Development of Rheumatoid Arthritis Classification from Electronic Image Sensor Using Ensemble Method

BackgroundSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused a global pandemic of severe coronavirus disease 2019 (COVID-19). Staphylococcus aureus is one of the most common pathogenic bacteria in humans, rheumatoid arthritis (RA) is among the most prevalent autoimmune conditions. RA is a significant risk factor for SARS-CoV-2 and S. aureus infections, although the mechanism of RA and SARS-CoV-2 infection in conjunction with S. aureus infection has not been elucidated. The purpose of this study is to investigate the biomarkers and disease targets between RA and SARS-CoV-2 and S. aureus infections using bioinformatics analysis, to search for the molecular mechanisms of SARS-CoV-2 and S. aureus immune escape and potential drug targets in the RA population, and to provide new directions for further analysis and targeted development of clinical treatments.MethodsThe RA dataset (GSE93272) and the S. aureus bacteremia (SAB) dataset (GSE33341) were used to obtain differentially expressed gene sets, respectively, and the common differentially expressed genes (DEGs) were determined through the intersection. Functional enrichment analysis utilizing GO, KEGG, and ClueGO methods. The PPI network was created utilizing the STRING database, and the top 10 hub genes were identified and further examined for functional enrichment using Metascape and GeneMANIA. The top 10 hub genes were intersected with the SARS-CoV-2 gene pool to identify five hub genes shared by RA, COVID-19, and SAB, and functional enrichment analysis was conducted using Metascape and GeneMANIA. Using the NetworkAnalyst platform, TF-hub gene and miRNA-hub gene networks were built for these five hub genes. The hub gene was verified utilizing GSE17755, GSE55235, and GSE13670, and its effectiveness was assessed utilizing ROC curves. CIBERSORT was applied to examine immune cell infiltration and the link between the hub gene and immune cells.ResultsA total of 199 DEGs were extracted from the GSE93272 and GSE33341 datasets. KEGG analysis of enrichment pathways were NLR signaling pathway, cell membrane DNA sensing pathway, oxidative phosphorylation, and viral infection. Positive/negative regulation of the immune system, regulation of the interferon-I (IFN-I; IFN-α/β) pathway, and associated pathways of the immunological response to viruses were enriched in GO and ClueGO analyses. PPI network and Cytoscape platform identified the top 10 hub genes: RSAD2, IFIT3, GBP1, RTP4, IFI44, OAS1, IFI44L, ISG15, HERC5, and IFIT5. The pathways are mainly enriched in response to viral and bacterial infection, IFN signaling, and 1,25-dihydroxy vitamin D3. IFI44, OAS1, IFI44L, ISG15, and HERC5 are the five hub genes shared by RA, COVID-19, and SAB. The pathways are primarily enriched for response to viral and bacterial infections. The TF-hub gene network and miRNA-hub gene network identified YY1 as a key TF and hsa-mir-1-3p and hsa-mir-146a-5p as two important miRNAs related to IFI44. IFI44 was identified as a hub gene by validating GSE17755, GSE55235, and GSE13670. Immune cell infiltration analysis showed a strong positive correlation between activated dendritic cells and IFI44 expression.ConclusionsIFI144 was discovered as a shared biomarker and disease target for RA, COVID-19, and SAB by this study. IFI44 negatively regulates the IFN signaling pathway to promote viral replication and bacterial proliferation and is an important molecular target for SARS-CoV-2 and S. aureus immune escape in RA. Dendritic cells play an important role in this process. 1,25-Dihydroxy vitamin D3 may be an important therapeutic agent in treating RA with SARS-CoV-2 and S. aureus infections.

Objectives This study aimed to investigate the potential role of synovial fibroblasts (SFs) in the development of rheumatoid arthritis (RA) to identify potential molecular targets and provide a theoretical basis for the treatment of RA. Methods GSE109449, a fibroblast transcriptome dataset of synovial tissue from RA and osteoarthritis (OA), were obtained from the GEO database. After standard cell quality control, this single-cell transcriptome data was used to perform routine single-cell analysis processes. After completing dimensionality reduction, clustering, and cell subset identification of fibroblasts, the SCENIC analysis helped calculate the significant gene regulatory networks in fibroblasts and their subsets. From these computed gene regulatory networks, the regulon in which follistatin-like protein 1 (FSTL1) resides was extracted and used to analyze the transcriptional regulatory status of fibroblasts. Finally, the gene set enrichment analysis (GSEA) was used to calculate the respective enriched gene sets of IRF1 and FSTL1. Results Three SF subgroups were identified from the single-cell transcriptome analysis; SF subset 3 was more abundant in RA than in OA (p < 0.001). From the SCENIC analysis, we obtained 269 regulons and the corresponding gene regulatory networks in SF from the RA datasets. Next, we screened and obtained a regulon-containing FSTL1, where IRF1 was the major transcription factor. The top five regulons in SF subset 3 were TWIST1, MECOM, KLF6, MAFB, and RUNX1. Among the 3 SF subsets, IRF1 regulon was ranked the highest in SF subset 3. Differential analysis of pseudobulk RNA-seq showed that IRF1 was up-regulated in RA compared to OA. Between the three SF subgroups, IRF1 and FSTL1 expression was more up-regulated in SF subset 3 compared to the other two subgroups. Conclusions IRF1 was found to regulate the invasiveness of SFs by regulating FSTL1, which may influence the disease progression of RA.

Invited Commentary

Rheumatoid Arthritis (RA) is a chronic autoimmune disease whose symptoms are hard to determine due to the overlapping indications of the condition with other illnesses such as dengue, malaria, etc. As the symptoms of RA disease are similar to inflammatory diseases, general physicians (GPs) find it difficult to detect the disease earlier. A computer aided framework is proposed in this study to assist and support the GPs to diagnose RA better. In this work Improved Harmony Search Optimization (IHSO) approach is proposed to select the significant feature subset of RA and Adaptive Neuro-Fuzzy Inference System (ANFIS) is used as a classification model. The performance of the proposed IHSO-ANFIS model is examined with metrics such as Balanced Accuracy (Bacc), Area under Curve (AUC), Sensitivity (Sen), Specificity (Spec), and Matthew’s Correlation Coefficient (MCC) using 10-Fold cross-validation. Additionally, the results of the IHSO-ANFIS are compared with HSO-ANFIS, ANFIS without any feature selection and standard bench mark datasets. IHSO-ANFIS attained 87.05% Bacc, 89.95% AUC and 0.6586 MCC on the RA dataset. From the results it is clear that IHSO-ANFIS could assist general physicians to diagnose RA earlier and pave the way for timely treatment.

Rheumatoid arthritis (RA) is an autoimmune disease characterized by chronic inflammation and joint destruction. Both inflammatory response and oxidative stress contribute to the pathogenesis of RA. Oxidative damage can induce and aggravate the imbalance of immune inflammation and promote cell and tissue damage. In this study, the expression of long non-coding RNA (lncRNA) LINC00638 in peripheral blood of patients with RA damp-heat arthralgia syndrome was observed, and the correlation between LINC00638 and disease activity, immune inflammation and oxidative stress indicator was investigated. Subsequently, the mechanisms for LINC00638 in regulating the inflammatory response and oxidative stress in RA fibroblast-like synoviocyte (FLS) under the condition of overexpression and interference were further explored. In this study, 48 RA patients with damp-heat arthralgia syndrome and 27 normal healthy subjects, who came from Department of Rheumatology, First Affiliated Hospital of Anhui University of Chinese Medicine, were included; and they were divided into a RA group and a control group. The expression of LINC00638 in peripheral blood mononuclear cells (PBMC) from the subjects was detected by real-time PCR. Enzyme linked immunosorbent assay (ELISA) was used to detect serum interleukin (IL)-10, IL-17, tumor necrosis factor-α (TNF-α), malondialdehyde (MDA), heme oxygenase 1 (HO-1) and superoxide dismutase 2 (SOD2) expression. Spearman method was used to study the relationship between LINC00638 and erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), rheumatoid factor (RF), anti-cyclic citrullinated peptide antibody (anti-CCP), and to observe the relation between LINC00638 and the Disease Activity Score of 28 Joint (DAS28), Quantitative Score of Damp Heat Syndrome, Visual Analogue Scale (VAS), Self-rating Anxiety Scale (SAS) and Self-rating Depression Scale (SDS). RA-FLS was induced by RA-PBMC, and the RA in vitro cell experimental model was established. LINC00638 overexpression plasmid and small interfering RNA (siRNA) were constructed and transfected into RA-FLS. The cell experiments were divided into 4 groups: a pcDNA3. 1- control group, a pcDNA3.1-LINC00638 group, a siRNA-control group, and a siRNA-LINC00638 group. The transfection efficiency of overexpression plasmid and siRNA was detected by real-time PCR, the expression of TNF-α and IL-10 was detected by ELISA, and the expression of antioxidant proteins HO-1 and SOD2 was detected by immunofluorescence. Compared with the control group, the expression of LINC00638 in the RA group was lower (P<0.01). The area under the curve (AUC) of the receiver operating characteristic (ROC) curve of LINC00638 was 0.9271. The DAS28 in RA group was 5.70 (5.40-6.50), the Quantitative Score of Damp-heat Syndrome was 20.0 (17.0-23.0), and the VAS score was 7.0 (6.3-8.0). Compared with the control group, the ESR, CRP, RF, anti-CCP, SAS and SDS scores in the RA group were significantly increased (all P<0.01). Spearman correlation analysis showed that: LINC00638 was negatively correlated with ESR (r=-0.532, P<0.01), CRP (r=-0.367, P<0.05), TNF-α (r=-0.375, P<0.01), MDA (r= -0.295, P<0.05), DAS28 (r=-0.450, P<0.01), and which was positively correlated with SOD2 (r=0.370, P<0.05). After the induction of RA-FLS, the expression level of LINC00638 was significantly decreased (P<0.01), indicating that the stimulation of PBMC could effectively reduce the expression of LINC00638 in RA-FLS, so the experimental model of RA-FLS-induced by PBMC was utilized. Compared with the pcDNA3.1-control group, the expressions of LINC00638, IL-10, SOD2, and HO-1 in the pcDNA3.1-LINC00638 group were significantly increased (all P<0.01), and the expression of TNF-α was decreased (P<0.01). Compared with siRNA-control group, LINC00638, IL-10, SOD2 and HO-1 in the siRNA-LINC00638 group were significantly decreased (all P<0.01), and TNF-α was significantly increased (P<0.01). LINC00638 is down-regulated in the peripheral blood of RA patients with damp-heat arthralgia syndrome, which is correlated with disease activity, immune inflammation and oxidative stress. Overexpression of LINC00638 can down-regulate pro-inflammatory factors, up-regulate anti-inflammatory factors, and increase antioxidant enzyme activity, thereby improving inflammation and oxidative stress in RA. LINC00638 is the differential lncRNA obtained by the research group's previous high-throughput sequencing of the whole transcriptome of peripheral blood PBMCs in RA patients and validation of clinical samples. In order to deepen the molecular biology research of this gene, the microRNA and mRNA targeted by LINC00638 can be further studied from the perspective of competing endogenous RNAs.

The popularity of ZigBee devices continues to grow in home automation, transportation, traffic management, and Industrial Control System (ICS) applications given their low-cost and low-power. However, the decentralized architecture of ZigBee ad-hoc networks creates unique security challenges for network intrusion detection and prevention. In the past, ZigBee device authentication reliability was enhanced by Radio Frequency-Distinct Native Attribute (RF-DNA) fingerprinting using a Fisher-based Multiple Discriminant Analysis and Maximum Likelihood (MDA-ML) classification process to distinguish between devices in low Signal-to-Noise Ratio (SNR) environments. However, MDA-ML performance inherently degrades when RF-DNA features do not satisfy Gaussian normality conditions, which often occurs in real-world scenarios where radio frequency (RF) multipath and interference from other devices is present. We introduce non-parametric Random Forest (RndF) and Multi-Class AdaBoost (MCA) ensemble classifiers into the RF-DNA fingerprinting arena, and demonstrate improved ZigBee device authentication. Results are compared with parametric MDA-ML and Generalized Relevance Learning Vector Quantization-Improved (GRLVQI) classifier results using identical input feature sets. Fingerprint dimensional reduction is examined using three methods, namely a pre-classification Kolmogorov-Smirnoff Test (KS-Test), a post-classification RndF feature relevance ranking, and a GRLVQI feature relevance ranking. Using the ensemble methods, an SNR=18.0 dB improvement over MDA-ML processing is realized at an arbitrary correct classification rate (%C) benchmark of %C=90%; for all SNR ∈ [0, 30] dB considered, %C improvement over MDA-ML ranged from 9% to 24%. Relative to GRLVQI processing, ensemble methods again provided improvement for all SNR, with a best improvement of %C=10% achieved at the lowest tested SNR=0.0 dB. Network penetration, measured using rogue ZigBee devices, show that at the SNR=12.0 dB (%C=90%) the ensemble methods correctly reject 31 of 36 rogue access attempts based on Receiver Operating Characteristic (ROC) curve analysis and an arbitrary Rogue Accept Rate of . This performance is better than MDA-ML, and GRLVQI which rejected 25/36, and 28/36 rogue access attempts respectively. The key benefit of ensemble method processing is improved rogue rejection in noisier environments; gains of 6.0 dB, and 18.0 dB are realized over GRLVQI, and MDA-ML, respectively. Collectively considering the demonstrated %C and rogue rejection capability, the use of ensemble methods improves ZigBee network authentication, and enhances anti-spoofing protection afforded by RF-DNA fingerprinting.

Exploring biological meaning from microarray data is very important but remains a great challenge. Here, we developed three new statistics: linear combination test, quadratic test and de-correlation test to identify differentially expressed pathways from gene expression profile. We apply our statistics to two rheumatoid arthritis datasets. Notably, our results reveal three significant pathways and 275 genes in common in two datasets. The pathways we found are meaningful to uncover the disease mechanisms of rheumatoid arthritis, which implies that our statistics are a powerful tool in functional analysis of gene expression data.Keywordsmicroarraypathwaylinear combination testquadratic testde-correlation testrheumatoid arthritis

Reviewing ensemble classification methods in breast cancer

BackgroudTaking the advan tage of high-throughput single nucleotide polymorphism (SNP) genotyping technology, large genome-wide association studies (GWASs) have been considered to hold promise for unravelling complex relationships between genotype and phenotype. At present, traditional single-locus-based methods are insufficient to detect interactions consisting of multiple-locus, which are broadly existing in complex traits. In addition, statistic tests for high order epistatic interactions with more than 2 SNPs propose computational and analytical challenges because the computation increases exponentially as the cardinality of SNPs combinations gets larger.ResultsIn this paper, we provide a simple, fast and powerful method using dynamic clustering and cloud computing to detect genome-wide multi-locus epistatic interactions. We have constructed systematic experiments to compare powers performance against some recently proposed algorithms, including TEAM, SNPRuler, EDCF and BOOST. Furthermore, we have applied our method on two real GWAS datasets, Age-related macular degeneration (AMD) and Rheumatoid arthritis (RA) datasets, where we find some novel potential disease-related genetic factors which are not shown up in detections of 2-loci epistatic interactions.ConclusionsExperimental results on simulated data demonstrate that our method is more powerful than some recently proposed methods on both two- and three-locus disease models. Our method has discovered many novel high-order associations that are significantly enriched in cases from two real GWAS datasets. Moreover, the running time of the cloud implementation for our method on AMD dataset and RA dataset are roughly 2 hours and 50 hours on a cluster with forty small virtual machines for detecting two-locus interactions, respectively. Therefore, we believe that our method is suitable and effective for the full-scale analysis of multiple-locus epistatic interactions in GWAS.

Abnormal glucose metabolism brings out joint inflammation and destruction in rheumatoid arthritis (RA). The aim of this study was to evaluate the potential of circulating hexokinase-2 (HK2) in peripheral blood mononuclear cells (PBMCs) of rheumatoid arthritis (RA) patients.PBMCs were obtained from patients with RA or osteoarthritis (OA) and healthy controls (HCs). The expression of HK2 was assessed by quantitative reverse transcription-polymerase chain reaction (qRT-PCR). The C-reactive protein (CRP) level, erythrocyte sedimentation rate (ESR), Calprotectin, rheumatoid factor (RF), anti-cyclic citrullinated peptides (anti-CCP) antibody level and 28-joint Disease Activity Score (DAS28), Clinical Disease Activity Index (CDAI) and Simplified Disease Activity Index (SDAI) were measured. Spearman's analysis was performed to determine the association between the level of HK2 and clinical characteristics. A receiver operating characteristic (ROC) curve was employed to evaluate the diagnostic value of HK2 in PBMCs. Logistic regression was used to identify risk factors. Sixty-five RA patients, 35 OA patients, and 40 HCs were included in the study.HK2 was upregulated in RA and OA patients compared with that in HCs (P < .05). The area under the ROC of HK2 for diagnosing RA and OA was 0.808 and 0.640, respectively. In addition, HK2 levels were increased in active RA compared with those in remittent RA (P = .03). Furthermore, HK2 correlated positively with the DAS28-ESR (P < .001), CDAI (P = .02) and SDAI scores (P = .02). Moreover, HK2 was independently associated with an increased risk of disease activity (DAS28-ESR>3.2, P = .02; CDAI score>10, P = .03; SDAI score>11, P = .04). Additionally, HK2 positivity was more frequently detected in patients treated with biologic disease-modifying antirheumatic drugs (bDMARDs) than in those not treated with bDMARDs.HK2 levels in PBMCs can be considered an ideal biomarker for diagnosing RA and involved in disease activity in RA. Dysregulation of HK2 may participate in the molecular mechanism of RA and could be an attractive selective metabolic target for RA treatment.

Accurate estimation of forest aboveground biomass is essential for the assessment of regional carbon cycle and the climate change in the terrestrial ecosystem. Currently, ensemble learning algorithms and cross-validation methods have been widely applied to estimate regional forest Above Ground Biomass (AGB). However, the effects of ensemble learning algorithms, validation methods, and their interactions on forest AGB estimation were rarely investigated. Based on Landsat 8 Operational Land Imager (OLI) imagery, Airborne Laser Scanning (ALS) data and China’s National Forest Continuous Inventory data, this study explored the effects of five ensemble learning algorithms, including Simple Averaging (SA), Weighted Averaging (WA), Stacked Generalization (SG), Random Forest (RF) and Extreme Gradient Boosting (XGBoost), and two validation approaches (i.e. 10-fold and leave-one-out cross-validation) on the AGB estimation of the Natural Secondary Forests (NSFs) in northeast China. The results revealed that the ensemble learning algorithms that combine heterogenous-based models (i.e. SA, WA, SG) generally produced higher accuracy than the base models (i.e. Decision Tree (DT), K-Nearest Neighbor (KNN), Support Vector Regression (SVR), Convolutional Neural Network (CNN)). Among all ensemble learning algorithms, the SG algorithm has the highest accuracy whereas the XGBoost algorithm has the lowest accuracy. Although prediction models considerably impact the accuracy of AGB estimation, the validation approach also plays a non-negligible role in AGB estimation. The leave-one-out cross-validation produced much higher accuracy than the 10-fold cross-validation using the same prediction model and tends to generate over-optimistic AGB estimates compared to 10-fold cross-validation, especially for the averaging and stacking ensemble learning algorithms (i.e. SA, WA, SG). This study highlights the potential challenges of applying a leave-one-out cross-validation approach and provides a scientific foundation for the feasibility of different ensemble learning algorithms and cross-validation approaches for accurate AGB estimation.

IntroductionThe role of adiponectin in the pathogenesis of arthritis is still controversial. This study was performed to examine whether adiponectin is involved in joint inflammation and destruction in rheumatoid arthritis (RA) in relation to the expression of vascular endothelial growth factor (VEGF) and matrix metalloproteinases (MMPs).MethodsSynovial cells from RA patients were treated with adiponectin or interleukin (IL)-1β for 24 hours. The culture supernatant was collected and analyzed for the levels of IL-6, IL-8, prostaglandin E2 (PGE2), VEGF, and MMPs by enzyme-linked immunosorbent assay. The levels of adiponectin, VEGF, MMP-1, and MMP-13 in the joint fluids from 30 RA or osteoarthritis (OA) patients were also measured.ResultsAdiponectin at the concentration of 10 μg/mL stimulated the production of IL-6, IL-8, and PGE2 in RA fibroblast-like synoviocytes (FLSs), although the level of these was much lower than with 1 ng/mL IL-1β. However, adiponectin stimulated the production of VEGF, MMP-1, and MMP-13 at the same level as IL-1β. In addition, the level of adiponectin and MMP-1 in the joint fluid of RA patients was significantly higher than in OA patients. Adiponectin was positively correlated with VEGF in RA patients but not in OA patients, while the level of MMPs in joint fluid was not correlated with adiponectin in either RA or OA patients.ConclusionsAdiponectin may play a significant role in the pathogenesis of RA by stimulating the production of VEGF and MMPs in FLSs, leading to joint inflammation and destruction, respectively.

An ensemble classifier is a method that combines output of multiple classifiers. It has been widely accepted that ensemble classifiers can improve the prediction accuracy. Recently, ensemble techniques have been successfully applied to the bankruptcy prediction. Bagging and random subspace are the most popular ensemble techniques. Bagging and random subspace have proved to be very effective in improving the generalization ability respectively. However, there are few studies which have focused on the integration of bagging and random subspace. In this study, we proposed a new hybrid ensemble model to integrate bagging and random subspace method using genetic algorithm for improving the performance of the model. The proposed model is applied to the bankruptcy prediction for Korean companies and compared with other models in this study. The experimental results showed that the proposed model performs better than the other models such as the single classifier, the original ensemble model and the simple hybrid model. Keywords：Random Subspace, Bagging, Bankruptcy Prediction, Genetic Algorithms

In this study the methodology of analyzing the data produced from Genome Wide Association Studies (GWAs) is presented, using appropriate techniques and software packages. The analytical process is applied to Rheumatoid Arthritis (RA) and Multiple Sclerosis (MS) data sets produced by experimental microarray assays. The purpose of the process aims to the filtering of the datasets in order to derive and define statistically significant genetic variants associated with each one of the diseases. Furthermore, the analysis is expanded using the Multifactor Dimensionality Reduction (MDR) data mining strategy. Using MDR we are reducing the dimensionality of the genotyped predictive factors from n-dimensions to one-dimension by characterizing the genotypes of the different locus as high-risk and low-risk. In this way, we can provide risk predictive models about the diseases' susceptibility and the methodology is further extended by adding clinical and environmental factors to the RA and MS genomic loci data. The methodology presented in this study aims to transform the analytical genomic datasets produced by microarray experiments to a knowledge modeling framework constituted of prognostic/diagnostic decision rules and risk predictive factors. Finally, the evaluation methodology of the MDR data mining strategy is presented.

The subject of the research is methods and means of identifying the state of a computer system . The purpose of the article is to improve the quality of computer system state identification by developing a method based on ensemble classifiers. Task: to investigate methods for constructing bagging classifiers based on decision trees, to configure them and develop a method for identifying the state of the computer system. Methods used: artificial intelligence methods, machine learning, ensemble methods. The following results were obtained: the use of bagging classifiers based on meta-algorithms were investigated: Pasting Ensemble, Bootstrap Ensemble, Random Subspace Ensemble, Random Patches Ensemble and Random Forest methods and their accuracy were assessed to identify the state of the computer system. The research of tuning parameters of individual decision trees was carried out and their optimal values were found, including: the maximum number of features used in the construction of the tree; the minimum number of branches when building a tree; minimum number of leaves and maximum tree depth. The optimal number of trees in the ensemble has been determined. A method for identifying the state of the computer system is proposed, which differs from the known ones by the choice of the classification meta-algorithm and the selection of the optimal parameters for its adjustment. An assessment of the accuracy of the developed method for identifying the state of a computer system is carried out. The developed method is implemented in software and investigated when solving the problem of identifying the abnormal state of the computer system functioning. Conclusions. The scientific novelty of the results obtained lies in the development of a method for identifying the state of the computer system by choosing a meta-algorithm for classification and determining the optimal parameters for its configuration.