Gapr for large-scale collaborative single-neuron reconstruction.

Whole-brain analysis of single-neuron morphology is crucial for unraveling the complex structure of the brain. However, large-scale neuron reconstruction from terabyte and even petabyte data of mammalian brains generated by state-of-the-art light microscopy is a daunting task. Here, we developed 'Gapr' (Gapr accelerates projectome reconstruction) that streamlines deep learning-based automatic reconstruction, 'automatic proofreading' that reduces human workloads at high-confidence sites, and high-throughput collaborative proofreading by crowd users through the Internet. Furthermore, Gapr offers a seamless user interface that ensures high proofreading speed per annotator, on-demand conversion for handling large datasets, flexible workflows tailored to diverse datasets and rigorous error tracking for quality control. Finally, we demonstrated Gapr's efficacy by reconstructing over 4,000 neurons in mouse brains, revealing the morphological diversity in cortical interneurons and hypothalamic neurons. Here, we present Gapr as a solution for large-scale single-neuron reconstruction projects.

MultiSC: a deep learning pipeline for analyzing multiomics single-cell data.

Single-cell technologies enable researchers to investigate cell functions at an individual cell level and study cellular processes with higher resolution. Several multi-omics single-cell sequencing techniques have been developed to explore various aspects of cellular behavior. Using NEAT-seq as an example, this method simultaneously obtains three kinds of omics data for each cell: gene expression, chromatin accessibility, and protein expression of transcription factors (TFs). Consequently, NEAT-seq offers a more comprehensive understanding of cellular activities in multiple modalities. However, there is a lack of tools available for effectively integrating the three types of omics data. To address this gap, we propose a novel pipeline called MultiSC for the analysis of MULTIomic Single-Cell data. Our pipeline leverages a multimodal constraint autoencoder (single-cell hierarchical constraint autoencoder) to integrate the multi-omics data during the clustering process and a matrix factorization-based model (scMF) to predict target genes regulated by a TF. Moreover, we utilize multivariate linear regression models to predict gene regulatory networks from the multi-omics data. Additional functionalities, including differential expression, mediation analysis, and causal inference, are also incorporated into the MultiSC pipeline. Extensive experiments were conducted to evaluate the performance of MultiSC. The results demonstrate that our pipeline enables researchers to gain a comprehensive view of cell activities and gene regulatory networks by fully leveraging the potential of multiomics single-cell data. By employing MultiSC, researchers can effectively integrate and analyze diverse omics data types, enhancing their understanding of cellular processes.

A model-based constrained deep learning clustering approach for spatially resolved single-cell data.

Spatially resolved scRNA-seq (sp-scRNA-seq) technologies provide the potential to comprehensively profile gene expression patterns in tissue context. However, the development of computational methods lags behind the advances in these technologies, which limits the fulfillment of their potential. In this study, we develop a deep learning approach for clustering sp-scRNA-seq data, named Deep Spatially constrained Single-cell Clustering (DSSC). In this model, we integrate the spatial information of cells into the clustering process in two steps: (1) the spatial information is encoded by using a graphical neural network model, and (2) cell-to-cell constraints are built based on the spatial expression pattern of the marker genes and added in the model to guide the clustering process. Then, a deep embedding clustering is performed on the bottleneck layer of autoencoder by Kullback-Leibler (KL) divergence along with the learning of feature representation. DSSC is the first model that can use information from both spatial coordinates and marker genes to guide cell/spot clustering. Extensive experiments on both simulated and real data sets show that DSSC boosts clustering performance significantly compared with the state-of-the-art methods. It has robust performance across different data sets with various cell type/tissue organization and/or cell type/tissue spatial dependency. We conclude that DSSC is a promising tool for clustering sp-scRNA-seq data.

Development and evaluation of a human CD47/HER2 bispecific antibody for Trastuzumab-resistant breast cancer immunotherapy.

The treatment for trastuzumab-resistant breast cancer (BC) remains a challenge in clinical settings. It was known that CD47 is preferentially upregulated in HER2+ BC cells, which is correlated with drug resistance to trastuzumab. Here, we developed a novel anti-CD47/HER2 bispecific antibody (BsAb) against trastuzumab-resistant BC, named IMM2902. IMM2902 demonstrated high binding affinity, blocking activity, antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and internalization degradation effects against both trastuzumab-sensitive and trastuzumab-resistant BC cells in vitro. The in vivo experimental data indicated that IMM2902 was more effective than their respective controls in inhibiting tumor growth in a trastuzumab-sensitive BT474 mouse model, a trastuzumab-resistant HCC1954 mouse model, two trastuzumab-resistant patient-derived xenograft (PDX) mouse models and a cord blood (CB)-humanized HCC1954 mouse model. Through spatial transcriptome assays, multiplex immunofluorescence (mIFC) and in vitro assays, our findings provided evidence that IMM2902 effectively stimulates macrophages to generate C-X-C motif chemokine ligand (CXCL) 9 and CXCL10, thereby facilitating the recruitment of T cells and NK cells to the tumor site. Moreover, IMM2902 demonstrated a high safety profile regarding anemia and non-specific cytokines release. Collectively, our results highlighted a novel therapeutic approach for the treatment of HER2+ BCs and this approach exhibits significant anti-tumor efficacy without causing off-target toxicity in trastuzumab-resistant BC cells.

Highly-Efficient and Robust Zn Anodes Enabled by Sub-1-µm Zincophilic CrN Coatings.

For exploring advanced Zn-ion batteries (ZIBs) with long lifespan and high Coulombic efficiency (CE), the critically important point is to limit the undesired Zn dendrite and parasitic reactions. Among the coating for electrode is a promising strategy, relying on the trade-off between its thickness and stability to achieve the ultra-stable Zn anodes in ZIBs. Herein, a submicron-thick (≈0.4 µm) zincophilic CrN coatings are fabricated by a facile and industry-compatible magnetron sputtering approach. It is exhilarating that the ultrathin and dense CrN coatings with strong adsorption ability for Zn2+ exhibit an impressive lifespan up to 3700 h with ≈100% CE at 1 mA cm-2. Along with the experiments and theoretical calculations, it is verified that the introduced CrN coatings cannot only effectively suppress the dendrite growth and notorious parasitic reactions, but also allow the uniform Zn deposition due to the reduced nucleation energy. Moreover, the as-assembled Zn@CrN‖MnO2 full cell delivers a high specific capacity of 171.1 mAh g-1 after 1000 cycles at 1 A g-1, much better than that of Zn‖MnO2 analog (97.8 mAh g-1). This work provides a facile strategy for scalable fabrication of ultrathin zincophilic coating to push forward the practical applications of ZIBs.

Autism spectrum disorders detection based on multi-task transformer neural network.

Autism Spectrum Disorders (ASD) are neurodevelopmental disorders that cause people difficulties in social interaction and communication. Identifying ASD patients based on resting-state functional magnetic resonance imaging (rs-fMRI) data is a promising diagnostic tool, but challenging due to the complex and unclear etiology of autism. And it is difficult to effectively identify ASD patients with a single data source (single task). Therefore, to address this challenge, we propose a novel multi-task learning framework for ASD identification based on rs-fMRI data, which can leverage useful information from multiple related tasks to improve the generalization performance of the model. Meanwhile, we adopt an attention mechanism to extract ASD-related features from each rs-fMRI dataset, which can enhance the feature representation and interpretability of the model. The results show that our method outperforms state-of-the-art methods in terms of accuracy, sensitivity and specificity. This work provides a new perspective and solution for ASD identification based on rs-fMRI data using multi-task learning. It also demonstrates the potential and value of machine learning for advancing neuroscience research and clinical practice.

Glycosylation of cellulase: a novel strategy for improving cellulase.

Protein glycosylation is the most complex posttranslational modification process. Most cellulases from filamentous fungi contain N-glycosylation and O-glycosylation. Here, we discuss the potential roles of glycosylation on the characteristics and function of cellulases. The use of certain cultivation, inducer, and alteration of engineering glycosylation pathway can enable the rational control of cellulase glycosylation. Glycosylation does not occur arbitrarily and may tend to modify the 3D structure of cellulases by using specially distributed glycans. Therefore, glycoengineering should be considered comprehensively along with the spatial structure of cellulases. Cellulase glycosylation may be an evolution phenomenon, which has been considered as an economical way for providing different functions from identical proteins. In addition to gene and transcription regulations, glycosylation may be another regulation on the protein expression level. Enhanced understanding of the potential regulatory role of cellulase glycosylation will enable synthetic biology approaches for the development of commercial cellulase.

Exploring the heterogeneity of brain structure in autism spectrum disorder based on individual structural covariance network.

Autism spectrum disorder (ASD) is characterized by highly structural heterogeneity. However, most previous studies analyzed between-group differences through a structural covariance network constructed based on the ASD group level, ignoring the effect of between-individual differences. We constructed the gray matter volume-based individual differential structural covariance network (IDSCN) using T1-weighted images of 207 children (ASD/healthy controls: 105/102). We analyzed structural heterogeneity of ASD and differences among ASD subtypes obtained by a K-means clustering analysis based on evidently different covariance edges relative to healthy controls. The relationship between the distortion coefficients (DCs) calculated at the whole-brain, intra- and interhemispheric levels and the clinical symptoms of ASD subtypes was then examined. Compared with the control group, ASD showed significantly altered structural covariance edges mainly involved in the frontal and subcortical regions. Given the IDSCN of ASD, we obtained 2 subtypes, and the positive DCs of the 2 ASD subtypes were significantly different. Intra- and interhemispheric positive and negative DCs can predict the severity of repetitive stereotyped behaviors in ASD subtypes 1 and 2, respectively. These findings highlight the crucial role of frontal and subcortical regions in the heterogeneity of ASD and the necessity of studying ASD from the perspective of individual differences.

Dysregulated dynamic time-varying triple-network segregation in children with autism spectrum disorder.

One of the remarkable characteristics of autism spectrum disorder (ASD) is the dysregulation of functional connectivity of the triple-network, which includes the salience network (SN), default mode network (DMN), and central executive network (CEN). However, there is little known about the segregation of the triple-network dynamics in ASD. This study used resting-state functional magnetic resonance imaging data including 105 ASD and 102 demographically-matched typical developing control (TC) children. We compared the dynamic time-varying triple-network segregation and triple-network functional connectivity states between ASD and TC groups, and examined the relationship between dynamic triple-network segregation alterations and clinical symptoms of ASD. The average dynamic network segregation value of the DMN with SN and the DMN with CEN in ASD was lower but the coefficient of variation (CV) of dynamic network segregation of the DMN with CEN was higher in ASD. Furthermore, partially reduced triple-network segregation associated with the DMN was found in connectivity states analysis of ASD. These abnormal average values and CV of dynamic network segregation predicted social communication deficits and restricted and repetitive behaviors in ASD. Our findings indicate abnormal dynamic time-varying triple-network segregation of ASD and highlight the crucial role of the triple-network in the neural mechanisms underlying ASD.

Rapid Identification of Chemical Constituents of Sanhua Decoction in Vivo and in Vitro and the Metabolites of Rhubarb Anthraquinone Aglycone by Ultra-High Performance Liquid Chromatography Coupled with Quadrupole Time-of-Flight Tandem Mass Spectrometry.

Sanhua Decoction (SHD) is a classic prescription for the treatment of stroke in the clinic. Based on the combination strategy in vitro and in vivo, the chemical constituents of SHD were characterized by ultra-high-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry, and the metabolites of five effective anthraquinone aglycones (emodin, chrysophanol, rhein, aloe-emodin, and physcion) of rhubarb in SHD were studied. A total of 222 compounds were isolated and identified in vitro, including 50 flavonoids and their glycosides, 41 coumarins, 22 anthraquinones, 22 tannins, 14 phenylpropanoids, 16 alkaloids, 18 organic acids and their esters, 14 lignans, six anthrones, and 19 other compounds. A total of 111 prototype components were isolated and identified in vivo. Among them, 26, 82, 101, and 46 prototype components and 87 metabolites were detected in plasma, urine, feces, and bile for the first time. This study provides a basis for the identification of chemical components in vivo and in vitro and the analysis of potential pharmacodynamic components of SHD, and provides a basis for further study of pharmacodynamic mechanism.

Self-assembled nanoparticles based on DNA origami and a nitrated T helper cell epitope as a platform for the development of personalized cancer vaccines.

Neoantigen vaccines constitute an emerging and promising cancer immunotherapy. However, not all neoantigens have anti-tumor activity, as poor CD4+ epitope recognition can lead to the lack of greatly limit the persistence of the CD8+ T cell response. Therefore, we designed a self-assembled nanoplatform hereinafter referred to as DNA-coupled nitrated T helper cell epitope nanoparticle (DCNP) based on DNA origami containing a nitrated CD4 + T cell epitope, which can facilitate the effective activation of neoantigen-specific CD8+ T cells. Moreover, we embedded the cytidine-phosphate-guanosine oligonucleotide (CpG ODN) motif sequence in the DNA skeleton to function as a built-in adjuvant to activate Toll-like receptor 9. DCNP can markedly improve adjuvant and neoantigen co-delivery to lymphoid organs and promote neoantigen presentation on dendritic cells. Moreover, DCNP induced robust, and long-lived neoantigen-specific CD8+ T cell responses that significantly delayed tumor growth. Further, these effects were largely dependent on the nitrated T cell epitope. Collectively, our findings indicate that DCNP is a promising platform that could improve the development of personalized therapeutic neoantigen vaccines for cancer immunotherapy.

Association between the pharmacological treatment of bipolar disorder and risk of traumatic injuries: a self-controlled case series study.

BACKGROUND: Patients with bipolar disorder (BPD) are prone to engage in risk-taking behaviours and self-harm, contributing to higher risk of traumatic injuries requiring medical attention at the emergency room (ER).We hypothesize that pharmacological treatment of BPD could reduce the risk of traumatic injuries by alleviating symptoms but evidence remains unclear. This study aimed to examine the association between pharmacological treatment and the risk of ER admissions due to traumatic injuries. METHODS: Individuals with BPD who received mood stabilizers and/or antipsychotics were identified using a population-based electronic healthcare records database in Hong Kong (2001-2019). A self-controlled case series design was applied to control for time-invariant confounders. RESULTS: A total of 5040 out of 14 021 adults with BPD who received pharmacological treatment and had incident ER admissions due to traumatic injuries from 2001 to 2019 were included. An increased risk of traumatic injuries was found 30 days before treatment [incidence rate ratio (IRR) 4.44 (3.71-5.31), p < 0.0001]. After treatment initiation, the risk remained increased with a smaller magnitude, before returning to baseline [IRR 0.97 (0.88-1.06), p = 0.50] during maintenance treatment. The direct comparison of the risk during treatment to that before and after treatment showed a significant decrease. After treatment cessation, the risk was increased [IRR 1.34 (1.09-1.66), p = 0.006]. CONCLUSIONS: This study supports the hypothesis that pharmacological treatment of BPD was associated with a lower risk of ER admissions due to traumatic injuries but an increased risk after treatment cessation. Close monitoring of symptoms relapse is recommended to clinicians and patients if treatment cessation is warranted.

Attenuation of IFN signaling due to m6A modification of the host epitranscriptome promotes EBV lytic reactivation.

BACKGROUND: Reactivation of Epstein Barr virus (EBV) leads to modulation of the viral and cellular epitranscriptome. N6-methyladenosine (m6A) modification is a type of RNA modification that regulates metabolism of mRNAs. Previous reports demonstrated that m6A modification affects the stability and metabolism of EBV encoded mRNAs. However, the effect of reactivation on reprograming of the cellular mRNAs, and how this contributes to successful induction of lytic reactivation is not known. METHODS: Methylated RNA immunoprecipitation sequencing (MeRIP-seq), transcriptomic RNA sequencing (RNA-seq) and RNA pull-down PCR were used to screen and validate differentially methylated targets. Western blotting, quantitative real-time PCR (RT-qPCR) and immunocytochemistry were used to investigate the expression and localization of different proteins. RNA stability and polysome analysis assays were used to detect the half-lives and translation efficiencies of downstream genes. Insertion of point mutation to disrupt the m6A methylation sites was used to verify the effect of m6A methylation on its stability and expression levels. RESULTS: We report that during EBV reactivation the m6A eraser ALKBH5 is significantly downregulated leading to enhanced methylation of the cellular transcripts DTX4 and TYK2, that results in degradation of TYK2 mRNAs and higher efficiency of translation of DTX4 mRNAs. This resulted in attenuation of IFN signaling that promoted progression of viral lytic replication. Furthermore, inhibition of m6A methylation of these transcripts led to increased production of IFN, and a substantial reduction in viral copy number, which suggests abrogation of lytic viral replication. CONCLUSION: Our findings illuminate the significance of m6A modification in overcoming the innate immune response during EBV reactivation. We now report that during lytic reactivation EBV targets the RNA methylation system of the host to attenuate the innate immune response by suppressing the interferon signaling which facilitates successful lytic replication of the virus.

Treatment with methylphenidate and the risk of fractures among children and young people: A systematic review and self-controlled case series study.

AIMS: Animal studies suggest that methylphenidate treatment for around 3 months may lead to less mineralized and weaker appendicular bones. A systematic review was conducted to summarize the evidence from observational studies, and a self-controlled case series study was used to compare the risk before and after treatment initiation. METHODS: Literature search was conducted using PubMed, Embase and the Cochrane Library to identify observational studies on methylphenidate and fractures. We also conducted a self-controlled case series study with individuals aged 5-24 years who received methylphenidate treatment and experienced fractures from 2001 to 2020 in Hong Kong. Incidence rate ratios and 95% confidence intervals were calculated by comparing the incidence rate in the methylphenidate-exposed period compared with nonexposed period. RESULTS: Six cohort studies and 2 case-control studies were included in the systematic review. For all-cause fractures, studies found a 39-74% lower risk in treated-attention deficit hyperactivity disorder (ADHD) group compared with untreated ADHD but no difference between stimulants and nonstimulants. Differences between sexes and treatment duration were also found-significant results were shown in males and those with longer treatment duration. Among 43 841 individuals with ADHD medication before the year 2020, 2023 were included in the self-controlled case series analysis. The risks of fractures were lower by 32-41% in different treatment periods when compared with 6 months before treatment initiation. CONCLUSION: Methylphenidate treatment may lower the risk of all-cause fractures from both study designs; however, further evidence is needed about the treatment duration and sex effect. Conclusions on stress fractures are not yet established, and further research is required.

Maternal Benzodiazepines and Z-Drugs Use during Pregnancy and Adverse Birth and Neurodevelopmental Outcomes in Offspring: A Population-Based Cohort Study.

INTRODUCTION: The use of benzodiazepines and/or z-drugs in women of childbearing age has increased. OBJECTIVE: The aim of the study was to evaluate whether gestational benzodiazepine and/or z-drug exposure is associated with adverse birth and neurodevelopmental outcomes. METHODS: A population-based cohort including mother-child pairs from 2001 to 2018 in Hong Kong was analysed to compare gestationally exposed and nonexposed children on the risk of preterm birth, small for gestational age, autism spectrum disorder (ASD), and attention-deficit/hyperactivity disorder (ADHD) through logistic/Cox proportional hazards regression with a 95% confidence interval (CI). Sibling-matched analyses and negative control analyses were applied. RESULTS: When comparing gestationally exposed with gestationally nonexposed children, the weighted odds ratio (wOR) was 1.10 (95% CI = 0.97-1.25) for preterm birth and 1.03 (95% CI = 0.76-1.39) for small for gestational age, while the weighted hazard ratio (wHR) was 1.40 (95% CI = 1.13-1.73) for ASD and 1.15 (95% CI = 0.94-1.40) for ADHD. Sibling-matched analyses showed no association between gestationally exposed children and their gestationally nonexposed siblings for all outcomes (preterm birth: wOR = 0.84, 95% CI = 0.66-1.06; small for gestational age: wOR = 1.02, 95% CI = 0.50-2.09; ASD: wHR = 1.10, 95% CI = 0.70-1.72; ADHD: wHR = 1.04, 95% CI = 0.57-1.90). Similarly, no significant differences were observed when comparing children whose mothers took benzodiazepines and/or z-drugs during pregnancy to children whose mothers took benzodiazepines and/or z-drugs before but not during pregnancy for all outcomes. CONCLUSIONS: The findings do not support a causal relationship between gestational benzodiazepines and/or z-drugs exposure and preterm birth, small for gestational age, ASD, or ADHD. Clinicians and pregnant women should carefully balance the known risks of benzodiazepines and/or z-drugs use against those of untreated anxiety and sleep problems.

Economical production of Pichia pastoris single cell protein from methanol at industrial pilot scale.

BACKGROUND: Methanol, synthesized from CO2, is a potentially sustainable one-carbon (C1) resource for biomanufacturing. The use of methanol as a feedstock to produce single cell protein (SCP) has been investigated for decades as an alternative to alleviate the high global demand for animal-derived proteins. The methylotrophic yeast Pichia pastoris is an ideal host for methanol-based SCP synthesis due to its natural methanol assimilation ability. However, improving methanol utilization, tolerance to higher temperature, and the protein content of P. pastoris are also current challenges, which are of great significance to the economical industrial application using methanol as a feedstock for SCP production. RESULTS: In the present work, adaptive laboratory evolution (ALE) has been employed to overcome the low methanol utilization efficiency and intolerance to a higher temperature of 33 °C in P. pastoris, associated with reduced carbon loss due to the lessened detoxification of intracellular formaldehyde through the dissimilation pathway and cell wall rearrangement to temperature stress resistance following long-term evolution as revealed by transcriptomic and phenotypic analysis. By strengthening nitrogen metabolism and impairing cell wall synthesis, metabolic engineering further increased protein content. Finally, the engineered strain via multi-strategy produced high levels of SCP from methanol in a pilot-scale fed-batch culture at 33 °C with a biomass of 63.37 g DCW/L, methanol conversion rate of 0.43 g DCW/g, and protein content of 0.506 g/g DCW. SCP obtained from P. pastoris contains a higher percentage of protein compared to conventional foods like soy, fish, meat, whole milk, and is a source of essential amino acids, including methionine, lysine, and branched-chain amino acids (BCAAs: valine, isoleucine, leucine). CONCLUSIONS: This study clarified the unique mechanism of P. pastoris for efficient methanol utilization, higher temperature resistance, and high protein synthesis, providing a P. pastoris cell factory for SCP production with environmental, economic, and nutritional benefits.

Analysis of histopathology and changes of major cytokines in the lesions caused by Mycoplasma ovipneumoniae infection.

BACKGROUND: Mycoplasma ovipneumoniae (M. ovipneumoniae) is one of the main pathogens of sheep pneumonia, causing a series of clinical symptoms, such as depression, anorexia, hyperthermia, cough, dyspnea, and tract secretions. In recent years, the prevalence of M. ovipneumoniae pneumonia has become increasingly serious in sheep farms in Ningxia, China, leading to the death of sheep, and causing significant economic losses. In this study, the pathological organs infected by M. ovipneumoniae were collected to observe histopathological change, to determine the tissue localization of M. ovipneumoniae, and to analyze the cytokine changes, which lays a basis for the diagnosis and pathogenesis of M. ovipneumoniae disease. RESULTS: In this study, M. ovipneumoniae was detected in 97 of 105 samples collected from 13 large-scale sheep farms for nucleic acid by PCR. One representative isolate per farm was isolated from 13 farms. The lesions caused by M. ovipneumoniae were mainly in the trachea, bronchus, and lung, including necrosis of tracheal mucosal epithelial cells, disintegration of some epithelial cells, edema of mucosal lamina propria, with inflammatory cell infiltration, cytoplasmic vacuolization of epithelial cells of bronchial mucosa, massive infiltration of inflammatory cells in the alveolar space of lung, necrosis and hyperplasia of alveolar epithelial cells. Immunohistochemical analysis showed that the proportion of M. ovipneumoniae positive area in the lung was the largest, followed by that in the bronchus and trachea. Compared to healthy animals, diseased animals exhibited up-regulated gene expression levels of IL-1ß, IL-6, and NF-κB in the trachea, bronchus, and lungs. In contrast, the expression of IL-10, IL-12, and IFN-γ was primarily limited to the trachea and bronchus. The expression of IL-1ß showed differential patterns across different lung regions, with variations observed among lung lobes. Additionally, other cytokines consistently showed significant up-regulation specifically in the bronchus. CONCLUSIONS: M. ovipneumoniae is primarily found in the lungs of infected individuals. NF-κB, an essential transcription factor, is involved in the regulation of IL-1ß transcription. IL-12 may enhance the cytotoxic function of natural killer cells during M. ovipneumoniae infection. Those findings demonstrate the distinct expression profiles of cytokines in various anatomical sites throughout disease progression, suggesting the potential role of bronchial tissue as a major site of immune response.

Machine learning based on the graph convolutional self-organizing map method increases the accuracy of pollution source identification: A case study of trace metal(loid)s in soils of Jiangmen City, south China.

Rapid economic development and industrialization may include environmentally harmful human activities that cause heavy-metal accumulation in soils, ultimately threatening the quality of the soil environment and human health. Therefore, accurate identification of pollution sources is an important weapon in efforts to control and prevent pollution. The self-organizing map (SOM) method is widely used in pollution source identification because of its capacity for visualization of high-dimensional data. The SOM ignores the graph structure relationship among chemical elements in soils; the SOM analysis of pollution sources has high uncertainty. Here, we propose a new analysis method, i.e., the graph convolutional self-organizing map (GCSOM), which uses a graph convolutional network (GCN) to extract the graph structure relationship among the chemical elements in soils, then performs data visualization using an SOM. We compared the performances of GCSOM and SOM, then assessed the pollution source characteristics of trace metal(loid)s (TMs, mostly heavy metals) in Jiangmen City using the GCSOM. Our experimental results showed that the GCSOM is superior to the SOM for identification of TM sources, while the TMs in the soil of Jiangmen originate from three main sources: agricultural activities (mainly in Taishan City, Jiangmen), traffic emissions (mainly in Xinhui and Pengjiang Districts), and industrial activities (mainly in Xinhui District). The risk assessment indicated that the risk of all TMs was within threshold.

Enhancing the Heterologous Expression of a Thermophilic Endoglucanase and Its Cost-Effective Production in Pichia pastoris Using Multiple Strategies.

Although Pichia pastoris was successfully used for heterologous gene expression for more than twenty years, many factors influencing protein expression remain unclear. Here, we optimized the expression of a thermophilic endoglucanase from Thermothielavioides terrestris (TtCel45A) for cost-effective production in Pichia pastoris. To achieve this, we established a multifactorial regulation strategy that involved selecting a genome-editing system, utilizing neutral loci, incorporating multiple copies of the heterologous expression cassette, and optimizing high-density fermentation for the co-production of single-cell protein (SCP). Notably, even though all neutral sites were used, there was still a slight difference in the enzymatic activity of heterologously expressed TtCel45A. Interestingly, the optimal gene copy number for the chromosomal expression of TtCel45A was found to be three, indicating limitations in translational capacity, post-translational processing, and secretion, ultimately impacting protein yields in P. pastoris. We suggest that multiple parameters might influence a kinetic competition between protein elongation and mRNA degradation. During high-density fermentation, the highest protein concentration and endoglucanase activity of TtCel45A with three copies reached 15.8 g/L and 9640 IU/mL, respectively. At the same time, the remaining SCP of P. pastoris exhibited a crude protein and amino acid content of up to 59.32% and 46.98%, respectively. These findings suggested that SCP from P. pastoris holds great promise as a sustainable and cost-effective alternative for meeting the global protein demand, while also enabling the production of thermophilic TtCel45A in a single industrial process.

PLDP-FL: Federated Learning with Personalized Local Differential Privacy.

As a popular machine learning method, federated learning (FL) can effectively solve the issues of data silos and data privacy. However, traditional federated learning schemes cannot provide sufficient privacy protection. Furthermore, most secure federated learning schemes based on local differential privacy (LDP) ignore an important issue: they do not consider each client's differentiated privacy requirements. This paper introduces a perturbation algorithm (PDPM) that satisfies personalized local differential privacy (PLDP), resolving the issue of inadequate or excessive privacy protection for some participants due to the same privacy budget set for all clients. The algorithm enables clients to adjust the privacy parameters according to the sensitivity of their data, thus allowing the scheme to provide personalized privacy protection. To ensure the privacy of the scheme, we have conducted a strict privacy proof and simulated the scheme on both synthetic and real data sets. Experiments have demonstrated that our scheme is successful in producing high-quality models and fulfilling the demands of personalized privacy protection.