Biology subjects

Yu, H.

Publications and source records attributed to Yu, H..

At least 289 records · Page 16Linked to original sources

bioRxiv2022

Longitudinal Characterization of Cerebral Hemodynamicsin the TgF344-AD Rat Model of Alzheimer's Disease

Alzheimers Disease (AD) is a global healthcare crisis. The TgF344-AD rat is an AD model exhibiting age-dependent AD pathological hallmarks. We confirmed that AD rats developed cognitive deficits at 6 months without alteration of any other major biophysical parameters. We longitudinally characterized cerebral hemodynamics in AD rats at 3, 4, 6, and 14 months. The myogenic responses of the cerebral arteries and arterioles were impaired at 4 months of age in the AD rats. Consistent with the ex vivo results, the AD rat exhibited poor autoregulation of surface and deep cortical cerebral blood flow two months preceding cognitive decline. The dysfunction of cerebral hemodynamics in AD is exacerbated with age associated with reduced cerebral perfusion. Further, abolished cell contractility contributes to cerebral hemodynamics imbalance in AD. This may be attributed to enhanced ROS production, reduced mitochondrial respiration and ATP production, and disrupted actin cytoskeleton in cerebral vascular contractile cells.

neuroscience↗

bioRxiv2022

SnapFISH: a computational pipeline to identify chromatin loops from multiplexed DNA FISH data

Multiplexed DNA fluorescence in situ hybridization (FISH) imaging technologies have been developed to map the folding of chromatin fibers at tens of nanometer and tens of kilobase resolution in single cells. However, computational methods to reliably identify chromatin loops from such imaging datasets are still lacking. Here we present a Single-Nucleus Analysis Pipeline for multiplexed DNA FISH (SnapFISH), to process the multiplexed DNA FISH data and identify chromatin loops. SnapFISH can identify known chromatin loops from mouse embryonic stem cells with high sensitivity and accuracy. In addition, SnapFISH obtained comparable results of chromatin loops across datasets generated from diverse imaging technologies.

bioinformatics↗

bioRxiv2022

An important role for triglyceride in regulating spermatogenesis

Drosophila is a powerful model to study how lipids affect spermatogenesis. Yet, the contribution of neutral lipids, a major lipid group which resides in organelles called lipid droplets (LD), to sperm development is largely unknown. Emerging evidence suggests LD are present in the testis and that loss of neutral lipid- and LD-associated genes causes subfertility; however, key regulators of testis neutral lipids and LD remain unclear. Here, we show LD are present in early-stage somatic and germline cells within the Drosophila testis. We identified a role for triglyceride lipase brummer (bmm) in regulating testis LD, and found that whole-body loss of bmm leads to defects in sperm development. Importantly, these represent cell-autonomous roles for bmm in regulating testis LD and spermatogenesis. Because lipidomic analysis of bmm mutants revealed excess triglyceride accumulation, and spermatogenic defects in bmm mutants were rescued by genetically blocking triglyceride synthesis, our data suggest that bmm-mediated regulation of triglyceride influences sperm development. This identifies triglyceride as an important neutral lipid that contributes to Drosophila sperm development, and reveals a key role for bmm in regulating testis triglyceride levels during spermatogenesis.

developmental biology↗

bioRxiv2022

An FcRn-targeted mucosal vaccine against SARS-CoV-2 infection and transmission

SARS-CoV-2 and its variants cause COVID-19, which is primarily transmitted through droplets and airborne aerosols. To prevent viral infection and reduce viral spread, vaccine strategies must elicit protective immunity in the airways. FcRn transfers IgG across epithelial barriers; we explore FcRn-mediated respiratory delivery of SARS-CoV-2 spike (S). A monomeric IgG Fc was fused to a stabilized S protein; the resulting S-Fc bound to S-specific antibodies (Ab) and FcRn. A significant increase in Ab responses was observed following the intranasal immunization of mice with S-Fc formulated in CpG as compared to the immunization with S alone or PBS. Furthermore, we intranasally immunize adult or aged mice and hamsters with S-Fc. A significant reduction of virus replication in nasal turbinate, lung, and brain was observed following nasal challenges with SARS-CoV-2, including Delta and Omicron variants. Intranasal immunization also significantly reduced viral transmission between immunized and naive hamsters. Protection was mediated by nasal IgA, serum-neutralizing Abs, tissue-resident memory T cells, and bone marrow S-specific plasma cells. Hence FcRn delivers an S-Fc antigen effectively into the airway and induces protection against SARS-CoV-2 infection and transmission. Based on these findings, FcRn-targeted non-invasive respiratory immunizations are superior strategies for preventing highly contagious respiratory viruses from spreading.

immunology↗

bioRxiv2022

Pretrained language models and weight redistributionachieve precise kcat prediction

The enzyme turnover number (kcat) is a meaningful and valuable kinetic parameter, reflecting the catalytic efficiency of an enzyme to a specific substrate, which determines the global proteome allocation, metabolic fluxes and cell growth. Here, we present a precise kcat prediction model (PreKcat) leveraging pretrained language models and a weight redistribution strategy. PreKcat significantly outperforms the previous kcat prediction method in terms of various evaluation metrics. We also confirmed the ability of PreKcat to discriminate enzymes of different metabolic contexts and different types. Additionally, the proposed weight redistribution strategies effectively reduce the prediction error of high kcat values and capture minor effects of amino acid substitutions on two crucial enzymes of the naringenin synthetic pathway, leading to obvious distinctions. Overall, the presented kcat prediction model provides a valuable tool for deciphering the mechanisms of enzyme kinetics and enables novel insights into enzymology and biomedical applications.

biochemistry↗

bioRxiv2022

Physiological adaptation of sulfate reducing bacteria in syntrophic partnership with anaerobic methanotrophic archaea

Sulfate-coupled anaerobic oxidation of methane (AOM) is performed by multicellular consortia of anaerobic methanotrophic archaea (ANME) in obligate syntrophic partnership with sulfate-reducing bacteria (SRB). Diverse ANME and SRB clades co-associate but the physiological basis for their adaptation and diversification is not well understood. In this work, we explore the metabolic adaptation of four syntrophic SRB clades (HotSeep-1, Seep-SRB2, Seep-SRB1a and Seep-SRB1g) from a phylogenomics perspective, tracing the evolution of conserved proteins in the syntrophic SRB clades, and comparing the genomes of syntrophic SRB to their nearest evolutionary neighbors in the phylum Desulfobacterota. We note several examples of gain, loss or biochemical adaptation of proteins within pathways involved in extracellular electron transfer, electron transport chain, nutrient sharing, biofilm formation and cell adhesion. We demonstrate that the metabolic adaptations in each of these syntrophic clades are unique, suggesting that they have independently evolved, converging to a syntrophic partnership with ANME. Within the clades we also investigated the specialization of different syntrophic SRB species to partnerships with different ANME clades, using metagenomic sequences obtained from ANME and SRB partners in individual consortia after fluorescent-sorting of cell aggregates from anaerobic sediments. In one instance of metabolic adaptation to different partnerships, we show that Seep-SRB1a partners of ANME-2c appear to lack nutritional auxotrophies, while the related Seep-SRB1a partners of a different methanotrophic archaeal lineage, ANME-2a, are missing the cobalamin synthesis pathway, suggesting that the Seep-SRB1a partners of ANME-2a may have a nutritional dependence on its partner. Together, our paired genomic analysis of AOM consortia highlights the specific adaptation and diversification of syntrophic SRB clades linked to their associated ANME lineages.

microbiology↗

bioRxiv2022

Diversified repertoire of phage defenses in Klebsiella pneumoniae: Bi-directional steering effects impacting antibiotic susceptibility

Bacteriophage (phage) therapy is rising as a promising anti-infective option to combat antimicrobial resistance; however, its clinical utilization is severely hindered by the potential emergence of phage resistance. Fortunately, certain phage resistance mechanisms can restore bacterial antibiotic susceptibility, making the combination of phages with antibiotics a potential strategic approach. Here, we demonstrated that phage resistance can also lead to increased antibiotic resistance and provided mechanistic insights into bacterial phage defense mechanisms. We discovered a repertoire of phage resistance mechanisms in Klebsiella pneumoniae, including the disruption of phage binding site (fhuA::Tn and tonB::Tn), extension of phage latent period (mnmE::Tn and rpoN::Tn) and increased mutation frequency (mutS::Tn and mutL::Tn). Different from the prevailing view that phage resistance re-sensitizes antibiotic-resistant bacteria, we revealed a bidirectional steering effect on the bacterial antibiotic susceptibility. Specifically, it was uncovered that, while rpoN::Tn became more susceptible to colistin, mutS::Tn and mutL::Tn caused increased resistance to rifampicin and colistin. Our findings highlight the diversified strategies utilized by K. pneumoniae to overcome phage infection and the parallel effect on the antibiotic susceptibility. Mechanism-guided phage steering represents a rational strategy that should be incorporated into phage therapy to better inform clinical decisions.

microbiology↗

bioRxiv2022

Proliferative Exhausted CD8 T Cells Exacerbate Long-lasting Anti-tumor Effects in Human Papillomavirus Positive Head and Neck Squamous Cell Carcinoma

The survival prognosis of human papillomavirus (HPV)-positive and HPV-negative head and neck squamous cell carcinoma (HNSCC) is largely different, and little is known about the anti-tumor mechanism of tumor-infiltrated CD8+ exhausted T cells (Tex) in HNSCC. We performed cell level multi-omics sequencing on human HNSCC samples to decipher the multi-dimensional characteristics of Tex cells. A proliferative exhausted CD8+ T cell cluster (P-Tex) which was beneficial to survival outcomes of patients with HPV-positive HNSCC was identified. Interestingly, P-Tex cells expressed CDK4 genes as high as cancer cells, which could be simultaneously inhibited by CDK4 inhibitors and might be a potential reason for the ineffectiveness of CDK4 inhibitors in treating HPV-positive HNSCC. P-Tex cells could aggregate in the antigen presenting cell niches and activate certain signaling pathways. Together, our findings suggest a promising role for P-Tex cells in the prognosis of patients with HPV-positive HNSCC by providing modest but persistent anti-tumor effects. Graphic abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=105 SRC="FIGDIR/small/513993v1_ufig1.gif" ALT="Figure 1"> View larger version (39K): org.highwire.dtl.DTLVardef@1f8810forg.highwire.dtl.DTLVardef@1e11d84org.highwire.dtl.DTLVardef@58fb6aorg.highwire.dtl.DTLVardef@1e20a13_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2022

Mechanobiology of Fibroblast Activation in Skin Grafting

Mechanical stretching of living tissues can activate long-lived changes in tissue cells such as fibroblasts, increasing their contractility and initiating phenotypic transformations. Increased mechanical stimulus typically leads to monotonically increasing activation of fibroblasts cultured in 2D, but activation levels are difficult to predict for cells in 3D fibrous tissues, leading to variable outcomes in procedures such as skin grafting. Here we report that the source of this variation is cell-extracellular matrix (ECM) interactions and their variation with the duration and magnitude of applied stretch, and present a model that can predict the degree to which stretch will either increase or decrease long-term activation levels of fibroblasts cultured within a stretched, three-dimensional collagen matrix. Combining experimental and mathematical approaches across multiple scales, we show that the viscoplasticity of the ECM regulates this nonmonotonic, long-term cell activation. Results demonstrate that feedback between cell and ECM determines how cells retain memory of mechanical stretch.

biophysics↗

bioRxiv2022

Network-based hub biomarker discovery for glaucoma

Glaucoma is an optic neuropathy, and the leading cause of irreversible blindness worldwide. However, the early detection of glaucoma remains challenging as chronic forms of glaucoma remain largely asymptomatic until considerable irreversible visual field deficits have ensued. Thus, biomarkers that facilitate early diagnosis and treatment for patients with a high risk of progression are critical. Network medicine approaches can be useful in identifying key relationships and important biomolecules for complex diseases. In this paper, we identified several hub biomarkers/drug targets for the diagnosis, treatment and prognosis for glaucoma and explored their associations for glaucoma based on human disease-biomarker and disease-target-drug networks. These results were verified by text-mining and genomic/epidemiology data. We also predicted the new application of BMP1 and MMP9 to diagnose glaucoma and confirm the theory of hub biomarkers with multiple clinical applications. Further, relevant pivotal pathways (regulation of the multicellular organismal process, regulation of localisation, and cytoplasmic vesicle for biomarkers; signal transduction and developmental process for targets) for these hub biomolecules were discovered, which may be foundations for future biomarker and drug target prediction for glaucoma. In conclusion, based on complex networks, hub biomolecules, essential pathways, and close diseases were identified for glaucoma in diagnosis, treatment and prognosis.

bioinformatics↗

bioRxiv2022

Feasibility of Using an Armband Optical Heart Rate Sensor in Naturalistic Environment

Consumer-grade heart rate (HR) sensors including chest straps, wrist-worn watches and rings have become very popular in recent years for tracking individual physiological state, training for sports and even measuring stress levels and emotional changes. While the majority of these consumer sensors are not medical devices, they can still offer insights for consumers and researchers if used correctly taking into account their limitations. Multiple previous studies have been done using a large variety of consumer sensors including Polar(R) devices, Apple(R) watches, and Fitbit(R) wrist bands. The vast majority of prior studies have been done in laboratory settings where collecting data is relatively straight-forward. However, using consumer sensors in naturalistic settings that present significant challenges, including noise artefacts and missing data, has not been as extensively investigated. Additionally, the majority of prior studies focused on wrist-worn optical HR sensors. Arm-worn sensors have not been extensively investigated either. In the present study, we validate HR measurements obtained with an arm-worn optical sensor (Polar OH1) against those obtained with a chest-strap electrical sensor (Polar H10) from 16 participants over a 2-week study period in naturalistic settings. We also investigated the impact of physical activity measured with 3-D accelerometers embedded in the H10 chest strap and OH1 armband sensors on the agreement between the two sensors. Overall, we find that the arm-worn optical Polar OH1 sensor provides a good estimate of HR (Pearson r = 0.90, p <0.01). Filtering the signal that corresponds to physical activity further improves the HR estimates but only slightly (Pearson r = 0.91, p <0.01). Based on these preliminary findings, we conclude that the arm-worn Polar OH1 sensor provides usable HR measurements in daily living conditions, with some caveats discussed in the paper.

bioinformatics↗

bioRxiv2022

Scratch-AID: A Deep-learning Based System for Automatic Detection of Mouse Scratching Behavior with High Accuracy

Mice are the most commonly used model animals for itch research and for development of antiitch drugs. Most labs manually quantify mouse scratching behavior to assess itch intensity. This process is labor-intensive and limits large-scale genetic or drug screenings. In this study, we developed a new system, Scratch-AID (Automatic Itch Detection), which could automatically identify and quantify mouse scratching behavior with high accuracy. Our system included a custom-designed videotaping box to ensure high-quality and replicable mouse behavior recording and a convolutional recurrent neural network (CRNN) trained with frame-labeled mouse scratching behavior videos, induced by nape injection of chloroquine (CQ). The best trained network achieved 97.6% recall and 96.9% precision on previously unseen test videos. Remarkably, Scratch-AID could reliably identify scratching behavior in other major mouse itch models, including the acute cheek model, the histaminergic model, and a chronic itch model. Moreover, our system detected significant differences in scratching behavior between control and mice treated with an anti-itch drug. Taken together, we have established a novel deep learning-based system that is ready to replace manual quantification for mouse scratching behavior in different itch models and for drug screening.

neuroscience↗

bioRxiv2022

Validation of the linear regression method to evaluate population accuracy and bias of predictions for non-linear models

BackgroundThe linear regression method (LR) was proposed to estimate population bias and accuracy of predictions, while addressing the limitations of commonly used cross-validation methods. The validity and behavior of the LR method have been provided and studied for linear model predictions but not for non-linear models. The objectives of this study were to 1) provide a mathematical proof for the validity of the LR method when predictions are based on conditional mean, 2) explore the behavior of the LR method in estimating bias and accuracy of predictions when the model fitted is different from the true model, and 3) provide guidelines on how to appropriately partition the data into training and validation such that the LR method can identify presence of bias and accuracy in predictions. ResultsWe present a mathematical proof for the validity of the LR method to estimate bias and accuracy of predictions based on the conditional mean, including for non-linear models. Using simulated data, we show that the LR method can accurately detect bias and estimate accuracy of predictions when an incorrect model is fitted when the data is partitioned such that the values of relevant predictor variables differ in the training and validation sets. But the LR method fails when the data are not partitioned in that manner. ConclusionsThe LR method was proven to be a valid method to evaluate the population bias and accuracy of predictions based on the conditional mean, regardless of whether it is a linear or non-linear function of the data. The ability of the LR method to detect bias and estimate accuracy of predictions when the model fitted is incorrect depends on how the data are partitioned. To appropriately test the predictive ability of a model using the LR method, the values of the relevant predictor variables need to be different between the training and validation sets.

genetics↗

bioRxiv2022

DeepBIO is an automated and interpretable deep-learning platform for biological sequence prediction, functional annotation, and visualization analysis

Here, we present DeepBIO, the first-of-its-kind automated and interpretable deep-learning platform for high-throughput biological sequence functional analysis. DeepBIO is a one-stop-shop web service that enables researchers to develop new deep-learning architectures to answer any biological question. Specifically, given any biological sequence data, DeepBIO supports a total of 42 state-of-the-art deep-learning algorithms for model training, comparison, optimization, and evaluation in a fully automated pipeline. DeepBIO provides a comprehensive result visualization analysis for predictive models covering several aspects, such as model interpretability, feature analysis, functional sequential region discovery, etc. Additionally, DeepBIO supports 9 base-level functional annotation tasks using deep-learning architectures, with comprehensive interpretations and graphical visualizations to validate the reliability of annotated sites. Empowered by high-performance computers, DeepBIO allows ultra-fast prediction with up to million-scale sequence data in a few hours, demonstrating its usability in real application scenarios. Case study results show that DeepBIO provides an accurate, robust, and interpretable prediction, demonstrating the power of deep learning in biological sequence functional analysis. Overall, we expect DeepBIO to ensure the reproducibility of deep-learning biological sequence analysis, lessen the programming and hardware burden for biologists, and provide meaningful functional insights at both sequence-level and base-level from biological sequences alone. DeepBIO is publicly available at http://inner.wei-group.net/DeepBIO.

bioinformatics↗

bioRxiv2022

CTCF and R-loops are boundaries of cohesin-mediated DNA looping

Cohesin and CCCTC-binding factor (CTCF) are key regulatory proteins of three-dimensional (3D) genome organization. Cohesin extrudes DNA loops that are anchored by CTCF in a polar orientation. Here, we present direct evidence that CTCF binding polarity controls cohesin-mediated DNA looping. Using single-molecule imaging of CTCF-cohesin collisions, we demonstrate that a critical N-terminal motif of CTCF blocks cohesin translocation and DNA looping. The cryo-electron microscopy structure of the intact cohesin-CTCF complex reveals that this CTCF motif ahead of zinc-fingers can only reach its binding site on the STAG1 cohesin subunit when the N-terminus of CTCF faces cohesin. Remarkably, a C-terminally oriented CTCF accelerates DNA compaction by cohesin. DNA-bound Cas9 and Cas12a ribonucleoproteins are also polar cohesin barriers, indicating that stalling is intrinsic to cohesin itself, and other proteins can substitute for CTCF in fruit flies and other eukaryotes. Finally, we show that RNA-DNA hybrids (R-loops) block cohesin-mediated DNA compaction in vitro and are enriched with cohesin subunits in vivo, likely forming TAD boundaries. Our results provide direct evidence that CTCF orientation and R-loops shape the 3D genome by directly regulating cohesin.

molecular biology↗

bioRxiv2022

Controllable self-replicating RNA vaccine delivered intradermally elicits predominantly cellular immunity

Intradermal delivery of self-replicating RNA (srRNA) is a promising vaccine platform. Considering that human skin temperature is around 33{degrees}C, lower than core body temperature of 37{degrees}C, we have developed an srRNA that functions optimally at skin temperature and is inactivated at or above 37{degrees}C as a safety switch. This temperature-controllable srRNA (c-srRNA), when tested as an intradermal vaccine against SARS-CoV-2, functions when injected naked without lipid nanoparticles. Unlike most currently available vaccines, c-srRNA vaccines predominantly elicit cellular immunity with little or no antibody production. Interestingly, c-srRNA-vaccinated mice produced antigen-specific antibodies upon subsequent stimulation with antigen protein. Antigen-specific antibodies were also produced when B-cell stimulation using antigen protein was followed by c-srRNA booster vaccination. Using c-srRNA, we have designed a pan-coronavirus booster vaccine that incorporates both spike receptor binding domains as viral surface proteins and evolutionarily conserved nucleoproteins as viral non-surface proteins, from both SARS-CoV-2 and MERS-CoV. It can thereby potentially immunize against SARS-CoV-2, SARS-CoV, MERS-CoV, and their variants. c-srRNA may provide a route to activate cellular immunity against a wide variety of pathogens.

immunology↗

bioRxiv2022

Tumor removal limits prostate cancer cell dissemination in bone cortex and osteoblasts induce cancer cell dormancy

BackgroundDisseminated tumor cells (DTCs) can enter a dormant state and cause no symptoms in cancer patients. On the other hand, the dormant DTCs can reactivate and cause metastases progression and lethal relapses. In prostate cancer (PCa), relapse can happen after curative treatments such as primary tumor removal. The impact of surgical removal on PCa dissemination and dormancy remains elusive. Furthermore, as dormant DTCs are asymptomatic, dormancy-inducing can be an operational cure for preventing metastases and relapse of PCa patients. MethodsWe used a PCa subcutaneous xenograft model and species-specific PCR to survey the DTCs in various organs at different time points of tumor growth and in response to tumor removal. We developed in vitro 2D and 3D co-culture models to recapitulate the dormant DTCs in the bone microenvironment. Proliferation assays, fluorescent cell cycle reporter, qRT-PCR, and Western Blot were used to characterize the dormancy phenotype. We performed RNA sequencing to determine the dormancy signature of PCa. A drug repurposing algorithm was applied to predict dormancy-inducing drugs and a top candidate was validated for the efficacy and the mechanism of dormancy induction. ResultsWe found DTCs in almost all mouse organs examined, including bones, at week 2 post-tumor cell injections. Surgical removal of the primary tumor reduced the overall DTC abundance, but the DTCs were enriched only in the bones. We found that osteoblasts, but not other cells of the bones, induced PCa cell dormancy. RNA-Seq revealed the suppression of mitochondrial-related biological processes in osteoblast-induced dormant PCa cells. Importantly, the mitochondrial-related biological processes were found up-regulated in both circulating tumor cells and bone metastases from PCa patients data. We predicted and validated the dormancy-mimicking effect of PF-562271, an inhibitor of focal adhesion kinase (FAK) in vitro. Decreased FAK phosphorylation and increased nuclear translocation were found in both co-cultured and PF-271-treated C4-2B cells, suggesting that FAK plays a key role in osteoblast-induced PCa dormancy. ConclusionsOur study provides the first insights into how primary tumor removal enriches PCa cell dissemination in the bones, defines a unique osteoblast-induced PCa dormancy signature, and identifies FAK as a PCa cell dormancy gatekeeper.

cancer biology↗

bioRxiv2022

Artificial Intelligence based Liver Portal Tract Region Identification and Quantification with Transplant Biopsy Whole-Slide Images

Liver fibrosis staging is clinically important for liver disease progression prediction. As the portal tract fibrotic quantity and size in a liver biopsy correlate with the fibrosis stage, an accurate analysis of portal tract regions is clinically critical. Manual annotations of portal tract regions, however, are time-consuming and subject to large inter- and intra-observer variability. To address such a challenge, we develop a Multiple Up-sampling and Spatial Attention guided UNet model (MUSA-UNet) to segment liver portal tract regions in whole-slide images of liver tissue slides. To enhance the segmentation performance, we propose to use depth-wise separable convolution, the spatial attention mechanism, the residual connection, and multiple up-sampling paths in the developed model. This study includes 53 histopathology whole slide images from patients who received liver transplantation. In total, 6,012 patches derived from 30 images are used for our deep learning model training and validation. The remaining 23 whole slide images are utilized for the model testing. The average liver portal tract segmentation performance of the developed MUSA-UNet is 0.94 (Precision), 0.85 (Recall), 0.89 (F1 Score), 0.89 (Accuracy), 0.80 (Jaccard Index), and 0.91 (Fowlkes-Mallows Index), respectively. The clinical Scheuer fibrosis stage presents a strong correlation with the resulting average portal tract fibrotic area (R=0.681, p<0.001) and portal tract percentage (R=0.335, p=0.02) computed from the MUSA-UNet segmentation results. In conclusion, our developed deep learning model MUSA-UNet can accurately segment portal tract regions from whole-slide images of liver tissue biopsies, presenting its promising potential to assist liver disease diagnosis in a computational manner.

bioinformatics↗