Biology subjects

Fang, Z.

Publications and source records attributed to Fang, Z..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

ADAMTS3 as a promising novel biomarker for the diagnosis of hepatocellular carcinoma

Early diagnosis of hepatocellular carcinoma (HCC) still represents a significant challenge. The rising of obesity and non-viral liver disease associated HCC further limits the effectiveness of conventional screening approaches such as ultrasonography and alpha-fetoprotein (AFP) testing. In this context, the identification of potential effective biomarkers for rapid and precise diagnosis of HCC still represents an extremely important task. Here, we identify ADAMTS3 as a novel mechanistic serum biomarker for HCC that addresses this shifting etiological landscape. Transcriptome analysis of four independent cohorts-including HCC patients, individuals with obesity-or non-viral liver disease, and healthy controls- reveals four secreted proteins associated with HCC progression. Among them, ADAMTS3 demonstrates strong diagnostic performance by ROC (receiver operating characteristic) and LASSO (least absolute shrinkage and selection operator) analysis. High ADAMTS3 expression correlates with aggressive molecular signature, characterized by enhanced proliferative signaling, suppressed immune effector responses, and a reprogrammed immune microenvironment marked by reduced NK cell infiltration and increased accumulation of immunosuppressive tumor-associated macrophages, driven by extracellular matrix stiffening. Knockdown of ADAMTS3 in cell lines significantly reduced proliferation, and clonogenic potential. Reintroduction of ADAMTS3 partially rescued both phenotypes, further confirming its role in promoting HCC cell growth. Optical tweezer-based measurements further reveal reduced cell stiffness upon ADAMTS3 deficiency, highlighting its role in extracellular matrix (ECM) remodeling. Finally, to provide a simple strategy for HCC diagnosis based on this novel biomarker, we demonstrated a DNA origami-based SERS biosensor capable of detecting ADAMTS3 at 10-11 M. Together, these findings identify ADAMTS3 as a mechanistic biomarker and demonstrate a translational sensing strategy for non-invasive HCC early screening.

cancer biology↗

bioRxiv2025

Impact of 2'-deoxyribo-purine substrates on nonenzymatic RNA template-directed primer extension

The composition of the primordial genetic material remains uncertain. Studies of duplex structure and stability, and of nonenzymatic template copying chemistry, provide insight into the viability of potentially primordial genetic polymers. Recent work suggests that 2'- deoxyribo-purine nucleotides may have been generated together with ribonucleotides on the early Earth. Since DNA/RNA duplexes are known to be less stable than RNA/RNA duplexes, we have examined the impact of dA, dI, and dG substitutions on RNA structure and nonenzymatic template copying. We find that single 2'-deoxyribo-purine substitutions reduce RNA duplex stability, as expected. Crystallographic studies show that such substitutions lead to minimal structural changes but point to diminished solvation as a likely reason for duplex destabilization. Kinetic studies show that dI and dG substrates exhibit slightly weaker template binding and slower rates of template-directed primer extension than the corresponding ribo-purine substrates. In contrast, dA substrates exhibit much slower reaction kinetics but higher template affinity than rA substrates. Our results suggest that a mixed RNA/DNA primordial genetic polymer would have suffered from moderately slower rates of template copying, but that this could have been offset by an advantage due to more facile strand separation or exchange. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=108 SRC="FIGDIR/small/673048v1_ufig1.gif" ALT="Figure 1"> View larger version (28K): org.highwire.dtl.DTLVardef@15bd09aorg.highwire.dtl.DTLVardef@dc49d4org.highwire.dtl.DTLVardef@730d20org.highwire.dtl.DTLVardef@162b84f_HPS_FORMAT_FIGEXP M_FIG C_FIG

biochemistry↗

bioRxiv2025

DrugLM: A Unified Framework to Enhance Drug-Target Interaction Predictions by Incorporating Textual Embeddings via Language Models

MotivationAccurate prediction of drug-target interactions (DTIs) is central to computational drug discovery, offering the potential to reduce experimental costs and accelerate development timelines. While existing deep learning approaches such as Graph Neural Networks and Transformers have shown promise, they often overlook the rich semantic information embedded in textual descriptions of drugs and targets. These descriptions encode critical biomedical knowledge, including mechanisms of action, biological pathways involved, and therapeutic effects of drugs, which can enhance DTI prediction performance. ResultsWe introduce DrugLM, a unified framework that integrates embeddings derived from large language models (LLMs) into DTI-specific model architectures. DrugLM leverages textual descriptions of drugs and targets to generate semantic embeddings using a range of pretrained LLMs. These embeddings can be seamlessly incorporated into existing DTI models. We systematically evaluate multiple LLMs on benchmark DTI datasets and demonstrate strong performance even without fine-tuning. Moreover, supervised parameter-efficient fine-tuning of the LLMs further improves embedding quality, leading to enhanced prediction accuracy. Notably, a simple multilayer perceptron (MLP) using only LLM-derived embeddings surpasses several established DTI methods, underscoring the power of semantic features. Our findings highlight the practical value of integrating LLMs into DTI pipelines and offer a straightforward recipe for improved drug discovery: LLM embeddings of drugs and targets are both effective and easy to use. AvailabilityOur code and dataset are available at https://github.com/ShPhoebus/DrugLM

bioinformatics↗

bioRxiv2025

SpatialMap: A Scalable Deep Learning Method for Cell Typing in Subcellular Spatial Transcriptomics

Subcellular imaging transcriptomics has offered unprecedented resolution of spatial mapping of gene expressions at subcellular level. However, accurate and robust cell type annotation at cellular level remains a challenge. Here we present SpatialMap, a deep weakly supervised cell typing framework that combines graph neural networks (GNN) with transfer learning to annotate cellular identities in subcellular spatial transcriptomics (ST). SpatialMap consistently outperforms state-of-the-art annotation methods on both simulated and real-world ST datasets. It demonstrates strong scalability across diverse tissue types and platforms, including NanoString CosMx, MERFISH and 10x Genomics Xenium. When applied to a mouse primary motor cortex dataset, SpatialMap successfully identified Micro, L4/5 IT, and L6 IT Car3 absent from the reference data, showcasing its capability to discover novel cell types. Furthermore, SpatialMap accurately annotated malignant cells in human liver cancer samples. Its design also enables seamless extension to cross-omics cell typing and cross-platform annotation within proteomics.

bioinformatics↗

bioRxiv2025

Pp6-Pfkfb1 axis modulates intracellular bacterial proliferation by orchestrating host-pathogen metabolic crosstalk

Intracellular bacterial pathogens exhibit heterogeneous replication rates within host macrophages, but the mechanisms by which they manipulate host factors for survival remain incompletely understood. Using a fluorescence-dilution reporter system in Salmonella typhimurium (Salmonella)-infected macrophages, we found that Protein Phosphatase 6 (Pp6) was downregulated in macrophages harboring growing bacteria. Conditional knockout of Pp6 elevated host susceptibility to Salmonella-mediated lethality due to compromised antimicrobial defenses. MicroRNA-31 (miR-31) was identified as a negative regulator of Pp6, and its conditional ablation enhanced bacterial clearance. Yeast two-hybrid screening identified 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 1 (Pfkfb1), a metabolic regulator, as a substrate of Pp6. Pp6 deficiency resulted in significantly elevated expression of Pfkfb1, which was highly expressed in macrophages containing replicating Salmonella. Pfkfb1 deletion restricted bacterial proliferation by promoting nitric oxide (NO) production while concurrently suppressing arginase-1 (Arg-1) expression and impairing arginine metabolism in macrophages. Collectively, these results establish the Pp6-Pfkfb1 axis as a key regulator of host metabolic adaptation and intracellular bacterial survival, offering potential therapeutic targets against multidrug-resistant pathogens. Author summaryIntracellular pathogens secrete effector proteins that intercept and modify host cells to usurp host defenses and establish habitable intracellular niches, yet the host factors they manipulate for survival and virulence are not fully understood. Using a fluorescence-dilution reporter system in Salmonella typhimurium (Salmonella)-infected macrophages, we found Protein Phosphatase 6 (Pp6) was significantly decreased in macrophages harboring replicating Salmonella and its deficiency indulged bacterial growth in macrophages. However, the deletion of microRNA-31 (miR-31), a negative regulator of Pp6, enhanced Salmonella clearance. Furthermore, through yeast two- hybrid screening, we identified 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 1 (Pfkfb1) is a substrate of Pp6. And we demonstrated Pfkfb1, abundantly expressed in macrophages containing growing bacteria, is crucial for Salmonella proliferation by regulating nitric oxide (NO) levels, arginase-1 (Arg-1) expression and arginine metabolism. Overall, the molecular changes in Pp6-Pfkfb1 axis drive host metabolic adaptations that enable intracellular bacterial survival.

microbiology↗

bioRxiv2025

DNA framework array enables ultra-high throughput DNA synthesis

High-throughput DNA synthesis has revolutionized synthetic biology, molecular diagnostics, genome engineering, and DNA data storage by enabling the scalable and precise construction of nucleotide-based information systems. However, existing high-throughput synthesis technologies rely on top-down fabrication strategies to maximize throughput, yet they are inherently constrained by physical limitations. These constraints make it exceedingly difficult to achieve single-molecule level synthesis control that is essential for advancing next-generation DNA synthesis. Here, we present a DNA framework-based bottom-up enzymatic synthesis strategy that enables single-molecule level control of DNA synthesis with high-throughput. Leveraging the nanoscale addressability of DNA frameworks, we achieved a synthesis site pitch of 10.9 nm--183-fold smaller than that of previously reported electrode arrays. Theoretical projections suggest that this approach could scale synthesis throughput to 5.3 x 1011 sequences per square centimeter, representing a four-order-of-magnitude improvement over existing synthesis arrays. This approach lays the foundation for gigabyte-scale DNA writing and offers a highly accessible pathway toward large-scale DNA synthesis for emerging applications such as molecular data storage.

synthetic biology↗

bioRxiv2025

A Tandem Repeat Atlas for the Genome of Inbred Mouse Strains: A Genetic Variation Resource

Tandem repeats (TRs) are a significant source of genetic variation in the human population; and TR alleles are responsible for over 60 human genetic diseases and for inter-individual differences in many biomedical traits. Therefore, we utilized long-read sequencing and state of the art computational programs to produce a database with 2,528,854 TRs covering 39 inbred mouse strains. As in humans, murine TRs are abundant and were primarily located in intergenic regions. However, there were important species differences: murine TRs did not have the extensive number of repeat expansions like those associated with human repeat expansion diseases and they were not associated with transposable elements. We demonstrate by analysis of two biomedical phenotypes, which were identified over 40 years ago, that this TR database can enhance our ability to characterize the genetic basis for trait differences among the inbred strains.

genetics↗

bioRxiv2025

AI-assisted Drug Re-purposing for Human Liver Fibrosis

Liver fibrosis is a severe disease with few treatment options due to the poor quality of the available animal and in vitro models. To address this, we investigated whether a hypothesis generating multi-agent AI system (AI co-scientist) could assist in re-purposing drugs for treatment of liver fibrosis and direct their experimental characterization. A multi-parameter image analysis workflow, which enabled anti-fibrotic efficacy and drug toxicity to be serially assessed in multi-lineage human hepatic organoids grown in microwells (i.e., microHOs), was used to assess the effects of 14 drugs. Remarkably, two of the three AI co-scientist-recommended drugs that targeted epigenomic modifiers exhibited significant anti-fibrotic activity. Analysis of the anti-fibrotic effects of five drugs indicated that two inhibited TGF{beta}-induced intracellular signaling and three drugs altered TGF{beta}-induced mesenchymal cell differentiation. Since all five of the anti-fibrotic drugs reduced TGF{beta}-induced chromatin structural changes, epigenomic changes play an important role in the pathogenesis of liver fibrosis. One AI co-scientist recommended drug is an FDA-approved anti-cancer treatment (Vorinostat) that reduced TGF{beta}-induced chromatin structural changes by 91% and promoted liver parenchymal cell regeneration in microHOs. Hence, the use of AI co-scientist and this microHO platform identified a potential new generation of liver fibrosis treatments that also promote liver regeneration.

bioinformatics↗

bioRxiv2025

Nonenzymatic RNA copying with a potentially primordial genetic alphabet

Nonenzymatic RNA copying is thought to have been responsible for the replication of genetic information during the origin of life. However, chemical copying with the canonical nucleotides (A, U, G, and C) strongly favors the incorporation of G and C and disfavors the incorporation of A and especially U, because of the stronger G:C vs. A:U base pair, and the weaker stacking interactions of U. Recent advances in prebiotic chemistry suggest that the 2-thiopyrimidines were precursors to the canonical pyrimidines, raising the possibility that they may have played an important early role in RNA copying chemistry. Furthermore, 2-thiouridine (s2U) and inosine (I) form by deamination of 2-thiocytidine (s2C) and A respectively. We used thermodynamic and crystallographic analyses to compare the I:s2C and A:s2U base pairs. We find that the I:s2C base pair is isomorphic and isoenergetic with the A:s2U base pair. The I:s2C base pair is weaker than a canonical G:C base pair, while the A:s2U base pair is stronger than the canonical A:U base pair, so that a genetic alphabet consisting of s2U, s2C, I and A generates RNA duplexes with uniform base pairing energies. Consistent with these results, kinetic analysis of nonenzymatic template-directed primer extension reactions reveals that s2C and s2U substrates bind similarly to I and A in the template, and vice versa. Our work supports the plausibility of a potentially primordial genetic alphabet consisting of s2U, s2C, I and A, and offers a potential solution to the long-standing problem of biased nucleotide incorporation during nonenzymatic template copying. Significance StatementA long-standing challenge in primordial nonenzymatic RNA copying chemistry is the biased incorporation of C and G over A and U due to differences in base pair strength. We hypothesized that 2-thiopyrimidine substitution could help overcome this bias since A:s2U is a stronger version of the A:U base pair, and I:s2C is a weaker version of the G:C base pair. This study explores the efficacy of a potentially primordial genetic alphabet consisting of s2U, s2C, A and I. Our results show that A:s2U and I:s2C pairs are isoenergetic and isomorphic. Our findings highlight the potential of this alternative genetic alphabet to yield a more balanced incorporation of all nucleotides, facilitating information propagation by nonenzymatic RNA copying during the origin of life.

biochemistry↗

bioRxiv2025

PELSA-Decipher: a software tool for the processing and interpretation of ligand protein interaction dataset acquired by PELSA

Knowledge of protein-ligand interactions is essential for comprehending various aspects of life science, including drug mechanisms of action, regulatory processes in cellular metabolism and signaling, etc. Recently, a robust ligand modification-free method, termed as peptide-centric local stability assay (PELSA), is developed to identify the protein targets and binding regions of diverse ligands at proteomics scale. This method has unprecedented sensitivity and can be broadly applied to ligands including drugs, metabolites, metal ions, antibodies tec. However, the extraction of key information on ligand-protein interaction, including the binding protein, binding site, and binding affinity, is not a trivial task, which hampers the wide utilization of PELSA strategy by the research community. To address this, we developed a software tool, PELSA-Decipher, to facilitate the efficient processing of dataset obtained in PELSA experiment, including raw data processing, result visualization, and the generation of reports and high-quality images, thereby will greatly promote the broader applications of PELSA strategy. The PELSA-Decipher software can be downloaded free of charge from the website https://github.com/DICP-1809.

bioinformatics↗

bioRxiv2025

A Murine Database of Structural Variants Enables the Genetic Architecture of a Spontaneous Murine Lymphoma to be Characterized

A more complete map of the pattern of genetic variation among inbred mouse strains is essential for characterizing the genetic architecture of the many available mouse genetic models of important biomedical traits. Although structural variants (SVs) are a major component of genetic variation, they have not been adequately characterized among inbred strains due to methodological limitations. To address this, we generated high-quality long-read sequencing data for 40 inbred strains; and designed a pipeline to optimally identify and validate different types of SVs. This generated a database for 40 inbred strains with 573,191SVs, which included 10,815 duplications and 2,115 inversions, which also has 70 million SNPs and 7.5 million insertions/deletions. Analysis of this SV database identified an SV that is one component of a bi-genic model for lymphoma susceptibility in SJL mice, which provides mechanistic insight into the genetic basis for susceptibility to murine (and potentially human) lymphomas.

genetics↗

bioRxiv2025

Toward a privacy-preserving predictive foundation model of single-cell transcriptomics with federated learning and tabular modeling

The ability to pre-train on vast amounts of data to build foundation models (FMs) has achieved remarkable success in numerous domains, including natural language processing, computer vision, and, more recently, single-cell genomics--epitomized by GeneFormer, scGPT, and scFoundation. However, as single-cell FMs begin to train on increasingly large corpora, significant privacy and ethical concerns arise. Moreover, unlike text data, single-cell data is unordered and exhibits a unique tabular structure that most existing single-cell FMs overlook. In this study, we propose Tabula, a privacy-preserving and tabular-structure aware FM designed with federated learning (FL) and tabular modeling. Tabula combines the advantages of FMs and FL, enabling collaborative model training across multiple clients without compromising data privacy. In contrast to earlier single-cell FMs--which treat single-cell data like natural language (viewing cells as "words" defined by genes)--Tabula introduces a novel pretraining strategy that explicitly models the tabular structure of single-cell data. Extensive experimental results show that Tabula outperforms state-of-the-art methods in various downstream tasks (including cell type annotation, gene imputation, gene perturbation, multi-batch integration, and multi-omics integration) while requiring only half the data for pretraining and preserving data privacy. Furthermore, Tabula accurately reveals pairwise and even combinatorial regulatory logic across diverse biological systems, including hematopoiesis, pancreatic endogenesis, neurogenesis, and cardiogenesis. Thus, Tabula provides a new foundation model that explicitly incorporates the tabular nature of single-cell data alongside FL, paving the way for creating a "virtual cell" for human health under critical privacy preservation.

bioinformatics↗

bioRxiv2024

GlioME: A Novel Organoid Model Retaining the Glioma Mi-croenvironment for Personalized Drug Screening and Thera-peutic Evaluation

Glioma is an aggressive brain tumor with a poor prognosis. Establishing an in vitro culture model that closely replicates the cellular composition and microenvironment of the original tumor has been challenging, limiting its clinical applications. Here, we present a novel approach to generate glioma organoids with a microenvironment (GlioME) from patient-derived glioma tissue. These organoids maintain the genetic and epigenetic characteristics of the primary tumor and preserve cell-to-cell interactions within the tumor microenvironment, including resident immune cells. Bulk RNA sequencing, whole exome sequencing, and DNA methylation analysis were used to confirm the molecular similarities between the organoids and primary glioma tissues. Immunofluorescence and flow cytometry were used to assess immune cell viability, comparing GlioME with floating glioma organoids. GlioME exhibited high responsiveness to chemotherapy and targeted therapy, demonstrating its potential for therapeutic screening applications. Notably, GlioME accurately predicted patient response to the recently approved MET inhibitor, vebreltinib. Thus, this organoid model provides a reliable in vitro platform for glioma microenvironment-related research and clinical drug screening.

cancer biology↗

bioRxiv2024

E-box independent chromatin recruitment turns MYOD into a transcriptional repressor

MYOD is an E-box sequence-specific basic Helix-Loop-Helix (bHLH) transcriptional activator that, when expressed in non-muscle cells, induces nuclear reprogramming toward skeletal myogenesis by promoting chromatin accessibility at previously silent loci. Here, we report on the identification of a previously unrecognized property of MYOD as repressor of gene expression, via E-box-independent chromatin binding within accessible genomic elements, which invariably leads to reduced chromatin accessibility. MYOD-mediated repression requires the integrity of functional domains previously implicated in MYOD-mediated activation of gene expression. Repression of mitogen-and growth factor-responsive genes occurs through promoter binding and requires a highly conserved domain within the first helix. Repression of cell-of-origin/alternative lineage genes occurs via binding and decommissioning of distal regulatory elements, such as super-enhancers (SE), which requires the N-terminal activation domain as well as two chromatin-remodeling domains and leads to reduced strength of CTCF-mediated chromatin interactions. Surprisingly, MYOD-mediated chromatin compaction and repression of transcription do not associate with reduction of H3K27ac, the conventional histone mark of enhancer or promoter activation, but with reduced levels of the recently discovered histone H4 acetyl-methyl lysine modification (Kacme). These results extend MYOD biological properties beyond the current dogma that restricts MYOD function to a monotone transcriptional activator and reveal a previously unrecognized functional versatility arising from an alternative chromatin recruitment through E-box or non-E-box sequences. The E-box independent repression of gene expression by MYOD might provide a promiscuous mechanism to reduce chromatin accessibility and repress cell-of-origin/alternative lineage and growth factor/mitogen-responsive genes to safeguard the integrity of cell identity during muscle progenitor commitment toward the myogenic lineage.

molecular biology↗

bioRxiv2024

Genomic Analysis of Progenitors in Viral Infection Implicates Glucocorticoids as Suppressors of Plasmacytoid Dendritic Cell Generation

Plasmacytoid Dendritic cells (pDCs) are the most potent producers of interferons, which are critical antiviral cytokines. pDC development is, however, compromised following a viral infection, and this phenomenon, as well as its relationship to conventional (c)DC development is still incompletely understood. By using lymphocytic choriomeningitis virus (LCMV) infection in mice as a model system, we observed that DC progenitors skewed away from pDC and towards cDC development during in vivo viral infection. Subsequent characterization of the transcriptional and epigenetic landscape of fms-like tyrosine kinase 3+ (Flt3+) DC progenitors and follow-up studies revealed increased apoptosis and reduced proliferation in different individual DC-progenitors as well as a profound IFN-I-dependent ablation of pre-pDCs, but not pre-DC precursor, after both acute and chronic LCMV infections. In addition, integrated genomic analysis identified altered activity of 34 transcription factors in Flt3+ DC progenitors from infected mice, including two regulators of Glucocorticoid (GC) responses. Subsequent studies demonstrated that addition of GCs to DC progenitors led to downregulated pDC-primed-genes while upregulating cDC-primed-genes, and that endogenous GCs selectively decreased pDC, but not cDC, numbers upon in-vivo LCMV infection. These findings demonstrate a significant ablation of pre-pDCs in infected mice and identify GCs as suppressors of pDC generation from early progenitors. This provides an explanation for the impaired pDC development following viral infection and links pDC generation to the hypothalamic-pituitary-adrenal axis. Significance StatementPlasmacytoid dendritic cells (pDCs) play critical roles in antiviral responses. However, adaptations of DC progenitors lead to compromised pDC generation after viral infection. Here, we characterized the transcriptional and epigenetic landscapes of DC progenitors after infection. We observed widespread changes in gene expression and chromatin accessibility, reflecting shifts in proliferation, apoptosis, and differentiation potential into various DC subsets. Notably, we identified alterations in the predicted activity of 34 transcription factors, including two regulators of glucocorticoid responses. Our data demonstrate that glucocorticoids inhibit pDC generation by reprogramming DC progenitors. These findings establish a molecular framework for understanding how DC progenitors adapt to infection and highlight the role of glucocorticoid signaling in this process.

immunology↗

bioRxiv2024

Selective Agonism of Liver and Gut FXR Prevents Cholestasis and Intestinal Atrophy in Parenterally Fed Neonatal Pigs

BACKGROUND & AIMSWe aimed to investigate the relative efficacy of feeding different bile acids in preventing PNALD in neonatal pigs. METHODSNewborn pigs given total parenteral nutrition (TPN) combined with minimal enteral feeding of chenodeoxycholic acid (CDCA), or increasing doses of obeticholic acid (OCA) for 19 days. RESULTSEnteral OCA (5 and 15 mg/kg), but not CDCA (30 mg/kg) reduced blood cholestasis markers compared to TPN controls and increased bile acids in the gallbladder and intestine. Major bile acids in the liver and distal intestine were CDCA, HCA, HDCA and OCA, and their relative proportions were increased by the type of bile acid (CDCA or OCA) given enterally. High doses of OCA increased the total NR1H4-agonistic bile acid profile in the liver and intestine above 50% total bile acids. Both CDCA and OCA treatments suppressed hepatic cyp7a1 expression, but only OCA increased hepatobiliary transporters, ABCB11, ABCC$ and ABCB1. Plasma phytosterol levels were reduced and biliary levels were increased by CDCA and OCA and hepatic sterol transporters, abcg5/8, expression were increased by OCA. Both CDCA and OCA increased plasma FGF19 and OCA increased intestinal FGF19, FABP6, and SLC51A. Both CDCA and OCA increased intestinal mucosal growth, whereas CDCA increased the plasma GLP-2, GLP-1 and GIP. CONCLUSIONSEnteral OCA prevented cholestasis and phytosterolemia by increased hepatic bile acid and sterol transport via induction of hepatobiliary transporter FXR target genes and not by suppression of bile acid synthesis genes. We also showed an intestinal trophic action of OCA that demonstrates a dual clinical benefit of FXR agonism in the prevention of PNALD in piglets.

physiology↗

bioRxiv2024

Predicting the bacterial host range of plasmid genomes using the language model-based one-class SVM algorithm

The prediction of the plasmid host range is crucial for investigating the dissemination of plasmids and the transfer of resistance and virulence genes mediated by plasmids. Several machine learning-based tools have been developed to predict plasmid host ranges. These tools have been trained and tested based on the bacterial host records of plasmids in related databases. Typically, a plasmid genome in databases such as NCBI is annotated with only one or a few bacterial hosts, which does not encompass all possible hosts. Consequently, existing methods may significantly underestimate the host ranges of mobilizable plasmids. In this work, we propose a novel method named HRPredict, which employs a word vector model to digitally represent the encoded proteins on plasmid genomes. Since it is difficult to confirm which host a particular plasmid definitely cannot enter, we develop a machine learning approach for predicting whether a plasmid can enter a specific bacterium as a no negative samples learning task. Using multiple one-class SVMs that do not require negative samples for training, the HRPredict predicts the host range of plasmids across 45 families, 56 genera, and 56 species. In the benchmark test set, we constructed reliable negative samples for each host taxonomic unit via two indirect methods, and we found that the AUC, F1-score, recall, precision, and accuracy of most taxonomic unit prediction models exceeded 0.9. Among the 13 broad-host-range plasmid types, HRPredict demonstrated greater coverage than HOTSPOT and PlasmidHostFinder, thus successfully predicting the majority of hosts previously reported. Through the feature importance calculation for each SVM model, we found that genes closely related to the plasmid host range are involved in functions such as bacterial adaptability, pathogenicity, and survival. These findings provide significant insight into the mechanisms through which bacteria adjust to diverse environments through plasmids. Impact StatementPlasmids are important vectors for horizontal gene transfer and play a crucial role in regulating bacterial host adaptation to the environment. The spread of plasmid-mediated antibiotic resistance genes and virulence factors is one of the most important public health issues today. Owing to the lack of highly efficient methods for predicting the host range of newly discovered plasmids, especially broad-host-range plasmids, it is difficult to fully elucidate the regulatory role of plasmids in microbial communities and to predict the risk of antibiotic resistance transmission in clinical settings. Existing prediction tools tend to underestimate the host range of mobilizable plasmids. The current paper aims to overcome this limitation. Based on the concept of a "no negative samples learning task," we propose a new plasmid host range prediction method (i.e., HRPredict) that uses an SVM algorithm based on language models. HRPredict may be a powerful tool that will improve biologists understanding of horizontal plasmid transfer and help predict the occurrence and development of bacterial resistance. Data SummaryHRPredict is freely available via https://github.com/FengTaoSMU/HRPredict.

bioinformatics↗

bioRxiv2024

Orally surface engineered probiotic system for ulcerative colitis therapy via modulating gut microbiota and immune homeostasis

Oral probiotics have emerged as a promising therapeutical strategy for effectively managing ulcerative colitis (UC) in the world. The existing researches successfully preserve probiotic viability in the upper gastrointestinal tract, but they fall short in achieving precise release profile and effective colonization of probiotics at the site of colitis in colon as well as the understanding of therapeutic mechanism with suboptimal efficacy. This constraint poses a significant barrier to rational design and effective development of oral probiotic system. Here, we fabricate an orally layered-by-layered probiotic (Escherichia coli Nissle 1917, EcN) system (so-called CML@EcN) based on carboxymethyl modified lignin (CML) which can significantly protect EcN in gastric and small intestinal microenvironment and efficiently control release EcN in colon. Interestingly, the release and proliferation process of EcN in the colon are detected and can be modeled as a plug-flow mode. We derive a mathematical expression that highly matches the experimental results, providing a theoretical basis for quantitatively calculation of the release process. Furthermore, CML@EcN significantly alleviates symptoms in dextran sulfate sodium (DSS)-induced UC mice by modulating the multiple organ immune disorder and gut microbiota (GM)/metabolites profile as well as reconstructing the colon barrier. Importantly, we explore the mechanism of immune homeostasis regulation via metabolism from GM. Moreover, the CML armour is excreted and would not influence the process. This work not only reports a novel oral therapeutic for UC treatment with high efficacy as well as the exploration of mechanism between GM and immune homeostasis, but also provides a general approach to engineer probiotics for oral administration.

microbiology↗