BAPCP: A comprehensive and user-friendly web tool for identifying biomarkers from protein microarray technologies.

BACKGROUND AND OBJECTIVE: Proteome microarrays are one of the popular high-throughput screening methods for large-scale investigation of protein interactions in cells. These interactions can be measured on protein chips when coupled with fluorescence-labeled probes, helping indicate potential biomarkers or discover drugs. Several computational tools were developed to help analyze the protein chip results. However, existing tools fail to provide a user-friendly interface for biologists and present only one or two data analysis methods suitable for limited experimental designs, restricting the use cases. METHODS: In order to facilitate the biomarker examination using protein chips, we implemented a user-friendly and comprehensive web tool called BAPCP (Biomarker Analysis tool for Protein Chip Platforms) in this research to deal with diverse chip data distributions. RESULTS: BAPCP is well integrated with standard chip result files and includes 7 data normalization methods and 7 custom-designed quality control/differential analysis filters for biomarker extraction among experiment groups. Moreover, it can handle cost-efficient chip designs that repeat several blocks/samples within one single slide. Using experiments of the human coronavirus (HCoV) protein microarray and the E. coli proteome chip that helps study the immune response of Kawasaki disease as examples, we demonstrated that BAPCP can accelerate the time-consuming week-long manual biomarker identification process to merely 3 min. CONCLUSIONS: The developed BAPCP tool provides substantial analysis support for protein interaction studies and conforms to the necessity of expanding computer usage and exchanging information in bioscience and medicine. The web service of BAPCP is available at https://cosbi.ee.ncku.edu.tw/BAPCP/.

The generalized spring-loaded inverted pendulum model for analysis of various planar reduced-order models and for optimal robot leg design.

This paper proposes a generalized spring-loaded inverted pendulum (G-SLIP) model to explore various popular reduced-order dynamic models' characteristics and suggest a better robot leg design under specified performance indices. The G-SLIP model's composition can be varied by changing the model's parameters, such as ground contacting type and spring property. It can be transformed into four widely used models: the spring-loaded inverted pendulum (SLIP) model, the two-segment leg model, the SLIP with rolling foot model, and the rolling SLIP model. The effects of rolling contact and spring configuration on the dynamic behavior and fixed-point distribution of the G-SLIP model were analyzed, and the basins of attraction of the four described models were studied. By varying the parameters of the G-SLIP model, the dynamic behavior of the model can be optimized. Optimized for general locomotion running at various speeds, the model provided leg design guidelines. The leg was empirically fabricated and installed on the hexapod for experimental evaluation. The results indicated that the robot with a designed leg runs faster and is more power-efficient.

TNFα modulates PANX1 activation to promote ATP release and enhance P2RX7-mediated antitumor immune responses after chemotherapy in colorectal cancer.

ATP and its receptor P2RX7 exert a pivotal effect on antitumor immunity during chemotherapy-induced immunogenic cell death (ICD). Here, we demonstrated that TNFα-mediated PANX1 cleavage was essential for ATP release in response to chemotherapy in colorectal cancer (CRC). TNFα promoted PANX1 cleavage via a caspase 8/3-dependent pathway to enhance cancer cell immunogenicity, leading to dendritic cell maturation and T-cell activation. Blockade of the ATP receptor P2RX7 by the systemic administration of small molecules significantly attenuated the therapeutic efficacy of chemotherapy and decreased the infiltration of immune cells. In contrast, administration of an ATP mimic markedly increased the therapeutic efficacy of chemotherapy and enhanced the infiltration of immune cells in vivo. High PANX1 expression was positively correlated with the recruitment of DCs and T cells within the tumor microenvironment and was associated with favorable survival outcomes in CRC patients who received adjuvant chemotherapy. Furthermore, a loss-of-function P2RX7 mutation was associated with reduced infiltration of CD8+ immune cells and poor survival outcomes in patients. Taken together, these results reveal that TNFα-mediated PANX1 cleavage promotes ATP-P2RX7 signaling and is a key determinant of chemotherapy-induced antitumor immunity.

ASPM stabilizes the NOTCH intracellular domain 1 and promotes oncogenesis by blocking FBXW7 binding in hepatocellular carcinoma cells.

Notch signaling is aberrantly activated in approximately 30% of hepatocellular carcinoma (HCC), significantly contributing to tumorigenesis and disease progression. Expression of the major Notch receptor, NOTCH1, is upregulated in HCC cells and correlates with advanced disease stages, although the molecular mechanisms underlying its overexpression remain unclear. Here, we report that expression of the intracellular domain of NOTCH1 (NICD1) is upregulated in HCC cells due to antagonism between the E3-ubiquitin ligase F-box/WD repeat-containing protein 7 (FBXW7) and the large scaffold protein abnormal spindle-like microcephaly-associated protein (ASPM) isoform 1 (ASPM-i1). Mechanistically, FBXW7-mediated polyubiquitination and the subsequent proteasomal degradation of NICD1 are hampered by the interaction of NICD1 with ASPM-i1, thereby stabilizing NICD1 and rendering HCC cells responsive to stimulation by Notch ligands. Consistently, downregulating ASPM-i1 expression reduced the protein abundance of NICD1 but not its FBXW7-binding-deficient mutant. Reinforcing the oncogenic function of this regulatory module, the forced expression of NICD1 significantly restored the tumorigenic potential of ASPM-i1-deficient HCC cells. Echoing these findings, NICD1 was found to be strongly co-expressed with ASPM-i1 in cancer cells in human HCC tissues (P < 0.001). In conclusion, our study identifies a novel Notch signaling regulatory mechanism mediated by protein-protein interaction between NICD1, FBXW7, and ASPM-i1 in HCC cells, representing a targetable vulnerability in human HCC.

Rituximab, but not other biologics, impairs humoral immunity in patients with rheumatoid arthritis-a study using CoVariant protein arrays.

Objectives: RA is an autoimmune disease characterized by chronic inflammation and joint destruction. Biologics are crucial to achieving treat-to-target goals in patients with RA. The global spread and continuous variation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) necessitate the monitoring of variant-specific humoral responses post-vaccination. The aim of this study was to investigate how different biologic treatments for vaccinated RA patients might affect their neutralizing antibodies against multiple SARS-CoV-2 variants. Methods: We recruited RA patients who had received three doses of conventional SARS-CoV-2 vaccines and were treated with various biologics, e.g. TNF inhibitor (etanercept), IL-6 inhibitor (tocilizumab), CTLA4-Ig (abatacept) or anti-CD20 (rituximab). Serum samples were used to profile the binding and neutralizing antibodies using our own SARS-CoV-2 variant (CoVariant) protein array, developed previously. Results: Compared with healthy controls, only RA therapy with rituximab showed a reduction in neutralizing antibodies capable of targeting spike proteins in SARS-CoV-2 wild-type and most variants. This reduction was not observed in binding antibodies against SARS-CoV-2 wild-type or its variants. Conclusion: After receiving three doses of SARS-CoV-2 vaccination, RA patients who underwent rituximab treatment generated sufficient antibodies but exhibited lower neutralizing activities against wild-type and multiple variants, including current Omicron. Other biological DMARDs, e.g. TNF inhibitor, IL-6 inhibitor and CTLA4-Ig, did not show obvious inhibition.

Antibody Profiling of Dengue Severities Using Flavivirus Protein Microarrays.

Dengue is a viral disease transmitted by Aedes aegypti mosquitoes. According to the World Health Organization, about half of the world's population is at risk of dengue. There are four serotypes of the dengue virus. After infection with one serotype, it will be immune to such a serotype. However, subsequent infection with other serotypes will increase the risk of severe outcomes, e.g., dengue hemorrhagic fever, dengue shock syndrome, and even death. Since severe dengue is challenging to predict and lacks molecular markers, we aim to build a multiplexed Flavivirus protein microarray (Flaviarray) that includes all of the common Flaviviruses to profile the humoral immunity and cross-reactivity in the dengue patients with different outcomes. The Flaviarrays we fabricated contained 17 Flavivirus antigens with high reproducibility (R-square = 0.96) and low detection limits (172-214 pg). We collected serums from healthy subjects (n = 36) and dengue patients within 7 days after symptom onset (mild dengue (n = 21), hospitalized nonsevere dengue (n = 29), and severe dengue (n = 36)). After profiling the serum antibodies using Flaviarrays, we found that patients with severe dengue showed higher IgG levels against multiple Flavivirus antigens. With logistic regression, we found groups of markers with high performance in distinguishing dengue patients from healthy controls as well as hospitalized from mild cases (AUC > 0.9). We further reported some single markers that were suitable to separate dengue patients from healthy controls (AUC > 0.9) and hospitalized from mild outcomes (AUC > 0.8). Together, Flaviarray is a valuable tool to profile antibody specificities, uncover novel markers for decision-making, and shed some light on early preventions and treatments.

Developing magnetic barcode bead fluorescence assay for high throughput analyzing humoral responses against multiple SARS-CoV-2 variants.

The continuous mutation of SARS-CoV-2 highlights the need for rapid, cost-effective, and high-throughput detection methods. To better analyze the antibody levels against SARS-CoV-2 and its variants in vaccinated or infected subjects, we developed a multiplex detection named Barcode Bead Fluorescence (BBF) assay. These barcode beads were magnetic, characterized by 2-dimensional edges, highly multiplexed, and could be decrypted with visible light. We conjugated 12 magnetic barcode beads with corresponding nine spike proteins (wild-type, alpha, beta, gamma, delta, and current omicrons), two nucleocapsid proteins (wild-type and omicron), and one negative control. First, the conjugated beads underwent serial quality controls via fluorescence labeling, e.g., reproducibility (R square = 0.99) and detection limits (119 pg via anti-spike antibody). Next, we investigated serums from vaccinated subjects and COVID-19 patients for clinical applications. A significant reduction of antibody levels against all variant beads was observed in both vaccinated and COVID-19 studies. Subjects with two doses of mRNA-1273 exhibited the highest level of antibodies against all spike variants compared to two doses of AZD1222 and unvaccinated. We also found that COVID-19 patients showed higher antibody levels against spike beads from wild-type, alpha, beta, and delta. Finally, the nucleocapsid beads served as markers to distinguish infections from vaccinated subjects. Overall, this study developed the BBF assay for analyzing humoral immune responses, which has the advantages of robustness, automation, scalability, and cost-effectiveness.

Profiling humoral responses to COVID-19 immunization in Kawasaki disease using SARS-CoV-2 variant protein microarrays.

Kawasaki disease (KD) is a form of acute systemic vasculitis syndrome that predominantly occurs in children under the age of 5 years. Its etiology has been postulated due to not only genetic factors but also the presence of foreign antigens or infectious agents. To evaluate possible associations between Kawasaki disease (KD) and COVID-19, we investigated humoral responses of KD patients against S-protein variants with SARS-CoV-2 variant protein microarrays. In this study, plasma from a cohort of KD (N = 90) and non-KD control (non-KD) (N = 69) subjects in categories of unvaccinated-uninfected (pre-pandemic), SARS-CoV-2 infected (10-100 days after infection), and 1-dose, 2-dose, and 3-dose BNT162b2 vaccinated (10-100 days after vaccination) was collected. The principal outcomes were non-KD-KD differences for each category in terms of anti-human/anti-His for binding antibodies and neutralizing percentage for surrogate neutralizing antibodies. Binding antibodies against spikes were lower in the KD subjects with 1-dose of BNT162b2, and mean differences were significant for the P.1 S-protein (non-KD-KD, 3401; 95% CI, 289.0 to 6512; P = 0.0252), B.1.617.2 S-protein (non-KD-KD, 4652; 95% CI, 215.8 to 9087; P = 0.0351) and B.1.617.3 S-protein (non-KD-KD, 4874; 95% CI, 31.41 to 9716; P = 0.0477). Neutralizing antibodies against spikes were higher in the KD subjects with 1-dose of BNT162b2, and mean percentage differences were significant for the 1-dose BNT162b2 B.1.617.3 S-protein (non-KD-KD, -22.89%; 95% CI, -45.08 to -0.6965; P = 0.0399), B.1.1.529 S-protein (non-KD-KD, -25.96%; 95% CI, -50.53 to -1.376; P = 0.0333), BA.2.12.1 S-protein (non-KD-KD, -27.83%; 95% CI, -52.55 to -3.115; P = 0.0195), BA.4 S-protein (non-KD-KD, -28.47%; 95% CI, -53.59 to -3.342; P = 0.0184), and BA.5 S-protein (non-KD-KD, -30.42%; 95% CI, -54.98 to -5.869; P = 0.0077). In conclusion, we have found that KD patients have a comparable immunization response to healthy individuals to SARS-CoV-2 infection and COVID-19 immunization.

Molecular and Genetics-Based Systems for Tracing the Evolution and Exploring the Mechanisms of Human Norovirus Infections.

Human noroviruses (HuNoV) are major causes of acute gastroenteritis around the world. The high mutation rate and recombination potential of noroviruses are significant challenges in studying the genetic diversity and evolution pattern of novel strains. In this review, we describe recent advances in the development of technologies for not only the detection but also the analysis of complete genome sequences of noroviruses and the future prospects of detection methods for tracing the evolution and genetic diversity of human noroviruses. The mechanisms of HuNoV infection and the development of antiviral drugs have been hampered by failure to develop the infectious virus in a cell model. However, recent studies have demonstrated the potential of reverse genetics for the recovery and generation of infectious viral particles, suggesting the utility of this genetics-based system as an alternative for studying the mechanisms of viral infection, such as cell entry and replication.

The propagation of European airports' on-time performance and on-time flights via air connectivity prior to the Covid-19 pandemic.

This research investigates the number of on-time flights (OTFs) at European airports and how this number is influenced by an airport's flight connectivity. We conduct a spatial statistical analysis of the spatial context relationship using econometric models, and the interaction between the number of airport's on-time flights (OTFs) and flight connectivity. Using 2017 and 2018 data, we characterize the relationship between a European airport's air connectivity index (ACI) and the number of flights that depart or arrive at a gate within 15 min of schedule (OTFs). We also analyze the relationship between OTFs at a given airport and those of neighboring airports. As the distances between airports increase, autocorrelation shifts from a positive to a negative sign meaning that at greater distances, airports' on-time performance is less dissimilar. We find that before the pandemic and the ensuing global travel shutdown, a spatially lagged term of ACI improves the model's ability to account for variations in OTFs across airports. Flight delay propagation in the air transport system caused delays to occur due to the shared resources underlying an initially delayed flight and subsequent flights. This analysis offers a rational for increasing airport connectivity as a way of improving the share of on-time flights of European airports.

Dynamic turning and running of a hexapod robot using a separated and laterally arranged two-leg model.

We report on the development of separated and laterally arranged two-leg (SLTL) models with/without differentiated leg properties and their use as the dynamic running and turning templates for a hexapod robot. The laterally arranged two-leg morphology enables differential driving for turning. The differentiable leg settings, such as stiffness, enables the model to adopt unbalanced leg arrangements of empirical legged gaits, such as a tripod gait, into consideration. The fixed-point motion of the model was utilized as the main methodology to plan dynamic running and turning, in which the plot of one-step distance versus period was constructed for the legs' operation point selection and matching. The proposed methodology was experimentally validated using four indices: turning curvature, flight phase, motion stability, and energy efficiency. The experimental results show that the running robot using the SLTL model with differentiated leg stiffness has better energy efficiency than one without by 4%, while the latter model has identical performance to the original spring-loaded inverted pendulum model with rolling contact. As for turning, the robot using the SLTL models with/without differentiated leg stiffness can preserve dynamic turning in all experiments with turning curvatures up to0.28m-1and0.30m-1, respectively,33%and43%more than the robot using the original model-less phase-shift turning strategy (0.21 m-1). Using the proposed model-based strategy, the flight phase of the robot turning in all curvatures (including straight running) maintains around 20%, the root-mean-squared (RMS) values of pitch and roll remains less than3 deg, and the specific resistance (SR) is bounded between0.64 and 0.73. By contrast, the robot using the phase-shifting turning strategy can maintain dynamic motion up to a turning curvature of0.21 m-1. A further increase in phase shifting not only does not increase the turning curvature but also changes the robot motion from running to walking. In this case, no flight phase exists, theSRjumps up significantly, and RMS values of pitch and roll also increase dramatically. In short, the experimental validation confirms the effectiveness of the proposed methodology for initiating the dynamic running and turning of the robot.

TCM as adjunctive therapy improves risks of respiratory hospitalizations in persons with type 2 diabetes: A retrospective cohort study.

Patients with type 2 diabetes are at a higher risk of chronic obstructive pulmonary disease (COPD) and asthma than the general population. In addition, emerging evidence suggests that traditional Chinese medicine (TCM) might be beneficial for patients with type 2 diabetes. We investigated whether TCM use was associated with a reduced risk of respiratory hospitalizations in patients with type 2 diabetes. Conducting a retrospective cohort study, we used data retrieved from the NDCMP database. Among 56,035 patients, 5226 were classified as TCM users; 50,809 were classified as TCM nonusers. Both groups were analyzed until the end of 2011 to examine the incidence of respiratory hospitalizations by using a Cox proportional hazards model to evaluate effects of TCM use on respiratory hospitalizations. During the 6-year study follow-up period, the incidence density rates of COPD- and asthma-related hospitalization were estimated to be 13.03 and 4.47 per 10,000 patient-years for TCM nonusers and 10.08 and 3.28 per 10,000 patient-years for TCM users, respectively. The HR of COPD-related hospitalization in TCM users was 0.88 (95% CI = 0.79-0.99); and the HR of asthma-related hospitalization in TCM users was 0.81 (95% CI = 0.66-1.00). Stratified analyses revealed that effects of TCM use were stronger among individuals who had diabetes for <3 years. As a part of Integrative Medicine, our study results demonstrate that TCM use was associated with a significant reduced risk of respiratory hospitalizations, especially in patients with diabetes for <3 years.

A Novel Engineered AAV-Based Neoantigen Vaccine in Combination with Radiotherapy Eradicates Tumors.

The potency of tumor-specific antigen (TSA) vaccines, such as neoantigen (neoAg)-based cancer vaccines, can be compromised by host immune checkpoint inhibitory mechanisms, such as programmed cell death protein 1 (PD-1)/programmed death ligand 1 (PD-L1), that attenuate neoAg presentation on dendritic cells (DC) and hinder T cell-mediated cytotoxicity. To overcome PD-1/PD-L1 inhibition in DCs, we developed a novel adeno-associated virus (meAAV) neoAg vaccine, modified with TLR9 inhibitory fragments, PD-1 trap, and PD-L1 miRNA, which extend the persistence of meAAV and activate neoAg-specific T-cell responses in immune-competent colorectal and breast cancer murine models. Moreover, we found that in combination with radiotherapy, the meAAV-based neoAg cancer vaccine not only elicited higher antigen presentation ability, but also maintained neoAg-specific cytotoxic T lymphocyte (CTL) responses. These functional PD-1 traps and PD-L1 miRNAs overcome host PD-1/PD-L1 inhibitory mechanisms and boost the therapeutic efficacy of radiotherapy. More importantly, combined radiotherapy and meAAV neoAg cancer vaccines significantly enhanced neoAg-specific CTL responses, increased CTL infiltration in tumor microenvironment, and decreased tumor-associated immunosuppression. This process led to the complete elimination of colorectal cancer and delayed tumor growth of breast cancer in tumor-bearing mice. Taken together, our results demonstrated a novel strategy that combines neoAg cancer vaccine and radiotherapy to increase the therapeutic efficacy against colorectal and breast cancers.

Effects of colonization-associated gene yqiC on global transcriptome, cellular respiration, and oxidative stress in Salmonella Typhimurium.

BACKGROUND: yqiC is required for colonizing the Salmonella enterica serovar Typhimurium (S. Typhimurium) in human cells; however, how yqiC regulates nontyphoidal Salmonella (NTS) genes to influence bacteria-host interactions remains unclear. METHODS: The global transcriptomes of S. Typhimurium yqiC-deleted mutant (ΔyqiC) and its wild-type strain SL1344 after 2 h of in vitro infection with Caco-2 cells were obtained through RNA sequencing to conduct comparisons and identify major yqiC-regulated genes, particularly those involved in Salmonella pathogenicity islands (SPIs), ubiquinone and menaquinone biosynthesis, electron transportation chains (ETCs), and carbohydrate/energy metabolism. A Seahorse XFp Analyzer and assays of NADH/NAD+ and H2O2 were used to compare oxygen consumption and extracellular acidification, glycolysis parameters, adenosine triphosphate (ATP) generation, NADH/NAD+ ratios, and H2O2 production between ΔyqiC and SL1344. RESULTS: After S. Typhimurium interacts with Caco-2 cells, yqiC represses gene upregulation in aspartate carbamoyl transferase, type 1 fimbriae, and iron-sulfur assembly, and it is required for expressing ilvB operon, flagellin, tdcABCD, and dmsAB. Furthermore, yqiC is required for expressing mainly SPI-1 genes and specific SPI-4, SPI-5, and SPI-6 genes; however, it diversely regulates SPI-2 and SPI-3 gene expression. yqiC significantly contributes to menD expression in menaquinone biosynthesis. A Kyoto Encyclopedia of Genes and Genomes analysis revealed the extensive association of yqiC with carbohydrate and energy metabolism. yqiC contributes to ATP generation, and the analyzer results demonstrate that yqiC is required for maintaining cellular respiration and metabolic potential under energy stress and for achieving glycolysis, glycolytic capacity, and glycolytic reserve. yqiC is also required for expressing ndh, cydA, nuoE, and sdhB but suppresses cyoC upregulation in the ETC of aerobically and anaerobically grown S. Typhimurium; priming with Caco-2 cells caused a reversed regulation of yiqC toward upregulation in these ETC complex genes. Furthermore, yqiC is required for maintaining NADH/NAD+ redox status and H2O2 production. CONCLUSIONS: Specific unreported genes that were considerably regulated by the colonization-associated gene yqiC in NTS were identified, and the key role and tentative mechanisms of yqiC in the extensive modulation of virulence factors, SPIs, ubiquinone and menaquinone biosynthesis, ETCs, glycolysis, and oxidative stress were discovered.

Scenario-Based Distributionally Robust Unit Commitment Optimization Involving Cooperative Interaction with Robots.

With the increasing penetration of renewable energy, uncertainty has become the main challenge of power systems operation. Fortunately, system operators could deal with the uncertainty by adopting stochastic optimization (SO), robust optimization (RO) and distributionally robust optimization (DRO). However, choosing a good decision takes much experience, which can be difficult when system operators are inexperienced or there are staff shortages. In this paper, a decision-making approach containing robotic assistance is proposed. First, advanced clustering and reduction methods are used to obtain the scenarios of renewable generation, thus constructing a scenario-based ambiguity set of distributionally robust unit commitment (DR-UC). Second, a DR-UC model is built according to the above time-series ambiguity set, which is solved by a hybrid algorithm containing improved particle swarm optimization (IPSO) and mathematical solver. Third, the above model and solution algorithm are imported into robots that assist in decision making. Finally, the validity of this research is demonstrated by a series of experiments on two IEEE test systems.

HLJ1 amplifies endotoxin-induced sepsis severity by promoting IL-12 heterodimerization in macrophages.

Heat shock protein (HSP) 40 has emerged as a key factor in both innate and adaptive immunity, whereas the role of HLJ1, a molecular chaperone in HSP40 family, in modulating endotoxin-induced sepsis severity is still unclear. During lipopolysaccharide (LPS)-induced endotoxic shock, HLJ1 knockout mice shows reduced organ injury and IFN-γ (interferon-γ)-dependent mortality. Using single-cell RNA sequencing, we characterize mouse liver nonparenchymal cell populations under LPS stimulation, and show that HLJ1 deletion affected IFN-γ-related gene signatures in distinct immune cell clusters. In CLP models, HLJ1 deletion reduces IFN-γ expression and sepsis mortality rate when mice are treated with antibiotics. HLJ1 deficiency also leads to reduced serum levels of IL-12 in LPS-treated mice, contributing to dampened production of IFN-γ in natural killer cells but not CD4+ or CD8+ T cells, and subsequently to improved survival rate. Adoptive transfer of HLJ1-deleted macrophages into LPS-treated mice results in reduced IL-12 and IFN-γ levels and protects the mice from IFN-γ-dependent mortality. In the context of molecular mechanisms, HLJ1 is an LPS-inducible protein in macrophages and converts misfolded IL-12p35 homodimers to monomers, which maintains bioactive IL-12p70 heterodimerization and secretion. This study suggests HLJ1 causes IFN-γ-dependent septic lethality by promoting IL-12 heterodimerization, and targeting HLJ1 has therapeutic potential in inflammatory diseases involving activated IL-12/IFN-γ axis.

An Image-Based Data-Driven Model for Texture Inspection of Ground Workpieces.

Nowadays, the grinding process is mostly automatic, yet post-grinding quality inspection is mostly carried out manually. Although the conventional inspection technique may have cumbersome setup and tuning processes, the data-driven model, with its vision-based dataset, provides an opportunity to automate the inspection process. In this study, a convolutional neural network technique with transfer learning is proposed for three kinds of inspections based on 750-1000 surface raw images of the ground workpieces in each task: classifying the grit number of the abrasive belt that grinds the workpiece, estimating the surface roughness of the ground workpiece, and classifying the degree of wear of the abrasive belts. The results show that a deep convolutional neural network can recognize the texture on the abrasive surface images and that the classification model can achieve an accuracy of 0.9 or higher. In addition, the external coaxial white light was the most suitable light source among the three tested light sources: the external coaxial white light, the high-angle ring light, and the external coaxial red light. Finally, the model that classifies the degree of wear of the abrasive belts can also be utilized as the abrasive belt life estimator.