Biology subjects

Zou, J.

Publications and source records attributed to Zou, J..

At least 91 records · Page 5Linked to original sources

bioRxiv2024

VirID: Beyond Virus Discovery - An Integrated Platform for Comprehensive RNA Virus Characterization

RNA viruses exhibit vast phylogenetic diversity and can significantly impact public health and agriculture. However, current bioinformatics tools for viral discovery from metagenomic data frequently generate false positive virus results, overestimate viral diversity, and misclassify virus sequences. Additionally, current tools often fail to determine virus-host associations, which hampers investigation of the potential threat posed by a newly detected virus. To address these issues we developed VirID, a software tool specifically designed for the discovery and characterization of RNA viruses from metagenomic data. The basis of VirID is a comprehensive RNA-dependent RNA polymerase (RdRP) database to enhance a workflow that includes RNA virus discovery, phylogenetic analysis, and phylogeny-based virus characterization. Benchmark tests on a simulated data set demonstrated that VirID had high accuracy in profiling viruses and estimating viral richness. In evaluations with real-world samples, VirID was able to identity RNA viruses of all type, but also provided accurate estimations of viral genetic diversity and virus classification, as well as comprehensive insights into virus associations with humans, animals, and plants. VirID therefore offers a robust tool for virus discovery and serves as a valuable resource in basic virological studies, pathogen surveillance, and early warning systems for infectious disease outbreaks.

bioinformatics↗

bioRxiv2024

Transformer-based modeling of Clonal Selection and Expression Dynamics (TraCSED) reveals resistance signatures in breast cancer

Understanding transcriptional heterogeneity in cancer cells and its implication for treatment response is critical to identify how resistance occurs and may be targeted. Such heterogeneity can be captured by in vitro studies through clonal barcoding methods. We present TraCSED (Transformer-based modeling of Clonal Selection and Expression Dynamics), a dynamic deep learning approach for modeling clonal selection. Using single-cell gene expression and the fitness of barcoded clones, TraCSED identifies interpretable gene programs and the timepoints at which they are associated with clonal selection. When applied to cells treated with either giredestrant, an estrogen receptor (ER) antagonist and degrader, or palbociclib, a CDK4/6 inhibitor, time-dependent resistance pathways are revealed. For example, ER activity is associated with positive selection around day four under palbociclib treatment and this adaptive response can be suppressed by combining the drugs. Yet, in the combination treatment, one clone still emerged. Clustering based on partial least squares regression found that high baseline expression of both SNHG25 and SNCG genes was the primary marker of positive selection to co-treatment and thus potentially associated with innate resistance - an aspect that traditional differential analysis methods missed. In conclusion, TraCSED enables associating pathways with phenotypes in a time-dependent manner from scRNA-seq data.

cancer biology↗

bioRxiv2024

Mutual information for detecting multi-class biomarkers when integrating multiple bulk or single-cell transcriptomic studies

MotivationBiomarker detection plays a pivotal role in biomedical research. Integrating omics studies from multiple cohorts can enhance statistical power, accuracy and robustness of the detection results. However, existing methods for horizontally combining omics studies are mostly designed for two-class scenarios (e.g., cases versus controls) and are not directly applicable for studies with multi-class design (e.g., samples from multiple disease subtypes, treatments, tissues, or cell types). ResultsWe propose a statistical framework, namely Mutual Information Concordance Analysis (MICA), to detect biomarkers with concordant multi-class expression pattern across multiple omics studies from an information theoretic perspective. Our approach first detects biomarkers with con-cordant multi-class patterns across partial or all of the omics studies using a global test by mutual information. A post hoc analysis is then performed for each detected biomarkers and identify studies with concordant pattern. Extensive simulations demonstrate improved accuracy and successful false discovery rate control of MICA compared to an existing MCC method. The method is then applied to two practical scenarios: four tissues of mouse metabolism-related transcriptomic studies, and three sources of estrogen treatment expression profiles. Detected biomarkers by MICA show intriguing biological insights and functional annotations. Additionally, we implemented MICA for single-cell RNA-Seq data for tumor progression biomarkers, highlighting critical roles of ribosomal function in the tumor microenvironment of triple-negative breast cancer and underscoring the potential of MICA for detecting novel therapeutic targets. Availabilityhttps://github.com/jianzou75/MICA

bioinformatics↗

bioRxiv2024

Loss of neuronal lysosomal acid lipase drives amyloid pathology in Alzheimer's disease

Underlying drivers of late-onset Alzheimers disease (LOAD) pathology remain unknown. However, multiple biologically diverse risk factors share a common pathological progression. To identify convergent molecular abnormalities that drive LOAD pathogenesis we compared two common midlife risk factors for LOAD, heavy alcohol use and obesity. This revealed that disrupted lipophagy is an underlying cause of LOAD pathogenesis. Both exposures reduced lysosomal flux, with a loss of neuronal lysosomal acid lipase (LAL). This resulted in neuronal lysosomal lipid (NLL) accumulation, which opposed A{beta} localization to lysosomes. Neuronal LAL loss both preceded (with aging) and promoted (targeted knockdown) A{beta} pathology and cognitive deficits in AD mice. The addition of recombinant LAL ex vivo and neuronal LAL overexpression in vivo prevented amyloid increases and improved cognition. In WT mice, neuronal LAL declined with aging and correlated negatively with entorhinal A{beta}. In healthy human brain, LAL also declined with age, suggesting this contributes to the age-related vulnerability for AD. In human LOAD LAL was further reduced, correlated negatively with A{beta}1-42, and occurred with polymerase pausing at the LAL gene. Together, this finds that the loss of neuronal LAL promotes NLL accumulation to impede degradation of A{beta} in neuronal lysosomes to drive AD amyloid pathology. SummaryCellular and molecular drivers of late-onset Alzheimers disease (LOAD) are unknown, though several risk factors account for the majority of disease incidence1-5. Though diverse in their biological natures, each of these risk exposures converge on a shared pathological progression with the accumulation of amyloid early in the disease. Human genetic and transcriptomic studies suggest a role for altered lipid metabolism6-9, though the mechanism has been unknown. Here, using two common midlife risk exposures for LOAD, we found that dysfunctional lipophagy caused by the loss of lysosomal acid lipase (LAL) promotes early LOAD pathogenesis. Both midlife obesity and heavy alcohol reduced neuronal LAL, causing an increase in neuronal lysosomal lipid, and a subsequent accumulation of A{beta} in the extra-lysosomal cytosol. This loss of LAL preceded and promoted A{beta} pathology and cognitive deficits in AD mice. The addition of recombinant LAL ex vivo and neuronal LAL overexpression in vivo prevented increases in amyloid and improved cognition. In human brain, LAL declined with age in healthy subjects, similar to rodents, showing robust losses in LOAD subjects with polymerase pausing. Together, this implicates neuronal LAL loss in LOAD pathogenesis and presents LAL as a promising diagnostic, preventative, and/or therapeutic target for AD.

neuroscience↗

bioRxiv2024

ProteinCLIP: enhancing protein language models with natural language

Language models have enabled a new era of biological sequence modeling. However, extracting meaningful sequence-level embeddings from these models remains challenging. In this work, we introduce ProteinCLIP, which applies contrastive learning between a proteins amino acid sequence and curated text describing its function. ProteinCLIP thus learns to take a pre-trained protein language models sequence embedding and refines it produce a function-centric embedding. We show that this embedding space yields sequence representations that enable state-of-the-art performance across a variety of important yet challenging tasks in the study of proteins - from predicting protein protein interactions to accurately detecting homologous proteins despite low sequence similarity. More broadly, ProteinCLIP demonstrates the effectiveness of multi-modal learning in biological contexts, and how such strategies can help isolate key signals from large models and further improve their utility.

bioinformatics↗

bioRxiv2024

Methyltransferase DNMT3B promotes colorectal cancer cell proliferation by inhibiting PLCG2

BackgroundAberrant DNA methylation patterns in the promoter region of PLCG2 have been associated with dysregulated signaling pathways and cellular functions. Its role in colorectal cancer cells is still unknown. MethodsqRT-PCR was used to examine DNMT3B expression in colorectal cancer. Western blot and immunohistochemistry were used to analyze DNMT3B and PLCG2 protein levels in colorectal tissues and cell lines. The cell counting kit-8 and colony experiments were used to identify the proliferation of colorectal cancer cells. Methylation-specific PCR (MSP) and bisulfite-sequencing PCR (BSP) was used to measure DNA methylation levels. ResultsDNMT3B is overexpressed in colorectal cell in TCGA datasets and Kaplan-Meier plots. DNMT3B is significantly overexpressed in tumor tissues compared to adjacent non-tumor tissues. Western blotting results demonstrated high expression of DNMT3B in tumor tissues. Compared to normal colonic epithelial cells, colorectal cancer cell lines exhibited elevated levels of PLCG2 methylation. oePLCG2 effectively prevented the in vivo xenograft tumor growth of colorectal cancer. ConclusionsPLCG2 is identified as a key downstream regulatory protein of DNMT3B in colorectal cancer. DNMT3B Inhibits PLCG2 transcription through methylation of the PLCG2 promoter region. DNMT3B controls colorectal cancer cell proliferation through the PLCG2, which is useful for creating therapeutic approaches that target PLCG2 expression for the treatment of colorectal cancer.

cancer biology↗

bioRxiv2024

Molecular basis for condensin enrichment at pericentromeres

Faithful chromosome segregation requires packaging of the genome on both global and local scales. Condensin plays a crucial role at pericentromeres to resist spindle forces and ensure the bioriented attachment of kinetochores to microtubules in mitosis. Here we demonstrate that budding yeast condensin is recruited to pericentromeres through a direct interaction between its Ycg1 subunit and the pericentromeric adaptor protein, shugoshin (Sgo1). We identify a Short Linear Motif (SLiM), termed CR1, within the C-terminal region of Sgo1 which inserts into a conserved pocket on Ycg1. Disruption of this interface abolishes the Sgo1-condensin interaction, prevents condensin recruitment to pericentromeres and results in defective sister kinetochore biorientation in mitosis. Similar motifs to CR1 are found in known and potential condensin binding partners and the Ycg1 binding pocket is broadly conserved, including in the mammalian homolog CAP-G. Overall, we uncover the molecular mechanism that targets condensin to define a specialized chromosomal domain.

cell biology↗

bioRxiv2024

Mapping single-cell developmental potential in health and disease with interpretable deep learning

Single-cell RNA sequencing (scRNA-seq) has transformed our understanding of cell fate in developmental systems. However, identifying the molecular hallmarks of potency - the capacity of a cell to differentiate into other cell types - has remained challenging. Here, we introduce CytoTRACE 2, an interpretable deep learning framework for characterizing potency and differentiation states on an absolute scale from scRNA-seq data. Across 31 human and mouse scRNA-seq datasets encompassing 28 tissue types, CytoTRACE 2 outperformed existing methods for recovering experimentally determined potency levels and differentiation states covering the entire range of cellular ontogeny. Moreover, it reconstructed the temporal hierarchy of mouse embryogenesis across 62 timepoints; identified pan-tissue expression programs that discriminate major potency levels; and facilitated discovery of cellular phenotypes in cancer linked to survival and immunotherapy resistance. Our results illuminate a fundamental feature of cell biology and provide a broadly applicable platform for delineating single-cell differentiation landscapes in health and disease.

genomics↗

bioRxiv2024

GGCX promotes Eurasian avian-like H1N1 swine influenza virus adaption to interspecies receptor binding

The Eurasian avian-like (EA) H1N1 swine influenza virus (SIV) possesses the capacity to instigate the next influenza pandemic, owing to its heightened affinity for the human-type -2,6 sialic acid (SA) receptor. Nevertheless, the molecular mechanisms underlying the switch in receptor binding preferences of EA H1N1 SIV remain elusive. In this study, we conducted a comprehensive genome-wide CRISPR/Cas9 knockout screen utilizing EA H1N1 SIV in porcine kidney cells. Knocking out the enzyme gamma glutamyl carboxylase (GGCX) reduced virus replication in vitro and in vivo by inhibiting the carboxylation modification of viral haemagglutinin (HA) and the adhesion of progeny viruses, ultimately impeding the replication of EA H1N1 SIV. Furthermore, GGCX was revealed to be the determinant of the D225E substitution of EA H1N1 SIV, and GGCX-medicated carboxylation modification of HA 225E contributed to the receptor binding adaption of EA H1N1 SIV to the -2,6 SA receptor. Taken together, our CRISPR screen has elucidated a novel function of GGCX in the support of EA H1N1 SIV adaption for binding to -2,6 SA receptor. Consequently, GGCX emerges as a prospective antiviral target against the infection and transmission of EA H1H1 SIV.

microbiology↗

bioRxiv2024

The Ubiquitin Ligase RBX2/SAG Regulates Mitochondrial Ubiquitination and Mitophagy

Clearance of damaged mitochondria via mitophagy is crucial for cellular homeostasis. While the role of ubiquitin (Ub) ligase PARKIN in mitophagy has been extensively studied, increasing evidence suggests the existence of PARKIN-independent mitophagy in highly metabolically active organs such as the heart. Here, we identify a crucial role for Cullin-RING Ub ligase 5 (CRL5) in basal mitochondrial turnover in cardiomyocytes. CRL5 is a multi-subunit Ub ligase comprised by the catalytic RING box protein RBX2 (also known as SAG), scaffold protein Cullin 5 (CUL5), and a substrate-recognizing receptor. Analysis of the mitochondrial outer membrane-interacting proteome uncovered a robust association of CRLs with mitochondria. Subcellular fractionation, immunostaining, and immunogold electron microscopy established that RBX2 and Cul5, two core components of CRL5, localizes to mitochondria. Depletion of RBX2 inhibited mitochondrial ubiquitination and turnover, impaired mitochondrial membrane potential and respiration, and increased cell death in cardiomyocytes. In vivo, deletion of the Rbx2 gene in adult mouse hearts suppressed mitophagic activity, provoked accumulation of damaged mitochondria in the myocardium, and disrupted myocardial metabolism, leading to rapid development of dilated cardiomyopathy and heart failure. Similarly, ablation of RBX2 in the developing heart resulted in dilated cardiomyopathy and heart failure. Notably, the action of RBX2 in mitochondria is not dependent on PARKIN, and PARKIN gene deletion had no impact on the onset and progression of cardiomyopathy in RBX2-deficient hearts. Furthermore, RBX2 controls the stability of PINK1 in mitochondria. Proteomics and biochemical analyses further revealed a global impact of RBX2 deficiency on the mitochondrial proteome and identified several mitochondrial proteins as its putative substrates. These findings identify RBX2-CRL5 as a mitochondrial Ub ligase that controls mitophagy under physiological conditions in a PARKIN-independent, PINK1-dependent manner, thereby regulating cardiac homeostasis.

pathology↗

bioRxiv2024

PLK1-Mediated Phosphorylation Cascade Activates the Mis18 Complex to Ensure Centromere Inheritance

Accurate chromosome segregation requires the attachment of spindle microtubules to centromeres, which are epigenetically defined by the enrichment of CENP-A nucleosomes. During DNA replication, existing CENP-A nucleosomes undergo dilution as they get redistributed among the two DNA strands. To preserve centromere identity, CENP-A levels must be restored in a cell-cycle controlled manner orchestrated by the Mis18 complex. Here we provide a comprehensive mechanistic basis for PLK1-mediated licensing of CENP-A loading. We demonstrate that PLK1 interacts with Mis18 and Mis18BP1 subunits of the Mis18 complex by recognising self-primed phosphorylations of Mis18 (S54) and Mis18BP1 (T78 and S93) through its Polo-box binding domain. Disrupting these PLK1 phosphorylations perturbed the centromere recruitment of HJURP and new CENP-A loading. Biochemical and functional analyses show that phosphorylation of Mis18 and subsequent PLK1 binding is required to activate the Mis18/{beta} complex for robust Mis18/{beta}-HJURP interaction. Thus, our study reveals key molecular events underpinning the licensing role of PLK1 in ensuring accurate centromere inheritance. One-Sentence SummaryPLK1 phosphorylation cascade licenses CENP-A loading by facilitating HJURP centromere recruitment via Mis18/{beta} activation.

cell biology↗

bioRxiv2024

SPRITE: improving spatial gene expression imputation with gene and cell networks

Spatially resolved single-cell transcriptomics have provided unprecedented insights into gene expression in situ, particularly in the context of cell interactions or organization of tissues. However, current technologies for profiling spatial gene expression at single-cell resolution are generally limited to the measurement of a small number of genes. To address this limitation, several algorithms have been developed to impute or predict the expression of additional genes that were not present in the measured gene panel. Current algorithms do not leverage the rich spatial and gene relational information in spatial transcriptomics. To improve spatial gene expression predictions, we introduce SPRITE (Spatial Propagation and Reinforcement of Imputed Transcript Expression) as a meta-algorithm that processes predictions obtained from existing methods by propagating information across gene correlation networks and spatial neighborhood graphs. SPRITE improves spatial gene expression predictions across multiple spatial transcriptomics datasets. Furthermore, SPRITE predicted spatial gene expression leads to improved clustering, visualization, and classification of cells. SPRITE is available as a software package and can be used in spatial transcriptomics data analysis to improve inferences based on predicted gene expression.

bioinformatics↗

bioRxiv2024

Cisplatin drives mitochondrial dysregulation in sensory hair cells

Cisplatin is a chemotherapy drug that causes permanent hearing loss by injuring cochlear hair cells. The mechanisms that initiate injury are not fully understood, but mitochondria have emerged as potential mediators of hair cell cytotoxicity. Using in vivo live imaging of hair cells in the zebrafish lateral-line organ expressing a genetically encoded indicator of cumulative mitochondrial activity, we first demonstrate that greater redox history increases susceptibility to cisplatin. Next, we conducted time-lapse imaging to understand dynamic changes in mitochondrial homeostasis and observe elevated mitochondrial and cytosolic calcium that surge prior to hair cell death. Furthermore, using a localized probe that fluoresces in the presence of cisplatin, we show that cisplatin directly accumulates in hair cell mitochondria, and this accumulation occurs before mitochondrial dysregulation and apoptosis. Our findings provide evidence that cisplatin directly targets hair cell mitochondria and support that the mitochondria are integral to cisplatin cytotoxicity in hair cells.

neuroscience↗

bioRxiv2024

A Murine Model of Post-acute Neurological Sequelae Following SARS-CoV-2 Variant Infection

Viral variant is one known risk factor associated with post-acute sequelae of COVID-19 (PASC), yet the pathogenesis is largely unknown. Here, we studied SARS-CoV-2 Delta variant-induced PASC in K18-hACE2 mice. The virus replicated productively, induced robust inflammatory responses in lung and brain tissues, and caused weight loss and mortality during the acute infection. Longitudinal behavior studies in surviving mice up to 4 months post-acute infection revealed persistent abnormalities in neuropsychiatric state and motor behaviors, while reflex and sensory functions recovered over time. Surviving mice showed no detectable viral RNA in the brain and minimal neuroinflammation post-acute infection. Transcriptome analysis revealed persistent activation of immune pathways, including humoral responses, complement, and phagocytosis, and reduced levels of genes associated with ataxia telangiectasia, impaired cognitive function and memory recall, and neuronal dysfunction and degeneration. Furthermore, surviving mice maintained potent T helper 1 prone cellular immune responses and high neutralizing antibodies against Delta and Omicron variants in the periphery for months post-acute infection. Overall, infection in K18-hACE2 mice recapitulates the persistent clinical symptoms reported in long COVID patients and may be useful for future assessment of the efficacy of vaccines and therapeutics against SARS-CoV-2 variants.

microbiology↗

bioRxiv2023

ADMET-AI: A machine learning ADMET platform for evaluation of large-scale chemical libraries

SummaryThe emergence of large chemical repositories and combinatorial chemical spaces, coupled with high-throughput docking and generative AI, have greatly expanded the chemical diversity of small molecules for drug discovery. Selecting compounds for experimental validation requires filtering these molecules based on favourable druglike properties, such as Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET). We developed ADMET-AI, a machine learning platform that provides fast and accurate ADMET predictions both as a website and as a Python package. ADMET-AI has the highest average rank on the TDC ADMET Benchmark Group leaderboard, and it is currently the fastest web-based ADMET predictor, with a 45% reduction in time compared to the next fastest ADMET web server. ADMET-AI can also be run locally with predictions for one million molecules taking just 3.1 hours. Availability and ImplementationThe ADMET-AI platform is freely available both as a web server at admet.ai.greenstonebio.com and as an open-source Python package for local batch prediction at github.com/swansonk14/admet_ai (also archived on Zenodo at doi.org/10.5281/zenodo.10372930). All data and models are archived on Zenodo at doi.org/10.5281/zenodo.10372418.

bioinformatics↗

bioRxiv2023

A Multi-Granularity Approach to Similarity Search in Multiplexed Immunofluorescence Images

Due to the rapid increase and importance of multiplexed immunofluorescence (mIF) imaging data in spatial biology, there is a pressing need to develop efficient image-to-image search pipelines for both diagnostic and research purposes. While several image search methods have been introduced for conventional images and digital pathology, mIF images present three main challenges: (1) high dimension-ality, (2) domain-specificity, and (3) complex additional molecular information. To address this gap, we introduce the MIISS framework, a Multi-granularity mIF Image Similarity Search pipeline that employs self-supervised learning models to extract features from mIF image patches and an entropy-based aggregation method to enable similarity searches at higher, multi-granular levels. We then benchmarked various feature generation approaches to handle high dimensional images and tested them on various foundation models. We conducted evaluations using datasets from different tissues on both patch- and patient-level, which demonstrate the frame-works effectiveness and generalizability. Notably, we found that domain-specific models consistently outperformed other models, further showing their robustness and generalizability across different datasets. The MIISS framework offers an effective solution for navigating the growing landscape of mIF images, providing tangible clinical benefits and opening new avenues for pathology research.

bioinformatics↗

bioRxiv2023

The N-terminal region of DNMT3A combines multiple chromatin reading motifs to guide recruitment.

DNA methyltransferase 3A (DNMT3A) plays a critical role in establishing and maintaining DNA methylation patterns. However, the mechanisms underlying DNMT3A recruitment to and function within different chromatin environments remain unclear. Using a combination of biochemical and structural approaches we find that DNMT3A interacts using multiple interfaces with chromatin; directly binding generic nucleosome features as well as site-specific post-translational histone modifications. The N-terminal region, unique to the DNMT3A1 isoform, is essential for these interactions and stabilises H3K36me2-nucleosome recruitment. Intriguingly, in the same region critical for nucleosome binding we also map a ubiquitylation-dependent recruitment motif (UDR). The UDR binds specifically to ubiquitylated H2AK119, explaining the previously observed recruitment to Polycomb-occupied heterochromatin. A cryo-EM structure of DNMT3A1-DNMT3L with a modified nucleosome reveals that the UDR interacts with the nucleosome surface including the acidic patch. Previously unexplained disease-associated mutations are present in the UDR and ablate nucleosome interactions. This leads to an increased understanding of how DNMT3A1 recruitment occurs in the genome and highlights the importance of multivalent binding of DNMT3A to histone modifications and the nucleosome.

biochemistry↗

bioRxiv2023

GENEPT: A SIMPLE BUT HARD-TO-BEAT FOUNDATION MODEL FOR GENES AND CELLS BUILT FROM CHATGPT

There has been significant recent progress in leveraging large-scale gene expression data to develop foundation models for single-cell biology. Models such as Geneformer and scGPT implicitly learn gene and cellular functions from the gene expression profiles of millions of cells, which requires extensive data curation and resource-intensive training. Here we explore a much simpler alternative by leveraging ChatGPT embeddings of genes based on literature. Our proposal, GenePT, uses NCBI text descriptions of individual genes with GPT-3.5 to generate gene embeddings. From there, GenePT generates single-cell embeddings in two ways: (i) by averaging the gene embeddings, weighted by each genes expression level; or (ii) by creating a sentence embedding for each cell, using gene names ordered by the expression level. Without the need for dataset curation and additional pretraining, GenePT is efficient and easy to use. On many downstream tasks used to evaluate recent single-cell foundation models -- e.g., classifying gene properties and cell types -- GenePT achieves comparable, and often better, performance than Geneformer and other models. GenePT demonstrates that large language model embedding of literature is a simple and effective path for biological foundation models.

bioinformatics↗