Biology subjects

Hu, Y.

Publications and source records attributed to Hu, Y..

At least 217 records · Page 12Linked to original sources

bioRxiv2024

Genome-wide investigation of outer membrane protein families under mosaic evolution in Escherichia coli

Several genes in Gram-negative bacteria encoding outer membrane proteins (OMPs) have been reported to show patterns of mosaic evolution featured by mixture of negative selection and local recombination. Here, we proposed a strategy, and applied it to screen OMPs under mosaic evolution in Escherichia coli in the genome level. In total, 21 OMP families, including 16 new ones, were detected with the typical patterns of mosaic evolution. An abosolute majority of the protein families are conserved in E. coli for the composition, genomic loci and the overall structures. Highly Variable Regions (HVRs) can be recognized, which are frequently located extracellularly within the protruding loops. There are only limited number of major HVR sequence types, within which positively selected sites can be detected ocassionally. Based on the simulated results of multiple models, the OMPs under mosaic evolution are often with good antigenicity, with HVRs of various sequence types coinciding with the B-cell epitopes of the strongest immunogenicity. The study futher broadened our understanding of the characteristics of mosaic evolution and the functions of OMPs in Gram-negative bacteria, laying an important foundation for their potential translational applications. ImportanceIt is important to understand the evolutionary mechanisms of bacterial OMP-encoding genes, which would facilitate the development of anti-bacterial reagents. This study made the first genome-wide screening of bacterial OMPs under mosaic evolution, and increased the list of candidate OMP families by 3 folds in E. coli, far more than we expected. The study further confirmed the hypothesis about the evolutionary, micro-evolutionary and structural features of these OMPs, and facilitated the functional theory of mosaic evolution. Moreover, the findings of limited HVR sequence types and strong immunogenicity of HVRs paved important foundation for application of these OMPs and their HVRs in development of antibodies or other anti-bacterial treatment.

microbiology↗

bioRxiv2024

NAD+ - and EVA1-C-dependent reversal of neurological deficits is mediated by differential alternative RNA splicing in tauopathic animal models

Aberrant alternative splicing (ASEs) is an aging hallmark to Alzheimers Disease (AD). Although NAD+ and related metabolites can slow down AD, NAD+ on ASEs in AD remain unclear. Mouse transcriptomic data revealed NR-induced ASEs, focusing on the Eva1-C locus. AI-based algorithms predicted EVA1-C protein structures and protein-protein interactions. AD postmortem brain samples and tauopathy models including transgenic mice and worm was used for validation. NAD+ abundance/metabolic status modulates ASEs and the expression of EVA1-C isoforms, which in turn regulate the interaction with BAG-1 and HSP70 proteins. Importantly, EVA1-C is dramatically reduced in 20 Braak 5/6 AD patients compared to cognitive normal humans in different brain regions. NAD+ metabolism modulates abundance of specific mRNA isoforms, and that ASEs influence disease progression in model tauopathies and potentially AD. These results could facilitate future development of NAD+-based splice-switching therapeutics for AD. TeaserUnveiling the Link Between NAD+ Metabolism and Alzheimers Disease: Discovering the Role of Alternative RNA Splicing in Disease Progression and Potential Therapeutic Targets

neuroscience↗

bioRxiv2024

Chemosensory modulation of eye-body coordination in larval zebrafish

Coordinated eye-body movements are essential for adaptive behavior, yet little is known about how multisensory input, particularly chemosensory cues, shapes this coordination. Using our enhanced Fish-on-Chips optofluidic platform, we uncovered complex dynamics in how larval zebrafish coordinate saccadic eye movements with tail flips. Under baseline conditions, spontaneous tail flips dynamically align with saccades in frequency and direction for coordinated turns. Chemosensory valence further modulates this coordination: death-associated cues intensify both the strength and relative frequency of coupled saccade-tail flips during turns, whereas food-related cues promote forward gliding without altering saccade coupling. Concurrent brain-wide neuronal imaging reveals that the zebrafish pallium represents the mapping between aversive valence and enhanced saccade-tail coordination, with stronger coupling associated with higher pallium activation. These findings uncover the neural representations and behavioral strategies by which chemosensory inputs of different valences are associated with amplified eye-body coordination and adaptive locomotion in a developing vertebrate, offering insights into principles of sensory-motor integration.

neuroscience↗

bioRxiv2024

Biotic interactions shape infection outcomes in Arabidopsis

The plant microbiome protects plants from stresses, including pathogen attacks. However, identifying microbes that provide plant protection remains challenging in complex microbial communities. In this study, we analysed samples from natural A. thaliana populations, including both plants infected with the pathogenic oomycete Albugo laibachii and uninfected plants, over six years. Using machine learning classification models, we achieved high accuracy in distinguishing infected and uninfected plants based on microbiome abundance. We identified 80 key taxa associated with health and disease. Among the health-associated microbes (HCom), we selected bacteria, fungi, and cercozoa that effectively reduced pathogen presence in co-inoculation assays. In comparison, disease-associated microbes (DCom) were less effective in conferring protection. Our findings highlight the complexity of plant-microbe interactions and advance our understanding of microbial roles in plant disease ecology. By integrating ecological insights with machine learning, we take a significant step towards designing robust microbial consortia that enhance plant resilience against pathogens.

microbiology↗

bioRxiv2024

scAGCI: an anchor graph-based method for cell clustering from integrated scRNA-seq and scATAC-seq data

Cell clustering plays a crucial role in the analysis of single-cell multi-omics research. Despite many methods for multi-omics integrated clustering, challenges such as noise, data sparsity, and biointerpretability analysis hider effective clustering. Recent studies have demonstrated that anchor graph learning clustering in the multiview graph domain can help alleviate sparsity and high noise, reducing runtime costs. However, addressing the heterogeneity and high noise levels in multi-omics data is crucial for obtaining a more representative anchor graph. Furthermore, the existing methods often directly obtain surface information from multi-omics for clustering purposes, neglecting the mining and utilization of higher-order correlations among different features within shared information. In response to these challenges, we propose scAGCI, a cell clustering method based on anchor graphs that integrates both scRNA-seq and scATAC-seq data. Our method captures specific and shared anchor graphs representing the properties of omics data in the process of dynamic anchor unification, and mines high-order shared information to complete the omics representation. Subsequently, clustering results are obtained by integrating the specific and shared omics representation. Extensive experiments show that our method not only outperforms 13 state-of-the-art methods in terms of clustering metrics and running time, but also that the completed omics retains the biological meaning of the original omics.

bioinformatics↗

bioRxiv2024

DeepVul: A Multi-Task Transformer Model for Joint Prediction of Gene Essentiality and Drug Response

Despite their potential, current precision oncology approaches benefit only a small fraction of patients due to their limited focus on actionable genomic alterations. To expand its applicability, we propose DeepVul, a multi-task transformer-based model designed to predict gene essentiality and drug response from cancer transcriptome data. DeepVul aligns gene expressions, gene perturbations, and drug perturbations into a latent space, enabling simultaneous and accurate prediction of cancer cell vulnerabilities to numerous genes and drugs. Benchmarking against existing precision oncology approaches revealed that Deep-Vul not only matches but also complements oncogene-defined precision methods. Through interpretability analyses, DeepVul identifies underlying mechanisms of treatment response and resistance, as demonstrated with BRAF vulnerability prediction. By leveraging whole-genome transcriptome data, DeepVul enhances the clinical actionability of precision oncology, aiding in the identification of optimal treatments across a broader range of cancer patients. DeepVul is publicly available at https://github.com/alaaj27/DeepVul.git.

bioinformatics↗

bioRxiv2024

Dissecting RNA Selectivity Mediated by Tandem RNA-Binding Domains

RNA-protein interactions are pivotal to proper gene regulation. Many RNA-binding proteins possess multiple RNA-binding domains; however, how these domains interplay to specify and regulate RNA targets remains poorly understood. Here, we investigate three multi-domain proteins, Musashi-1, Musashi-2, and Unkempt, three factors which share a high degree of RNA specificity. We use a combination of massively parallel in vitro assays with random or naturally derived RNA sequences and find that individual domains within a protein can have differing affinities, specificities, and spacing preferences. Further, we emphasize that while all three proteins have overlapping motif specificities, non-overlapping sequences may allow for target discrimination. We carry out large scale competition assays between these proteins and determine how individual protein specificities and affinities influence competitive binding. Integration of in vivo binding and regulation with in vitro specificities shows that target selection involves a combination of the protein intrinsic specificities described here, but cellular context is critical to drive these proteins to motifs in specific transcript regions. Finally, evolutionarily conserved RNA regions display evidence of binding multiple RBPs in vivo, and these RNA regions recapitulate this trend with the highest affinity in vitro. We highlight the importance of understanding features of complex RNA-protein interactions and how protein-target discrimination can be established.

biochemistry↗

bioRxiv2024

Benchmarking alternative polyadenylation detection in single-cell and spatial transcriptomes

Background3-tag-based sequencing methods have become the predominant approach for single-cell and spatial transcriptomics, with some protocols proven effective in detecting alternative polyadenylation (APA). While numerous computational tools have been developed for APA detection from these sequencing data, the absence of comprehensive benchmarks and the diversity of sequencing protocols and tools make it challenging to select appropriate methods for APA analysis in these contexts. ResultsWe systematically compared seven 3-tag-based sequencing protocols and identified key peak features affecting APA detection performance. We developed a simulation pipeline that generates realistic datasets preserving protocol-specific characteristics. Using simulated and real data, we comprehensively assessed six computational tools for their ability to identify polyA sites, quantify polyA site expression, detect differentially expressed (DE) APA genes, filter sequencing artifacts, and their computational efficiency. We also investigated factors influencing APA detection. Our evaluation revealed that SCAPE and scAPAtrap generally outperformed other tools across various performance metrics and protocols. ConclusionOur systematic evaluation provides guidance for tool selection, experiment design, and future tool development in APA analysis for singlecell and spatial transcriptomics, paving the way for investigating APA in these contexts.

bioinformatics↗

bioRxiv2024

Deciphering RNA regulation with a foundation language model

RNA metabolism is tightly regulated by cis-elements and trans-acting factors. Most information guiding such regulation is encoded in RNA sequences. Considering the similarities in semantic and syntactic features between RNAs and human language, we developed LAMAR, a transformer-based foundation language model for RNA regulation, to decipher general rules underlying RNA processing. The model was pretrained on approximately 15 million sequences from both genome and transcriptome of 225 mammals and 1569 viruses, and further fine-tuned with labeled datasets for various tasks. The resulting fine-tuned models outperformed the state-of-the-art methods in predicting mRNA translation efficiency and mRNA half-life, while achieving comparable accuracy to specifically designed methods in predicting splice sites of pre-mRNAs and internal ribosome entry sites. Our results indicated that a single foundation language model is applicable in the comprehensive analysis of different aspects of RNA regulation, providing new insight into the design and optimization of RNA drugs.

bioinformatics↗

bioRxiv2024

Red sandstone surface microbiota: functional convergence despite taxonomic divergence under dispersal limitation and hetero selection

Understanding the role of microbiota on stone surface is essential for developing effective grottoes conservation strategies. However, ecological feature of microbial communities on stone surface have been rarely investigated systematically. In this study, we explored diversity, assembly, and functional profiles of microbial communities on the red sandstone surface of the Leshan Giant Buddha from microbial ecology perspective. The results show that Proteobacteria, Actinobacteria, Cyanobacteria, and Ascomycota are the dominant phyla. Fundamental metabolic pathways are maintained during the formation of visually distinguishable microbial communities, but gene profiles vary across microbial communities of different colors. Ecological modelling suggests that selective pressure from the harsh stone surface environment fostered the interplay of dispersal limitation and heterogeneous selection during community assembly. The assembly of visually distinct microbial communities is linked to a narrower ecological niche, higher proportion of habitat specialists, and sparser network structure. Microbial-mediated ammonium assimilation and nitrogen mineralization might be the two prominent processes that contribute to stone biodeterioration. This study deepens our understanding of the assembly mechanisms and functional potentials of microbial communities on stone cultural heritage surfaces, provides microbial ecological insights for the conservation of these cultural treasures. ImportanceMinimal systematically research on the ecological interpretation of stone biodeterioration. This study reports dispersal limitation and heterogeneous selection shape the microbial community assembly responsible for the biodeterioration of red sandstone. Furthermore, fundamental metabolic processes of microbial communities, such as ammonium assimilation and nitrogen mineralization, are identified as contributors to stone biodeterioration. This study improves our understanding of microbial community assembly and their functional roles, providing a microbial ecological basis for developing effective strategies for the conservation of stone cultural heritage.

ecology↗

bioRxiv2024

MedGraphNet: Leveraging Multi-Relational Graph Neural Networks and Text Knowledge for Biomedical Predictions

Genetic, molecular, and environmental factors influence diseases through complex interactions with genes, phenotypes, and drugs. Current methods often fail to integrate diverse multi-relational biological data meaningfully, limiting the discovery of novel risk genes and drugs. To address this, we present MedGraphNet, a multi-relational Graph Neural Network (GNN) model designed to infer relationships among drugs, genes, diseases, and phenotypes. MedGraphNet initializes nodes using informative embeddings from existing text knowledge, allowing for robust integration of various data types and improved generalizability. Our results demonstrate that MedGraphNet matches and often outperforms traditional single-relation approaches, particularly in scenarios with isolated or sparsely connected nodes. The model shows generalizability to external datasets, achieving high accuracy in identifying disease-gene associations and drug-phenotype relationships. Notably, MedGraphNet accurately inferred drug side effects without direct training on such data. Using Alzheimers disease as a case study, MedGraphNet successfully identified relevant phenotypes, genes, and drugs, corroborated by existing literature. These findings demonstrate the potential of integrating multi-relational data with text knowledge to enhance biomedical predictions and drug repurposing for diseases.MedGraphNet code is available at https://github.com/vinash85/MedGraphNet

bioinformatics↗

bioRxiv2024

A genome-wide CRISPR screen in Anopheles mosquito cells identifies essential genes and required components of clodronate liposome function

Anopheles mosquitoes are the sole vector of human malaria, the most burdensome vector-borne disease worldwide. Strategies aimed at reducing mosquito populations and limiting their ability to transmit disease show the most promise for disease control. Therefore, gaining an improved understanding of mosquito biology, and specifically that of the immune response, can aid efforts to develop new approaches that limit malaria transmission. Here, we use a genome-wide CRISPR screening approach for the first time in mosquito cells to identify essential genes in Anopheles and identify genes for which knockout confers resistance to clodronate liposomes, which have been widely used in mammals and arthropods to ablate immune cells. In the essential gene screen, we identified a set of 1280 Anopheles genes that are highly enriched for genes involved in fundamental cell processes. For the clodronate liposome screen, we identified several candidate resistance factors and confirm their roles in the uptake and processing of clodronate liposomes through in vivo validation in Anopheles gambiae, providing new mechanistic detail of phagolysosome formation and clodronate liposome function. In summary, we demonstrate the application of a genome-wide CRISPR knockout platform in a major malaria vector and the identification of genes that are important for fitness and immune-related processes.

genetics↗

bioRxiv2024

Mesenchymal-epithelial transition reduces proliferation but increases immune evasion in tumor spheroids

Mesenchymal-epithelial transition (MET) has been associated with secondary tumor outgrowth during metastasis but the underlying mechanism remains elusive. Using MET-inducible mesenchymal breast cancer cells, we investigated whether MET benefits tumor outgrowth by enhancing proliferation. We found that crowding inhibition of proliferation is present before and after MET, but mesenchymal cells gain a proliferative advantage through more effective escape from crowded cell islands. In 3D culture, proliferation is reduced upon MET with differential effects of focal-adhesion-signalling and actomyosin activity. In particular, inhibition of Src-signalling leads to increased growth after MET. Finally, in co-culture experiments, MET-induced tumor spheroids evade immune cell attack to a larger extent, likely due to more confined epithelial spheroid shape and changes in immunomodulatory molecules. Our data suggest that, contrary to previous assumptions in the field, MET might promote secondary tumor outgrowth not through a proliferation boost but through increased survival rate in the presence of immune cells.

cell biology↗

bioRxiv2024

Integrating various Experimental Information to Assist Protein Complex Structure Prediction by GRASP

Protein complex structures are essential for understanding of biological activities and drug development. Improving complex structure prediction accuracy of AI models for cases such as antigen-antibody complexes is expected to further enhance their applicability. Meanwhile, a large variety of experimental methods are used to provide structural insights for protein complexes, with only sparse or approximate knowledge obtained. A general tool is needed to integrate AI models with limited experimental information for high-throughput and accurate protein complex structure prediction. To efficiently and flexibly incorporate the different forms of experimental information, we introduce here GRASP. GRASP outperforms existing tools in handling both simulated and real-world experimental restraints including those obtained from XL, CL, CSP, and DMS. As an example, GRASP excels in predicting antigen-antibody complex structures, even surpassing AF3 when utilizing experimental DMS and CL restraints. In addition to accelerating the restrained modeling process, its ability to integrate multiple forms of restraints makes it capable of integrative modeling. We also showcase its potential in modeling protein structural interactome in the near-cellular condition based on large-scale in vivo XL data for mitchondria.

bioinformatics↗

bioRxiv2024

Higher resolution pooled genome-wide CRISPR knockout screening in Drosophila cells using integration and anti-CRISPR (IntAC)

CRISPR screens enable systematic, scalable genotype-to-phenotype mapping. We previously developed a pooled CRISPR screening method for Drosophila melanogaster and mosquito cell lines using plasmid transfection and site-specific integration to introduce single guide (sgRNA) libraries, followed by PCR and sequencing of integrated sgRNAs. While effective, the method relies on early constitutive Cas9 activity that potentially can lead to discrepancies between genome edits and sgRNAs detected by PCR, reducing screen accuracy. To address this issue, we introduce a new method to co-transfect a plasmid expressing the anti-CRISPR protein AcrIIa4 to suppress Cas9 activity during early sgRNA expression, which we term "IntAC" (integrase with anti-CRISPR). IntAC allowed us to construct a new CRISPR screening approach driven by the high strength dU6:3 promoter. This new library dramatically improved precision-recall of fitness genes across the genome, retrieving 90-95% of essential gene groups within 5% error, allowing us to generate the most comprehensive list of cell fitness genes yet assembled for Drosophila. Our analysis determined that elevated sgRNA levels, made permissible by the IntAC approach, drove much of the improvement. The Drosophila fitness genes show strong correlation with human fitness genes and underscore the effects of paralogs on gene essentiality. We further demonstrate that IntAC combined with a targeted sgRNA sub-library enabled precise positive selection of a transporter under solute overload. IntAC represents a straightforward enhancement to existing Drosophila CRISPR screening methods, dramatically increasing accuracy, and might also be broadly applicable to virus-free CRISPR screens in other cell types, including mosquito, lepidopteran, tick, and mammalian cells.

genetics↗

bioRxiv2024

The transcription factor ZmbZIP75 promotes both grain filling and kernel dehydration in maize

Selecting both high-yield and low-kernel-moisture varieties is essential for modern maize production, but relevant breeding efforts are hindered by a lack of valuable regulatory genes. Here, we demonstrate that the transcription factor (TF) basic leucine zipper 75 (ZmbZIP75) promotes grain yield and reduces kernel moisture in maize. Knockout of ZmbZIP75 results in defective grain filling and kernel dehydration, whereas ZmbZIP75 overexpression confers increased grain yield per plant and decreased kernel moisture without altering plant architecture. Mechanistically, during the grain filling stage, ZmbZIP75 is transcriptionally induced by maternal-derived basal abscisic acid (ABA) and directly activates multiple core starch synthesis-related genes and key TFs, thereby promoting grain filling and final yield. In the late stage of kernel development, high concentrations of zygotic ABA enhance ZmbZIP75 phosphorylation through SnRK2.10. The phosphorylated ZmbZIP75 subsequently transactivates and interacts with TF VP1 to synergistically promote kernel dehydration. This study thus highlights the potential of ZmbZIP75 for engineering both high-yield and low-kernel-moisture varieties to meet the demands of high-efficient maize production. IN A NUTSHELLO_ST_ABSBackgroundC_ST_ABSHigh grain yield in maize is generally associated with elevated kernel moisture at harvest, which is the main limiting factor for modern maize production. Therefore, it is of importance to select varieties with both high-yield and low-kernel-moisture. These traits are largely controlled by grain filling and kernel dehydration, two tightly connected processes during maize kernel development. Abscisic acid (ABA) is well-documented for its vital role in grain filling and dehydration maturation. However, the molecular mechanisms by which ABA coordinates these two processes remain unclear. QuestionWhat are the regulatory factors involved, and how do they mediate ABA signaling to coordinate grain filling and dehydration maturation in maize? FindingsZmbZIP75 is transcriptionally induced by ABA and directly activates multiple core starch synthesis-related genes and key TFs in developing maize endosperm, thereby promoting grain filling. In developing embryo, ABA enhances ZmbZIP75 phosphorylation via SnRK2.10. The phosphorylated ZmbZIP75 then transactivates and interacts with VP1 to synergistically promote kernel dehydration. Moreover, ZmbZIP75 overexpression confers increased grain yield and reduced kernel moisture in maize. Next stepsWhile ZmbZIP75 is directly phosphorylated to mediate ABA signaling during dehydration maturation, ZmbZIP75 is transcriptionally induced by ABA during grain filling. We plan to identify upstream factors that mediate ABA signaling to regulate ZmbZIP75 expression, thereby enhancing our understanding of ABA-promoted grain filling in maize.

plant biology↗

bioRxiv2024

A 15-Gene prognostic signature with TFAP2B functioning in Platinum Resistance of Ovarian Carcinoma

Resistance to platinum is the main challenge in the chemotherapy of ovarian cancer (OV). Therefore, developing a response signature to platinum is essential for the precision therapy of OV. Existing quantitative signatures of platinum are susceptible to batch effects and sequencing platform variations. To address this, we developed a transcriptome-based platinum signature, named PRSM, consisting of 15 genes, based on within-sample prognostic and relative expression ordering of genes, to predict individual responses to platinum in OV. The PRSM model demonstrated superior classification accuracy compared to previous quantitative signatures. Resistant samples classified by PRSM exhibited poorer overall survival, lower SNV neoantigen load, tumor mutational burden, and distinct methylation patterns compared to sensitive samples. Pathway analysis revealed the activation of MYC targets V2 and oxidative phosphorylation in resistant tumors. Single-cell analysis highlighted the roles of NK and epithelial cells in resistance. Among the 15 core genes, five (TFAP2B, KRT81, PAGE1, CRNN, UGT2B17) were linked to poor prognosis, with TFAP2B having the highest contribution to PRSM. Overexpression of TFAP2B in A2780 cells enhanced cisplatin sensitivity, while in A2780cis cells, it inhibited growth. In brief, our findings provide a multi-dimensional view of platinum resistance in ovarian cancer, introducing a robust predictive model and identifying potential therapeutic targets.

pharmacology and toxicology↗

bioRxiv2024

Deep Generative Optimization of mRNA Codon Sequences for Enhanced Protein Production and Therapeutic Efficacy

Messenger RNA (mRNA) therapeutics show immense promise, but their efficacy is limited by suboptimal protein expression. Here, we present RiboDecode, a deep learning framework that generates mRNA codon sequences for enhanced mRNA translation. RiboDecode introduces several advances, including direct learning from large-scale ribosome profiling data and generative exploration of a large sequence space. In silico analysis demonstrates RiboDecode s robust predictive accuracy for unseen genes and cellular environments. In vitro experiments showed substantial improvements in protein expression, significantly outperforming past methods. In addition, RiboDecode enables mRNA design with consideration of cellular context and demonstrates robust performance across different mRNA formats, including m1{Psi}-modified and circular mRNAs, an important feature for mRNA therapeutics. In vivo mouse studies showed that optimized influenza hemagglutinin mRNAs induce ten times stronger neutralizing antibody responses against influenza virus compared to the unoptimized sequence. In an optic nerve crush model, optimized nerve growth factor mRNAs achieve equivalent neuroprotection of retinal ganglion cells at one-fifth the dose of the unoptimized sequence. Collectively, RiboDecode represents a paradigm shift from rule-based to a data-driven, context-aware approach for mRNA therapeutic applications, enabling the development of more potent and dose-efficient treatments.

bioinformatics↗