-1 of the retrograded rice starch and amylose were significantly increased by annealing treatment, while that of retrograded amylopectin did not change. These results clarified that amylose was more sensitive to annealing treatment than amylopectin, and amylose was more responsible for annealing of starch than amylopectin. The findings contributed to a better understanding of the annealing behavior of starch.

[Knockdown the expression of ku70 and lig4 in HEK293T cells by CRISPR/Cas13 system].

Clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated proteins (Cas) system is a hotspot of gene editing and gene expression research, in which CRISPR/Cas13 system provides a new direction for RNA interference and editing. In this study, we designed and synthesized the corresponding gRNAs of CRISPR/Cas13a and CRISPR/Cas13b systems in non-homologous end joining (NHEJ) pathway, such as Ku70 and Lig4, and then detected the expression of ku70 and lig4 in HEK293T cells. The CRISPR/Cas13a system could efficiently knockdown the mRNA expression of ku70 and lig4 more than 50%, and CRISPR/Cas13b system also suppressed ku70 and lig4 about 92% and 76%, respectively. Also, CRISPR/Cas13a, b systems could down-regulate Ku70 and Lig4 proteins level to 68% and 53%, respectively. The study demonstrates that the CRISPR/Cas13 system could effectively knockdown the expression of RNA and protein in HEK293T cells, providing a new strategy for gene function and regulation research.

Increasing CRISPR/Cas9-mediated homology-directed DNA repair by histone deacetylase inhibitors.

Histone deacetylase inhibitors (HDACis) affect DNA repair pathways by modulating multiple cellular machineries, including chromatin state, DNA repair factor modification, and the cell cycle. These machineries can differentially affect DNA repair outcomes. With the aim to investigate the impacts of HDACis on DNA repair following CRISPR/Cas9 cleavage from the mixed actions, we used two pan-HDACis, trichostatin A (TSA) and PCI-24781, to treat animal immortalized and primary cells, and studied CRISPR/Cas9-mediated genome editing results by nonhomologous end joining (NHEJ) and homology-directed repair (HDR) pathways. We first found that TSA and PCI-24781 increased NHEJ efficiency. However, further analysis of the total NHEJ events demonstrated that alternative end joining (alt-EJ) mainly contributed to the enhanced total NHEJ by HDACis. We then analyzed HDR efficiency with HDACi treatment and found that multiple HDR pathways, including homologous recombination, single strand annealing and single-stranded oligonucleotide (ssODN)-mediated HDR, were all increased with HDACi treatment. TSA also increased CRISPR-induced ssODN-mediated HDR rate in pig parthenogenetic embryos. Analyzing acetylation status of DNA repair factors showed that acetylation levels of classical NHEJ (c-NHEJ) factors KU70 and KU80 and alt-EJ factor PARP1 were significantly enhanced, but alt-EJ factor LIG3 and HDR factors Rad51 and Rad52 were not affected greatly, implying a differential impact on these repair pathways by HDACis. In addition, TSA and PCI-24781 can enrich cells in G2/M phase of the cell cycle which is beneficial for occurrence of HDR. These findings show that HDACis can effectively promote CRISPR-mediated homology-involved DNA repair, including HDR and alt-EJ pathways, through concerted action of multiple cellular machineries.

Bcl6 Preserves the Suppressive Function of Regulatory T Cells During Tumorigenesis.

During tumorigenesis, tumor infiltrating regulatory T (Treg) cells restrict the function of effector T cells in tumor microenvironment and thereby promoting tumor growth. The anti-tumor activity of effector T cells can be therapeutically unleashed, and is now being exploited for the treatment of various types of human cancers. However, the immune suppressive function of Treg cells remains a major hurdle to broader effectiveness of tumor immunotherapy. In this article, we reported that the deletion of Bcl6 specifically in Treg cells led to stunted tumor growth, which was caused by impaired Treg cell responses. Notably, Bcl6 is essential in maintaining the lineage stability of Treg cells in tumor microenvironment. Meanwhile, we found that the absence of follicular regulatory T (Tfr) cells, which is a result of Bcl6 deletion in Foxp3+ cells, was dispensable for tumor control. Importantly, the increased Bcl6 expression in Treg cells is associated with poor prognosis of human colorectal cancer and lymph node metastasis of skin melanoma. Furthermore, Bcl6 deletion in Treg cells exhibits synergistic effects with immune checkpoint blockade therapy. Collectively, these results indicate that Bcl6 actively participates in regulating Treg cell immune responses during tumorigenesis and can be exploited as a therapeutic target of anti-tumor immunity.

One-step preparation of single-layered graphene quantum dots for the detection of Fe3.

Single-layer graphene quantum dots are highly desirable while their facile and controllable preparations remain challenging. Herein, single-layered graphene quantum dots (sl-GQDs) were developed via a facile one-step hydrothermal synthesis, with citric acid and ß-cyclodextrin (CD) as starting materials. The sl-GQDs decorated with CD molecules emit green fluorescence with a quantum yield of 5.34%, and exhibit a good response exclusively to ferric ions for their structural oxygenous groups. The linear range of the proposed sensor for ferric ions was found in a wide concentration range of 0-85 µM. The detection limit is about 0.26 µM. The sl-GQDs based sensing platform also demonstrates its feasibility in real water sample analysis with recoveries of 93.8%-101.5%.

CRISPR/Cas9-Mediated Integration of Large Transgene into Pig CEP112 Locus.

Clustered regularly interspaced short palindromic repeats (CRISPR)-associated protein 9 (Cas9) is a precise genome manipulating tool that can produce targeted gene mutations in various cells and organisms. Although CRISPR/Cas9 can efficiently generate gene knockout, the gene knock-in (KI) efficiency mediated by homology-directed repair remains low, especially for large fragment integration. In this study, we established an efficient method for the CRISPR/Cas9-mediated integration of large transgene cassette, which carries salivary gland-expressed multiple digestion enzymes (≈ 20 kbp) in CEP112 locus in pig fetal fibroblasts (PFFs). Our results showed that using an optimal homology donor with a short and a long arm yielded the best CRISPR/Cas9-mediated KI efficiency in CEP112 locus, and the targeting efficiency in CEP112 locus was higher than in ROSA26 locus. The CEP112 KI cell lines were used as nuclear donors for somatic cell nuclear transfer to create genetically modified pigs. We found that KI pig (705) successfully expressed three microbial enzymes (ß-glucanase, xylanase, and phytase) in salivary gland. This finding suggested that the CEP112 locus supports exogenous gene expression by a tissue-specific promoter. In summary, we successfully targeted CEP112 locus in pigs by using our optimal homology arm system and established a modified pig model for foreign digestion enzyme expression in the saliva.