Identification of Lipid Metabolism-Related Genes in Rheumatoid Arthritis Using Bioinformatics and Machine Learning

Identification of Lipid Metabolism-Related Genes in Rheumatoid Arthritis Using Bioinformatics and Machine Learning | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Identification of Lipid Metabolism-Related Genes in Rheumatoid Arthritis Using Bioinformatics and Machine Learning Zhi Feng, Xiangping Luo, Linyue Tang, Wei Long, Lvheng He, Wei Wang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7913208/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Rheumatoid arthritis (RA) is a common autoimmune inflammatory joint disease. Recent studies suggest that lipid metabolism dysregulation plays a crucial role in RA pathogenesis; however, its precise mechanisms remain unclear. This study aims to identify lipid metabolism-related diagnostic biomarkers in RA and analyze their potential pathogenic mechanisms and clinical significance. Methods Multiple RA-related microarray datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were obtained from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified using the “limma” R package. Weighted gene co-expression network analysis (WGCNA) was performed to identify key module genes associated with RA. Two machine learning algorithms, least absolute shrinkage and selection operator (LASSO) and random forest (RF), were used to screen hub genes closely related to synovial lipid metabolism in RA. A nomogram and receiver operating characteristic (ROC) curve were constructed to predict RA risk. Additionally, immune cell infiltration was analyzed, followed by single-sample gene set enrichment analysis (ssGSEA) and validation in an independent dataset. Results PIK3CD was identified as a key gene with strong diagnostic value for RA. Abnormal immune cell infiltration was observed in RA patients and was positively correlated with PIK3CD expression. Conclusion These findings suggest that PIK3CD is involved in RA-related lipid metabolism and may serve as a reliable diagnostic biomarker and potential therapeutic target for RA. Rheumatoid Arthritis Lipid metabolism Machine learning Immune infiltration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Highlights Our study integrates multiple datasets to enhance the robustness of our findings The combination of WGCNA and machine learning algorithms allowed us to identify key genes with high precision. The construction of a nomogram and ROC curve provides a practical tool for predicting RA risk. Our analysis of immune cell infiltration and validation in an independent dataset strengthens the clinical relevance of our findings.. 1 Introduction Rheumatoid arthritis (RA) is a common chronic inflammatory autoimmune disease characterized by symmetric joint inflammation, leading to pain, swelling, dysfunction, and, in severe cases, joint deformity and loss of function[ 1 ]. The global prevalence of RA is estimated to be 0.5%–1%, and its chronic progressive and disabling nature significantly reduces patients’ quality of life while imposing a substantial burden on healthcare systems[ 2 ]. Current therapeutic strategies primarily include nonsteroidal anti-inflammatory drugs (NSAIDs), disease-modifying antirheumatic drugs (DMARDs), and glucocorticoids. Although these treatments can alleviate symptoms to some extent, they fail to halt disease progression[ 3 – 5 ]. Therefore, a deeper understanding of RA pathogenesis and the identification of novel therapeutic targets are critical for improving patient outcomes and quality of life. Lipids, including fatty acids, triglycerides, cholesterol, phospholipids, and sphingolipids, serve not only as energy storage molecules but also as key mediators of signal transduction and inflammation [ 6 , 7 ]. Emerging evidence suggests that lipid metabolism dysregulation is closely associated with various inflammatory diseases, such as atherosclerosis and osteoarthritis[ 8 – 11 ]. Recent studies have indicated that lipid reprogramming within the synovial microenvironment of RA patients drives joint destruction by regulating inflammatory cytokine secretion and oxidative stress[ 12 ]. The lipid composition of synovial fluid is strongly correlated with inflammation severity. For instance, fatty acids promote fibroblast-like synoviocytes (FLSs) secretion of pro-inflammatory cytokines and matrix metalloproteinases (e.g., IL-6, IL-8, MMP-1, and MMP-3)[ 13 , 14 ]. Additionally, the arachidonic acid metabolite 15s-HETE enhances MMP-2 expression in FLS via its downstream enzyme 15-LOX, while another metabolite, LTB4, binds to LTB4 receptor 2 on FLS, significantly upregulating TNF-α and IL-1β production[ 15 – 17 ]. Increasing evidence supports the critical role of lipid metabolism in RA pathogenesis; however, the precise mechanisms by which lipid metabolism mediates RA pathophysiology remain unclear. To address this issue, RA-related public datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were retrieved from the Gene Expression Omnibus (GEO) database. WGCNA was performed to identify RA-associated genes, while lipid metabolism-related gene sets were obtained from the MigSB database. Machine learning algorithms were then applied to identify hub genes involved in lipid metabolism in RA. Differentially expressed lipid metabolism-related genes (LMRGs) were further analyzed through enrichment analysis and ROC curve evaluation to identify potential diagnostic biomarkers. A nomogram predictive model based on diagnostic genes was subsequently constructed and validated to support clinical decision-making. Additionally, single-sample gene set enrichment analysis (ssGSEA) was conducted to explore the relationship between diagnostic genes and immune cell infiltration. Finally, the findings were validated in an independent dataset. This study aims to identify potential LMRGs with diagnostic value for RA, providing new therapeutic insights for RA management. 2 Materials and methods 2.1 Data acquisition and processing Five datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were retrieved from the GEO database ( https://www.ncbi.nlm.nih.gov/geo/ ). Among them, GSE206848 and GSE77298 were used as the training set, while GSE55235, GSE55584, and GSE55235 served as the validation set. To ensure analytical accuracy, batch effect removal was performed before merging the datasets, and datasets with substantial batch differences were excluded to maintain data consistency and compatibility. The “sva” R package was used to integrate and normalize the datasets. Additionally, lipid metabolism-related gene sets were obtained from the MigSB database ( https://www.gsea-msigdb.org/gsea/index.jsp ). 2.2 Differentially expressed genes To identify DEGs between RA and healthy samples, the “limma” R package was used. The selection criteria were set to P 1.5. The “ggplot2” R package was then employed to generate a volcano plot for visualization. 2.3 WGCNA and module Identification Data from GSE206848 (control = 7, RA = 2) and GSE77298 (control = 7, RA = 16) are merged and batch-corrected. WGCNA is performed to identify trait-associated modules. A topological overlap matrix is constructed based on expression profiles. The soft-threshold power is set to 5, and the minimum module size is set to 50 to filter core modules. A height cutoff of 0.25 is applied as a guideline for module merging. Pearson correlation analysis is then conducted to assess module significance, with a significance threshold of P < 0.05. 2.4 Functional enrichment analyses Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses are performed using the “clusterProfiler” R package. GO annotation includes biological process (BP), cellular component (CC), and molecular function (MF) categories. A P-value threshold of 0.05 is applied for filtering, with P < 0.05 considered statistically significant. The enrichment analysis results are visualized using bubble plots. 2.5 Machine learning‑based hub gene screening Machine learning is an artificial intelligence technique that identifies patterns from data using models and algorithms, optimizing predictions and improving performance. In the medical field, machine learning models have widespread applications. In this study, DEGs identified from differential analysis are intersected with RA-associated gene modules obtained from WGCNA to generate a set of candidate genes. These candidate genes are then further intersected with lipid metabolism-related gene sets extracted from the MigSB database to identify key genes. For the key genes obtained from this intersection, the “glmnet” R package in R is used to perform LASSO-Cox regression, integrating variables and conducting survival analysis to derive the first gene module. Simultaneously, the randomForest function is employed to construct a random forest model, yielding the second gene module. Finally, the VennDiagram package in R is used to intersect the two modules, identifying four hub genes. 2.6 Screening of hub genes for its’diagnostic ability For the four hub genes identified through machine learning algorithms, the “pROC” R package is used to generate ROC curves. The area under the curve (AUC) is calculated to evaluate the diagnostic performance of these genes. 2.7 Immune infiltration analysis and Gene Set Enrichment Analysis (GSEA) To compare immune infiltration between RA and healthy samples, ssGSEA is performed using the “GSVA” R package to assess the relative abundance of 28 immune cell subsets. The “ggpubr” R package is used to visualize differences in immune infiltration, while the “corrplot” R package is applied to calculate Pearson correlation coefficients for each immune cell type and to evaluate the correlation between identified hub genes and immune cells. Additionally, ssGSEA analysis is conducted specifically for the hub genes. 2.8 Establishment of Nomogram model To assess the relationship between diagnostic genes and RA risk estimation, a nomogram model is constructed using the rms package in R and visualized with the plot(nom) function. A calibration curve is used to evaluate the predictive accuracy of the nomogram model, while decision curve analysis (DCA) is performed to assess its clinical applicability. 2.9 Validation results of the validation set The expression matrices of GSE55235, GSE55584, and GSE55235 are extracted using R, and the datasets are merged after excluding outlier samples. The Wilcoxon rank-sum test is performed to validate the differential expression of these biomarker candidates. The results are visualized using box plots. Hub genes that exhibit statistically significant differential expression are identified as key genes. 3 Results 3.1 Identification of DEGs and WGCNA to Identify Lipid Metabolism-Related Cross Genes Research Flowchart (Fig. 1 ), After integrating, standardizing, and normalizing the datasets, a total of 1,938 DEGs were identified across 14 healthy controls and 18 RA samples, including 1,250 upregulated and 688 downregulated genes (Fig. 2 A). WGCNA was performed on the integrated dataset, selecting a soft threshold of 5 for the scale-free network (Fig. 2 B). Based on this optimal threshold, a co-expression network was constructed, and a gene clustering heatmap was generated (Fig. 2 C). Notably, the module correlation heatmap revealed that the MEturquoise module was identified as the hub module (Fig. 2 D), which contains 1,312 genes. The gene distribution results of the MEturquoise module showed a high correlation between RA and the module members, indicating that genes in this module are significantly associated with RA (Fig. 2 E). The top three correlated module genes were intersected with the DEGs, and a Venn diagram was generated to display the overlap: 933 cross genes were identified between the DEGs and the WGCNA dataset. LMRGs were obtained from the MigSB database and intersected with the above cross genes, resulting in the identification of 29 intersections genes (Fig. 2 F). 3.2 GO and KEGG Enrichment of LMRGs To explore the potential biological functions and signaling pathways of the identified intersections genes, GO and KEGG enrichment analyses were performed on the 29 intersections genes. The GO analysis results revealed that the top 10 enriched BP were primarily associated with lipid catabolic processes, fatty acid metabolism, phospholipid metabolism, glycerolipid biosynthesis, metabolism, glycerophospholipid metabolism, long-chain fatty acid metabolism, sphingolipid catabolic processes, and membrane lipid degradation (Fig. 3 A). The top 7 enriched CC were located in lysosomal lumens, vascular lumens, major lysosomes, azurophilic granules, phosphoinositide 3-kinase complex, peroxisomal matrix, and microtubule lumens. The top 10 MF were mainly related to phosphatase activity, lipase activity, hydrolase activity, phosphatidylinositol 3-kinase activity, acylglycerol O-acyltransferase activity, O-acyltransferase activity, hydrogenase activity, 1-phosphatidylinositol-3-kinase activity, phosphoinositide phosphatase activity, and phosphoinositide 3-kinase activity. KEGG pathway enrichment analysis showed that the top 15 enriched pathways were primarily involved in sphingolipid metabolism, inositol phosphate metabolism, lysosomes, phosphoinositide signaling system, glycerophospholipid metabolism, platelet activation, biosynthesis of unsaturated fatty acids, ether lipid metabolism, ovarian steroidogenesis, fatty acid metabolism, regulation of adipocyte lipolysis, arachidonic acid metabolism, GnRH secretion, FcεRI signaling pathway, and prolactin signaling pathway (Fig. 3 B). These results provide valuable insights into the potential biological roles and pathways associated with lipid metabolism in RA and may offer new therapeutic targets. 3.3 Identification of Hub Genes Using the LASSO Cox regression algorithm, five genes were selected from the intersection genes: INPP5D, PLD3, PIK3CD, NEU1, and CYP2C19 (Fig. 4 A-B). The RF method identified 10 feature genes: HSD17B4, FDXR, NR1H3, CYP2C19, PLBD1, TNFAIP8L2, PLD3, PIK3CD, NEU1, and ABCC3 (Fig. 4 C-D). The overlap of feature genes from both methods resulted in four hub genes: PLD3, PIK3CD, NEU1, and CYP2C19 (Fig. 4 E). These four overlapping genes are considered key candidates for further investigation as potential biomarkers for RA. 3.4 ROC Curve Analysis To validate the diagnostic potential of the four feature genes, we plotted ROC curves and compared their AUC values to assess their diagnostic value. The AUC values for PLD3, PIK3CD, NEU1, and CYP2C19 were 0.813, 0.857, 0.830, and 0.790, respectively, indicating moderate diagnostic value, with PIK3CD showing the highest diagnostic potential (Fig. 5 A-D). Additionally, a combined ROC curve analysis of the hub genes revealed an AUC value of 0.913, significantly higher than the individual gene AUCs, suggesting that the model as a whole is more effective for disease prediction (Fig. 5 E). Furthermore, we also validated the hub genes associated with ferroptosis, as identified by Yihua Fan et al [18] , The AUC values for VEGFA, PTGS2, and JUN were 0.591, 0.567, and 0.631, respectively (Fig. 6 ). These results indicate that our identified hub genes provide higher diagnostic value compared to previously identified ferroptosis-related genes. 3.5 Nomogram and Immune Infiltration Correlation Analysis A Nomogram model was constructed based on the identified hub genes to evaluate the contribution of each gene to the risk of developing RA (Fig. 7 A). Calibration curves were employed to assess the model, and the results demonstrated that the four hub genes closely aligned with the ideal model (Fig. 7 B-C). Subsequently, the correlation of immune cell infiltration was further analyzed using the Timer CellMarker database. The findings indicated that the four key genes were likely negatively correlated with the activation of several immune cell types, including B cells, CD4 + T cells, CD8 + T cells, dendritic cells, central memory CD8 + T cells, myeloid-derived suppressor cells, regulatory T cells, follicular T cells, Th1 cells, memory B cells, macrophages, and γδT cells. In contrast, positive correlations were observed with effector memory CD4 + T cells and immature dendritic cells (Fig. 7 D). Furthermore, the correlation between the four hub genes and various immune checkpoints was assessed (Fig. 7 E). The analysis revealed that: PLD3 was predominantly upregulated in immature dendritic cells and CD56bright NK cells. PIK3CD was highly expressed in Th1 cells, regulatory T cells, monocytes, myeloid-derived suppressor cells, macrophages, immature B cells, γδT cells, eosinophils, central memory CD4 + T cells, and activated CD4 + and CD8 + T cells, as well as dendritic cells. NEU1 was predominantly expressed in immature dendritic cells, and its expression was upregulated in memory B cells, myeloid-derived suppressor cells, immature B cells, γδT cells, central memory CD4 + and CD8 + T cells, and NK cells. CYP2C19 exhibited downregulation in most of the immune checkpoints where the expression of the other three genes was upregulated. These results suggest that these key genes play a significant role in modulating the immune landscape of RA and may contribute to immune cell modulation in the disease. 3.6 GSEA Analysis To further investigate the role of the hub genes in disease, single-gene GSEA analysis was performed for the four hub genes. In-depth analysis revealed that genes such as PLD3 were upregulated in pathways related to glutathione metabolism, lysosome, and viral myocarditis (Fig. 8 A). PIK3CD exhibited high expression in pathways associated with allograft rejection, IgA in the intestinal immune network, and systemic lupus erythematosus (Fig. 8 B). In contrast, NEU1 was downregulated in pathways related to the hematopoietic cell lineage, lysosome, and Vibrio cholerae infection (Fig. 8 C). CYP2C19 was upregulated in pathways involving DNA replication, proteasome, and systemic lupus erythematosus (Fig. 8 D). 3.7 Validation of Results in the Validation Set The expression of the four RA lipid metabolism hub genes in the validation set was visualized. Ultimately, the expression difference of PIK3CD was statistically significant and consistent with the training set, showing upregulation in RA (Fig. 9 A-D). 4 Discussion RA is a chronic autoimmune disease that involves multiple bilateral joints. Its clinical characteristics include tenosynovitis, synovial tissue proliferation, vascular proliferation, cartilage destruction, and bone erosion[ 19 , 20 ]. The hallmark of the disease is the presence of various autoantibodies, such as rheumatoid factor (RF), anti-citrullinated protein antibodies (ACPA), and antibodies against post-translationally modified proteins, such as carbamylated proteins (anti-CarP antibodies). These autoantibodies can form immune complexes within the joints, activating the immune system and triggering aggressive synovial inflammation and cartilage destruction, ultimately leading to irreversible joint deformities or disabilities[ 21 ]. Therefore, exploring novel diagnostic biomarkers and targeted therapeutic strategies holds significant clinical value for early intervention in RA. Recent studies have revealed that lipid metabolism dysregulation plays a role in the pathological progression of RA by regulating the inflammatory microenvironment and oxidative stress responses. Notably, lipid peroxidation, as a key inducer of ferroptosis, may exacerbate joint damage by activating the ferroptosis pathway in synovial fibroblasts[ 22 ]. However, the specific molecular mechanisms underlying lipid metabolism dysregulation and RA progression remain unclear, which serves as a critical focus of this study. Based on the above background, this study identified 1,938 DEGs by comparing synovial tissue samples from the RA group and the control group. Furthermore, a gene co-expression network was constructed using the WGCNA approach to identify the top three gene modules most strongly associated with the RA phenotype. By integrating DEGs, lipid metabolism-related gene sets, and core gene modules, 29 candidate risk genes associated with lipid metabolism in RA were ultimately identified. To analyze the biological functions of the candidate genes, we performed KEGG and GO enrichment analyses on the 29 identified genes. These genes were found to be associated with lipid catabolic processes, fatty acid metabolism, phosphate metabolism, phospholipid metabolism, and triglyceride biosynthesis in biological processes. In terms of cellular components, these genes are primarily linked to the lysosomal lumen, blood vessel lumen, major lysosomes, azurophilic granules, and phosphoinositide 3-kinase complex. In molecular function, they are mainly involved in phosphatase activity, lipase activity, hydrolase activity, phosphatidylinositol kinase activity, and acylglycerol O-acyltransferase activity. These findings suggest that these genes may influence the progression of RA by regulating the energy metabolism imbalance in synovial cells. Notably, significant heterogeneity in the fatty acid profile has been observed in the synovial fluid (SF) of RA patients[ 23 , 24 ]. The ratio of arachidonic acid (AA) to docosahexaenoic acid (DHA) is higher in the shoulder joint SF, while oleic acid predominates in the knee joint SF[ 25 ], This spatially specific lipid distribution may influence disease progression by altering joint lubrication, regulating synovial inflammation, and affecting cartilage degradation enzyme activity. To further identify the core regulatory factors, we integrated Lasso regression and random forest machine learning algorithms, ultimately determining PLD3, PIK3CD, NEU1, and CYP2C19 as key RA lipid metabolism-related hub genes. The nomogram model based on these four genes demonstrated excellent predictive performance in the training set (AUC > 0.8), confirming their potential as RA diagnostic biomarkers. PLD3, a member of the phospholipase D family, encodes a 5'-3' nucleic acid exonuclease. It participates in phospholipid metabolism, affecting cell membrane lipid composition and signal transduction, and can contribute to inflammatory responses via lipid metabolism-related pathways, such as NF-κB [26] . PIK3CD encodes a subunit of phosphoinositide 3-kinase and plays a role in converting phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP3), thereby activating downstream signaling pathways like AKT/mTOR [27] , Previous studies have shown that inhibiting the PIK3CD pathway reduces autoreactive B cells and autoantibody levels, significantly improving inflammation in autoimmune diseases [28] . NEU1, an enzyme that hydrolyzes sialic acid on the cell surface, has been shown to activate AMPKα and subsequently SIRT3, regulating fibrosis, inflammation, apoptosis, and oxidative stress in diabetic cardiac tissue [29] . CYP2C19, a member of the CYP450 enzyme family, is closely related to drug metabolism in the body [30] . Research indicates that upregulation of CYP2C19 is associated with decreased C-reactive protein levels in patients with invasive aspergillosis treated with voriconazole [31] . To further assess the clinical value of lipid metabolism hub genes, we compared their ROC curves with those of the ferroptosis-related hub genes reported by the Yihua Fan team [18] . The results indicated that the lipid metabolism hub genes identified in this study had higher diagnostic efficacy than the ferroptosis-related genes. This finding suggests that lipid metabolism, as a chronic mechanism continuously regulating the energy homeostasis of the synovial microenvironment, may be more deeply involved in the pathological progression of RA than acute cell death pathways like ferroptosis. We speculate that lipid metabolism reprogramming, through long-term effects on immune cell activation, synovial cell proliferation, and inflammatory mediator release, plays a more central regulatory role in the development and progression of RA. Given that immune dysregulation is a core pathological feature of RA [32] , we further analyzed the relationship between hub genes and immune cell infiltration. The results indicated a significant increase in the proportion of macrophages, CD8⁺ T cells, and Treg cells in RA synovial tissue. Additionally, PLD3, PIK3CD, and NEU1 showed positive correlations with immune cell infiltration, whereas CYP2C19 exhibited a negative correlation. Notably, the exonuclease activity of PLD3 plays a crucial role in regulating immune cell inflammatory responses. When PLD3 function is impaired, the accumulation of ssDNA in the lysosome triggers downstream inflammatory signaling pathways (such as TLR9 and cGAS-STING), which in turn activate pro-inflammatory factors (e.g., IFN-α and TNF-α) [26] . Furthermore, activation of PIK3CD leads to abnormal activation and differentiation of CD4⁺ T cells, upregulating the expression of Th1 cells and suppressing the generation of Treg cells [33] . Further research revealed that gain-of-function mutations in PIK3CD resulted in an increase in memory T cells and follicular helper T cells (TFH), but with dysfunction, as indicated by increased expression of PD-1, CXCR3, and IFN-γ [34] . On the other hand, NEU1 deficiency leads to macrophage polarization towards the M2 phenotype, disrupting the balance between M1 and M2 macrophages [35] . Finally, CYP2C19 is highly expressed in M2 macrophages and, through metabolism, generates 11,12- and 14,15-epoxyeicosatrienoic acids (EETs). These metabolites act as PPARγ receptor agonists, promoting M2 macrophage polarization. This polarization helps regulate inflammatory responses and reduces the secretion of pro-inflammatory factors, thus exerting anti-inflammatory effects [36] . The above results demonstrate the significant role of these four hub genes in RA progression. Importantly, these genes are closely associated with immune cell dysfunction in immune infiltration, suggesting their potential as adjuncts for RA diagnosis and to mitigate immune responses during adjunctive therapy. Finally, we integrated three external datasets for external validation, and the results demonstrated that PIK3CD is involved in the lipid metabolism process of RA, indicating its potential as a valuable diagnostic biomarker. Additionally, it is closely associated with lipid metabolism-related pathways, showing further potential as a therapeutic target. However, more experiments are needed to validate the detailed mechanisms and functions of PIK3CD. 5 Conclusion This study demonstrated through bioinformatics analysis that the central gene PIK3CD exhibits differential expression between RA patients and healthy individuals, suggesting its potential as an immune-related biomarker and therapeutic target for RA. This provides a new perspective and direction for future in-depth research and treatment strategies. Declarations Authors ' Contributions : Z.F, X-PL and L-YT were responsible for writing first drafts and revising manuscripts. W.L, L-HH, W.W, K.Z, H-XX were responsible for finding literature. D-BL and B.W provided ideas for the work and revised it critically. All authors read and approved the final manuscript. Funding: This work was supported by the Key Fund Project of Hunan Provincial Department of Education (Grant No.23A0340), the Natural Science Foundation of Hunan Province (Grant No.2025JJ80606 and Grant No. 2023JJ50098), the Scientific Research Project of Health Commission of Hunan Province (Grant No. D202309016796) and Students’ platform for innovation and entrepreneurship training program (Grant No D202405241030541188) Declaration of Interest Statement: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Ethics statement ： Not applicable. Clinical trial number: not applicable. References Gravallese EM, Firestein GS: Rheumatoid Arthritis - Common Origins, Divergent Mechanisms. N Engl J Med 2023, 388:529–542. https://doi.org/10.1056/NEJMra2103726 Wang F, Palmer N, Fox K, Liao KP, Yu KH, Kou SC: Large-scale real-world data analyses of cancer risks among patients with rheumatoid arthritis. Int J Cancer 2023, 153:1139–1150. https://doi.org/10.1002/ijc.34606 Brown P, Pratt AG, Hyrich KL: Therapeutic advances in rheumatoid arthritis. Bmj 2024, 384:e070856 .https://doi.org/10.1136/bmj-2022-070856 Simon D, Minopoulou I, Kemenes S, Bayat S, Tascilar K, Mutlu MY, Valor-Méndez L, Krönke G, Hueber AJ, Schett G, Kleyer A: Baricitinib Improves Bone Properties and Biomechanics in Patients With Rheumatoid Arthritis: Results of the Prospective Interventional BARE BONE Trial. Arthritis Rheumatol 2023, 75:1923–1934. https://doi.org/10.1002/art.42617 Liu Y, Chen L, Chen Z, Liu M, Li X, Kou Y, Hou M, Wang H, Li X, Tian B, Dong J: Multifunctional Janus Nanoplatform for Efficiently Synergistic Theranostics of Rheumatoid Arthritis. ACS Nano 2023, 17:8167–8182. https://doi.org/10.1021/acsnano.2c11777 Iglesias-Artola JM, Nadler A: The Road to Quantitative Lipid Biochemistry in Living Cells. Acc Chem Res 2023, 56:810–820. https://doi.org/10.1021/acs.accounts.2c00804 Yuan Y, Li P, Li J, Zhao Q, Chang Y, He X: Protein lipidation in health and disease: molecular basis, physiological function and pathological implication. Signal Transduct Target Ther 2024, 9:60. https://doi.org/10.1038/s41392-024-01759-7 Martini T, Gobet C, Salati A, Blanc J, Mookhoek A, Reinehr M, Knott G, Sordet-Dessimoz J, Naef F: A sexually dimorphic hepatic cycle of periportal VLDL generation and subsequent pericentral VLDLR-mediated re-uptake. Nat Commun 2024, 15:8422. https://doi.org/10.1038/s41467-024-52751-2 Sinturel F, Chera S, Brulhart-Meynet MC, Montoya JP, Stenvers DJ, Bisschop PH, Kalsbeek A, Guessous I, Jornayvaz FR, Philippe J, et al: Circadian organization of lipid landscape is perturbed in type 2 diabetic patients. Cell Rep Med 2023, 4:101299. https://doi.org/10.1016/j.xcrm.2023.101299 Luo YX, Yang LL, Yao XQ: Gut microbiota-host lipid crosstalk in Alzheimer's disease: implications for disease progression and therapeutics. Mol Neurodegener 2024, 19:35. https://doi.org/10.1186/s13024-024-00720-0 Yang Y, Luo D, Shao Y, Shan Z, Liu Q, Weng J, He W, Zhang R, Li Q, Wang Z, Li X: circCAPRIN1 interacts with STAT2 to promote tumor progression and lipid synthesis via upregulating ACC1 expression in colorectal cancer. Cancer Commun (Lond) 2023, 43:100–122. https://doi.org/10.1002/cac2.12380 Lei Q, Yang J, Li L, Zhao N, Lu C, Lu A, He X: Lipid metabolism and rheumatoid arthritis. Front Immunol 2023, 14:1190607. https://doi.org/10.3389/fimmu.2023.1190607 Frommer KW, Schäffler A, Rehart S, Lehr A, Müller-Ladner U, Neumann E: Free fatty acids: potential proinflammatory mediators in rheumatic diseases. Ann Rheum Dis 2015, 74:303–310. https://doi.org/10.1136/annrheumdis-2013-203755 Wei J, Huang X, Zhang X, Chen G, Zhang C, Zhou X, Qi J, Zhang Y, Li X: Elevated fatty acid β-oxidation by leptin contributes to the proinflammatory characteristics of fibroblast-like synoviocytes from RA patients via LKB1-AMPK pathway. Cell Death Dis 2023, 14:97. https://doi.org/10.1038/s41419-023-05641-2 Chen K, Yan Y, Li C, Yuan J, Wang F, Huang P, Qian N, Qi J, Zhou H, Zhou Q, et al: Increased 15-lipoxygenase-1 expression in chondrocytes contributes to the pathogenesis of osteoarthritis. Cell Death Dis 2017, 8:e3109 .https://doi.org/10.1038/cddis.2017.511 Miyabe Y, Miyabe C, Luster AD: LTB(4) and BLT1 in inflammatory arthritis. Semin Immunol 2017, 33:52–57. https://doi.org/10.1016/j.smim.2017.09.009 Xu S, Lu H, Lin J, Chen Z, Jiang D: Regulation of TNFalpha and IL1beta in rheumatoid arthritis synovial fibroblasts by leukotriene B4. Rheumatol Int 2010, 30:1183–1189. https://doi.org/10.1007/s00296-009-1125-y Fan Y, Li Y, Fu X, Peng J, Chen Y, Chen T, Zhang D: Identification of potential ferroptosis key genes and immune infiltration in rheumatoid arthritis by integrated bioinformatics analysis. Heliyon 2023, 9:e21167 .https://doi.org/10.1016/j.heliyon.2023.e21167 Lin YJ, Anzaghe M, Schülke S: Update on the Pathomechanism, Diagnosis, and Treatment Options for Rheumatoid Arthritis. Cells 2020, 9. https://doi.org/10.3390/cells9040880 Jang S, Kwon EJ, Lee JJ: Rheumatoid Arthritis: Pathogenic Roles of Diverse Immune Cells. Int J Mol Sci 2022, 23. https://doi.org/10.3390/ijms23020905 Pertsinidou E, Saevarsdottir S, Manivel VA, Klareskog L, Alfredsson L, Mathsson-Alm L, Hansson M, Cornillet M, Serre G, Holmdahl R, et al: In early rheumatoid arthritis, anticitrullinated peptide antibodies associate with low number of affected joints and rheumatoid factor associates with systemic inflammation. Ann Rheum Dis 2024, 83:277–287. https://doi.org/10.1136/ard-2023-224728 Cheng Q, Chen M, Liu M, Chen X, Zhu L, Xu J, Xue J, Wu H, Du Y: Semaphorin 5A suppresses ferroptosis through activation of PI3K-AKT-mTOR signaling in rheumatoid arthritis. Cell Death Dis 2022, 13:608. https://doi.org/10.1038/s41419-022-05065-4 Mustonen AM, Käkelä R, Lehenkari P, Huhtakangas J, Turunen S, Joukainen A, Kääriäinen T, Paakkonen T, Kröger H, Nieminen P: Distinct fatty acid signatures in infrapatellar fat pad and synovial fluid of patients with osteoarthritis versus rheumatoid arthritis. Arthritis Res Ther 2019, 21:124. https://doi.org/10.1186/s13075-019-1914-y Mustonen AM, Käkelä R, Joukainen A, Lehenkari P, Jaroma A, Kääriäinen T, Kröger H, Paakkonen T, Sihvo SP, Nieminen P: Synovial Fluid Fatty Acid Profiles Are Differently Altered by Inflammatory Joint Pathologies in the Shoulder and Knee Joints. Biology (Basel) 2021, 10.https://doi.org/10.3390/biology10050401 Wang N, Zhao X, Huai J, Li Y, Cheng C, Bi K, Dai R: Arachidonic acid metabonomics study for understanding therapeutic mechanism of Huo Luo Xiao Ling Dan on rat model of rheumatoid arthritis. J Ethnopharmacol 2018, 217:205–211. https://doi.org/10.1016/j.jep.2018.02.027 Gavin AL, Huang D, Huber C, Mårtensson A, Tardif V, Skog PD, Blane TR, Thinnes TC, Osborn K, Chong HS, et al: PLD3 and PLD4 are single-stranded acid exonucleases that regulate endosomal nucleic-acid sensing. Nat Immunol 2018, 19:942–953. https://doi.org/10.1038/s41590-018-0179-y Wang X, Han H, Shi X, Nie X, Zhu R, Jin J, Zhou H: Genetic insights of blood lipid metabolites on polycystic ovary syndrome risk: a bidirectional two-sample Mendelian randomization study. Front Endocrinol (Lausanne) 2024, 15:1391826. https://doi.org/10.3389/fendo.2024.1391826 Scuron MD, Fay BL, Connell AJ, Oliver J, Smith PA: The PI3Kδ inhibitor parsaclisib ameliorates pathology and reduces autoantibody formation in preclinical models of systemic lupus erythematosus and Sjӧgren's syndrome. Int Immunopharmacol 2021, 98:107904. https://doi.org/10.1016/j.intimp.2021.107904 Guo Z, Tuo H, Tang N, Liu FY, Ma SQ, An P, Yang D, Wang MY, Fan D, Yang Z, Tang QZ: Neuraminidase 1 deficiency attenuates cardiac dysfunction, oxidative stress, fibrosis, inflammatory via AMPK-SIRT3 pathway in diabetic cardiomyopathy mice. Int J Biol Sci 2022, 18:826–840. https://doi.org/10.7150/ijbs.65938 Wiese MD, Schnabl M, O'Doherty C, Spargo LD, Sorich MJ, Cleland LG, Proudman SM: Polymorphisms in cytochrome P450 2C19 enzyme and cessation of leflunomide in patients with rheumatoid arthritis. Arthritis Res Ther 2012, 14:R163 .https://doi.org/10.1186/ar3911 Klomp SD, Veringa A, Alffenaar JC, de Boer MGJ, Span LFR, Guchelaar HJ, Swen JJ: Inflammation altered correlation between CYP2C19 genotype and CYP2C19 activity in patients receiving voriconazole. Clin Transl Sci 2024, 17:e13887 .https://doi.org/10.1111/cts.13887 Zhang J, Wang C, Wu X, Shen Q, Du Y: Nanozyme-based therapeutic strategies for rheumatoid arthritis. J Control Release 2025, 377:716–734. https://doi.org/10.1016/j.jconrel.2024.11.072 Hanna BS, Roessner PM, Scheffold A, Jebaraj BMC, Demerdash Y, Öztürk S, Lichter P, Stilgenbauer S, Seiffert M: PI3Kδ inhibition modulates regulatory and effector T-cell differentiation and function in chronic lymphocytic leukemia. Leukemia 2019, 33:1427–1438. https://doi.org/10.1038/s41375-018-0318-3 Bier J, Rao G, Payne K, Brigden H, French E, Pelham SJ, Lau A, Lenthall H, Edwards ESJ, Smart JM, et al: Activating mutations in PIK3CD disrupt the differentiation and function of human and murine CD4(+) T cells. J Allergy Clin Immunol 2019, 144:236–253. https://doi.org/10.1016/j.jaci.2019.01.033 Escalona E, Olate-Briones A, Albornoz-Muñoz S, Bonacic-Doric E, Rodríguez-Arriaza F, Herrada AA, Escobedo N: Neu1 deficiency and fibrotic lymph node microenvironment lead to imbalance in M1/M2 macrophage polarization. Front Immunol 2024, 15:1462853. https://doi.org/10.3389/fimmu.2024.1462853 Cho HY, Ahn S, Cho YS, Seo SK, Kim DH, Shin JG, Lee SJ: CYP2C19 Contributes to THP-1-Cell-Derived M2 Macrophage Polarization by Producing 11,12- and 14,15-Epoxyeicosatrienoic Acid, Agonists of the PPARγ Receptor. Pharmaceuticals (Basel) 2023, 16.https://doi.org/10.3390/ph16040593 Additional Declarations No competing interests reported. Supplementary Files Graphicalabstract.tif Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7913208","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":535420147,"identity":"be368b53-c7d2-4328-9275-18765f81b544","order_by":0,"name":"Zhi Feng","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Zhi","middleName":"","lastName":"Feng","suffix":""},{"id":535420148,"identity":"60568e58-c1c7-457c-9174-bfaf4798f2b3","order_by":1,"name":"Xiangping Luo","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Xiangping","middleName":"","lastName":"Luo","suffix":""},{"id":535420149,"identity":"81ee593b-e6fc-4636-9f00-0f077d938e19","order_by":2,"name":"Linyue Tang","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Linyue","middleName":"","lastName":"Tang","suffix":""},{"id":535420150,"identity":"d5636692-e0cf-4d4b-82e4-8478c61af7a3","order_by":3,"name":"Wei Long","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Long","suffix":""},{"id":535420151,"identity":"83005221-9bdf-4e77-a7da-66c5c9cd4abb","order_by":4,"name":"Lvheng He","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Lvheng","middleName":"","lastName":"He","suffix":""},{"id":535420152,"identity":"951bac29-549a-4800-aeda-cc1b5a028cb4","order_by":5,"name":"Wei Wang","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Wang","suffix":""},{"id":535420153,"identity":"86cc563b-45fc-455d-af6f-76cb342464e9","order_by":6,"name":"Kai Zhang","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Zhang","suffix":""},{"id":535420154,"identity":"ba4685e2-8a18-43eb-9f4c-6049fa0f14ae","order_by":7,"name":"Haixia Xu","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Haixia","middleName":"","lastName":"Xu","suffix":""},{"id":535420155,"identity":"0c3c0cd9-2cee-4034-b1fb-3c7e1c526959","order_by":8,"name":"Debin Li","email":"","orcid":"","institution":"University of South China","correspondingAuthor":false,"prefix":"","firstName":"Debin","middleName":"","lastName":"Li","suffix":""},{"id":535420156,"identity":"63ff6369-b9bb-4268-9ae5-9ca78280b69d","order_by":9,"name":"Bo Wei","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAt0lEQVRIiWNgGAWjYLACxgYbCIOHBC1ppGs5TIIWg/OHHzDz7jifOL/9AOODt20M8uYEtdxIM2DmPXM7ccOZBGbDuW0MhjsbCGrhYf+d2wbUIsHAJs3bxpBgcICgw84wMOe2nUucP4OB/TdxWg7kgLQcSGy4wcDGTJQWSZBf/p5JNt5wJrFZcs45CcMNhLTwAUOMceYOO9n57YcPfnhTZiNP0BYkwNgAJCSIVz8KRsEoGAWjADcAAFiVPvC/72L8AAAAAElFTkSuQmCC","orcid":"","institution":"University of South China","correspondingAuthor":true,"prefix":"","firstName":"Bo","middleName":"","lastName":"Wei","suffix":""}],"badges":[],"createdAt":"2025-10-21 10:04:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7913208/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7913208/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95263496,"identity":"38db1092-f243-4da5-9857-ab7db137428f","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2567247,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/559ed88bfd053a04653c7d69.docx"},{"id":95263507,"identity":"cf8d362d-47f7-4963-9a79-b0310ebfd196","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34191716,"visible":true,"origin":"","legend":"","description":"","filename":"Graphicalabstract.tif","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/7c3e59e4380d3960995e21d4.tif"},{"id":95263481,"identity":"5530b65a-c7e8-43cc-b813-a29635d922d7","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"json","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10195,"visible":true,"origin":"","legend":"","description":"","filename":"1c2089222d364135aa7d1130b7d4be4b.json","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/664d729045f086c9f4352bba.json"},{"id":95263482,"identity":"10c3b89c-e87c-4c1f-b76b-3e66999a3acd","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17252,"visible":true,"origin":"","legend":"","description":"","filename":"Highlight.docx","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/8ea1b37565986352f1bf84bc.docx"},{"id":95312985,"identity":"88620dfc-2e81-4d16-ada4-4c1b4cba778a","added_by":"auto","created_at":"2025-11-06 15:50:41","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117465,"visible":true,"origin":"","legend":"","description":"","filename":"1c2089222d364135aa7d1130b7d4be4b1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/ac65f847d8d5b5369859ea1b.xml"},{"id":95263515,"identity":"cfdc7e1a-b11f-483a-9107-c9309a81d82e","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34191716,"visible":true,"origin":"","legend":"","description":"","filename":"Graphicalabstract.tif","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/d21a509ed90a116c69a08cf5.tif"},{"id":95313539,"identity":"c49b3ce0-60a1-432f-a77a-bf93d8d6b36e","added_by":"auto","created_at":"2025-11-06 15:51:39","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":210924,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/d4d4ef0805bb90dc583129f8.jpeg"},{"id":95263500,"identity":"c56f19af-cf32-4e33-b6ca-e9998ceed2e2","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":654360,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/fe5c2eca79a626278cee4b02.jpeg"},{"id":95313554,"identity":"3c624eb2-d602-49ce-9fca-def2fc289475","added_by":"auto","created_at":"2025-11-06 15:51:41","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":264204,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/0deb848146e835adbe2d2875.jpeg"},{"id":95313304,"identity":"2b0518e0-01a4-4b7e-bd19-bd862cf00889","added_by":"auto","created_at":"2025-11-06 15:51:14","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":202264,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/ae92992911734d8171151c85.jpeg"},{"id":95263487,"identity":"bcd55859-3eba-4442-9103-9a7e66f67af9","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":113830,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/acc109fc4a64c3f5f7d1222c.jpeg"},{"id":95263489,"identity":"3dc956a5-cc98-4553-8dcb-bdf1fa7ddc34","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":65494,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/ab479abd1021b364d7c02b4d.jpeg"},{"id":95263493,"identity":"56a8c58d-48aa-498e-9046-1ca0f5d81978","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":381494,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/2b609ffa4ae4dec25abf2603.jpeg"},{"id":95313653,"identity":"9a04eca4-5993-4dbc-b704-11bbb8e8717b","added_by":"auto","created_at":"2025-11-06 15:51:49","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":399342,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/2905f7f4427192c347fd9530.jpeg"},{"id":95263514,"identity":"204db19b-ce66-440a-a702-ef8fa74106a7","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":165220,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/3cf3f7420d877c93eb45ab03.jpeg"},{"id":95313143,"identity":"604c7e42-7798-4b7c-a0fd-650d12238292","added_by":"auto","created_at":"2025-11-06 15:51:00","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1028523,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineGraphicalabstract.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/dfd3c96f360269a082d025e0.png"},{"id":95313389,"identity":"a3514a0b-024e-4aff-ba6c-1faf4942e8d2","added_by":"auto","created_at":"2025-11-06 15:51:20","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31550,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/09f92146dc3f4f734fe02965.png"},{"id":95263501,"identity":"6f9e9f12-3663-470b-a680-b414718a4080","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90243,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/b9e961aef68d19f80893b25e.png"},{"id":95263508,"identity":"1dcbb0a0-a829-41b9-9b48-6d19188d49e9","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40776,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/fbbc94f9a032ad19e7ea1185.png"},{"id":95312893,"identity":"277b332f-d34c-4ee4-acd7-19759c2fe72f","added_by":"auto","created_at":"2025-11-06 15:50:32","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29278,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/455cc08be442bb3f51de99e2.png"},{"id":95312890,"identity":"38bcb7ea-9366-4611-bff5-47e8ed67271d","added_by":"auto","created_at":"2025-11-06 15:50:32","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16764,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/94588bd043a95ae773c703ce.png"},{"id":95263503,"identity":"c0c72401-ce4f-4bb6-8ca2-05db68b70cb7","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12328,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/d55ced333d7165fffa733476.png"},{"id":95313522,"identity":"fa819c21-ab69-4f5f-9362-ddbbebc2fbc0","added_by":"auto","created_at":"2025-11-06 15:51:36","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56931,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/9256f4726c067f585acbdfe5.png"},{"id":95263505,"identity":"86851e29-714b-41c4-ae32-85b9c25c0725","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46171,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/1a75abdb82db0a27ccb72493.png"},{"id":95312826,"identity":"f6e60ad8-58cb-465f-bd6c-cb53d0c227aa","added_by":"auto","created_at":"2025-11-06 15:50:23","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29283,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/6689214d81ea3bec6e551851.png"},{"id":95263511,"identity":"c066aef8-81e7-4df1-ad3b-eab92b64f22d","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"xml","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118049,"visible":true,"origin":"","legend":"","description":"","filename":"1c2089222d364135aa7d1130b7d4be4b1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/fa189b26e5b630f901eb887f.xml"},{"id":95313108,"identity":"74ed2ab9-bdbd-4b1d-b2aa-5b398d403cad","added_by":"auto","created_at":"2025-11-06 15:50:54","extension":"html","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127123,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/13f7289e9c50fa9e0ad6f022.html"},{"id":95313593,"identity":"6ecafe00-c510-45c0-a206-e11c47e2043c","added_by":"auto","created_at":"2025-11-06 15:51:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":682731,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow chart.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/bf118523c942a68f19b21829.png"},{"id":95263478,"identity":"c9c7eecd-2bd3-4295-b086-d1a5c59d9d76","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2089976,"visible":true,"origin":"","legend":"\u003cp\u003eDEGs screening between RA and healthy control. Identification of critical modules by WGCNA. \u003cstrong\u003eA \u003c/strong\u003eVolcano graphic visualizing DEGs of RA and normal samples. \u003cstrong\u003eB\u003c/strong\u003e Analysis of the scale free ft index and analysis of the mean connectivity for various soft-thresholding powers. \u003cstrong\u003eC\u003c/strong\u003e Gene dendrogram obtained by average linkage hierarchical clustering. The color row underneath the dendrogram shows the module assignment determined by the Dynamic Tree Cut, in which 18 modules were identified. \u003cstrong\u003eD\u003c/strong\u003e The heatmap showed the correlation between module eigengenes and the two modification patterns. \u003cstrong\u003eE\u003c/strong\u003e Scatter plot of the turquoise module. \u003cstrong\u003eF\u003c/strong\u003e Venn diagram for overlapped genes.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/c20bc72e39bd9bd7b6a825cc.png"},{"id":95263479,"identity":"6597f4fc-0b97-4550-a073-14c55bed28bb","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":919175,"visible":true,"origin":"","legend":"\u003cp\u003eFunctional Intersection genes enrichment. \u003cstrong\u003eA\u003c/strong\u003e GO analysis. \u003cstrong\u003eB\u003c/strong\u003e KEGG pathway analysis.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/0d008e975c98a17cbfa60d93.png"},{"id":95263484,"identity":"b87f2432-ce15-4ffd-9b19-f396108f1ae7","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":621333,"visible":true,"origin":"","legend":"\u003cp\u003eHub gene identification. \u003cstrong\u003eA\u003c/strong\u003e LASSO coefficient profiles of candidate genes. \u003cstrong\u003eB\u003c/strong\u003eCross-validation to select the optimal tuning parameter log (Lambda) in LASSO regression analysis. \u003cstrong\u003eC\u003c/strong\u003e Prediction accuracy of the RF model. \u003cstrong\u003eD\u003c/strong\u003eGene importance scores of RF model. \u003cstrong\u003eE\u003c/strong\u003e Venn diagram of four hub genes shared by the RF and LASSO algorithms.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/3b846a13c52f65bef2f31dd9.png"},{"id":95263485,"identity":"19fc8627-36d0-4f67-9418-d68f22c78806","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":301511,"visible":true,"origin":"","legend":"\u003cp\u003eVerification of the 4 specifically expressed hub genes. \u003cstrong\u003eA \u003c/strong\u003eROC curve of PLD3.\u003cstrong\u003e B \u003c/strong\u003eROC curve of PIK3CD.\u003cstrong\u003e C \u003c/strong\u003eROC curve of NEU2.\u003cstrong\u003e D \u003c/strong\u003eROC curve of CYP2C19.\u003cstrong\u003eE \u003c/strong\u003eROC curve of hub gene.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/8f40e24a94894ecd3c5faf7d.png"},{"id":95263494,"identity":"9cabc514-8bac-4c49-a679-8119e818c981","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":184642,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve of ferroptosis hub genes. \u003cstrong\u003eA \u003c/strong\u003eROC curve of VEGFA.\u003cstrong\u003e B \u003c/strong\u003eROC curve of PTGS2.\u003cstrong\u003eC \u003c/strong\u003eROC curve of JUN.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/b2dd79a8ed76f26014c8d92a.png"},{"id":95313261,"identity":"c0a42a8e-9d3b-444c-8e03-b83ba29eb63c","added_by":"auto","created_at":"2025-11-06 15:51:11","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1468029,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram and Immune Infiltration of hub gene. \u003cstrong\u003eA\u003c/strong\u003e The nomogram of hub genes was visible for RA. \u003cstrong\u003eB\u003c/strong\u003e Calibration curve to evaluate nomogram model. \u003cstrong\u003eC\u003c/strong\u003e The DCA curve suggests that decision-making based on the nomogram may benefit in RA patients because the blue lines are consistently maintained above the gray and black lines of 0 to 1. \u003cstrong\u003eD\u003c/strong\u003e Comparison of immune cell infiltration levels in hub genes. \u003cstrong\u003eE\u003c/strong\u003e Immune cell correlation heatmaps of patients with RA. \u003cstrong\u003eF\u003c/strong\u003eCorrelation analysis among 4 differentially expressed hub genes in RA patients. *p \u0026lt; 0.05; **p \u0026lt; 0.01; ***p \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/f25164163816b5169cca08ff.png"},{"id":95263486,"identity":"881524f5-7328-4215-a4a2-1f09c1a6c801","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1029261,"visible":true,"origin":"","legend":"\u003cp\u003eGSEA analysis for hub genes. Top 3 GSEA enrichment in RA. A Top 3 enrichment terms for upregulated PLD3. B Top 3 enrichment terms for upregulated PIK3CD. C Top 3 enrichment terms for downregulated NEU1. D Top 3 enrichment terms for upregulated CYP2C19.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/abe4748ae9b5d8ea21838750.png"},{"id":95263490,"identity":"40bccaeb-ec99-462f-b914-c49186714fcc","added_by":"auto","created_at":"2025-11-06 05:21:57","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":478107,"visible":true,"origin":"","legend":"\u003cp\u003eValidation set validation gene expression. \u003cstrong\u003eA\u003c/strong\u003e Box plot showing the expression of PIK3CD. \u003cstrong\u003eB\u003c/strong\u003e Box plot showing the expression of PLD3. \u003cstrong\u003eC\u003c/strong\u003e Box plot showing the expression of NEU1. \u003cstrong\u003eD\u003c/strong\u003e Box plot showing the expression of CYP2C19. P \u0026lt; 0.05 was considered statistically different when compared to the control group. *p \u0026lt; 0.05; **p \u0026lt; 0.01; ***p \u0026lt; 0.001, ****p \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/a5e7e8d7943b4696795e8888.png"},{"id":99787993,"identity":"4571cdee-aec8-4f4b-849e-88d2575b30b4","added_by":"auto","created_at":"2026-01-08 12:43:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7653739,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/2b825861-1fab-44d6-bdce-134e23cb90c5.pdf"},{"id":95313080,"identity":"6534c91d-b75e-4b82-893c-086dc68425ba","added_by":"auto","created_at":"2025-11-06 15:50:52","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2009471,"visible":true,"origin":"","legend":"","description":"","filename":"Graphicalabstract.tif","url":"https://assets-eu.researchsquare.com/files/rs-7913208/v1/7cfa7da248730d9b90690caf.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identification of Lipid Metabolism-Related Genes in Rheumatoid Arthritis Using Bioinformatics and Machine Learning","fulltext":[{"header":"Highlights","content":"\u003cul\u003e\n \u003cli\u003eOur study integrates multiple datasets to enhance the robustness of our findings\u003c/li\u003e\n \u003cli\u003eThe combination of WGCNA and machine learning algorithms allowed us to identify key genes with high precision.\u003c/li\u003e\n \u003cli\u003eThe construction of a nomogram and ROC curve provides a practical tool for predicting RA risk.\u003c/li\u003e\n \u003cli\u003eOur analysis of immune cell infiltration and validation in an independent dataset strengthens the clinical relevance of our findings..\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"1 Introduction","content":"\u003cp\u003eRheumatoid arthritis (RA) is a common chronic inflammatory autoimmune disease characterized by symmetric joint inflammation, leading to pain, swelling, dysfunction, and, in severe cases, joint deformity and loss of function[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The global prevalence of RA is estimated to be 0.5%\u0026ndash;1%, and its chronic progressive and disabling nature significantly reduces patients\u0026rsquo; quality of life while imposing a substantial burden on healthcare systems[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Current therapeutic strategies primarily include nonsteroidal anti-inflammatory drugs (NSAIDs), disease-modifying antirheumatic drugs (DMARDs), and glucocorticoids. Although these treatments can alleviate symptoms to some extent, they fail to halt disease progression[\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Therefore, a deeper understanding of RA pathogenesis and the identification of novel therapeutic targets are critical for improving patient outcomes and quality of life.\u003c/p\u003e\u003cp\u003eLipids, including fatty acids, triglycerides, cholesterol, phospholipids, and sphingolipids, serve not only as energy storage molecules but also as key mediators of signal transduction and inflammation [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Emerging evidence suggests that lipid metabolism dysregulation is closely associated with various inflammatory diseases, such as atherosclerosis and osteoarthritis[\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Recent studies have indicated that lipid reprogramming within the synovial microenvironment of RA patients drives joint destruction by regulating inflammatory cytokine secretion and oxidative stress[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. The lipid composition of synovial fluid is strongly correlated with inflammation severity. For instance, fatty acids promote fibroblast-like synoviocytes (FLSs) secretion of pro-inflammatory cytokines and matrix metalloproteinases (e.g., IL-6, IL-8, MMP-1, and MMP-3)[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Additionally, the arachidonic acid metabolite 15s-HETE enhances MMP-2 expression in FLS via its downstream enzyme 15-LOX, while another metabolite, LTB4, binds to LTB4 receptor 2 on FLS, significantly upregulating TNF-α and IL-1β production[\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Increasing evidence supports the critical role of lipid metabolism in RA pathogenesis; however, the precise mechanisms by which lipid metabolism mediates RA pathophysiology remain unclear.\u003c/p\u003e\u003cp\u003eTo address this issue, RA-related public datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were retrieved from the Gene Expression Omnibus (GEO) database. WGCNA was performed to identify RA-associated genes, while lipid metabolism-related gene sets were obtained from the MigSB database. Machine learning algorithms were then applied to identify hub genes involved in lipid metabolism in RA. Differentially expressed lipid metabolism-related genes (LMRGs) were further analyzed through enrichment analysis and ROC curve evaluation to identify potential diagnostic biomarkers. A nomogram predictive model based on diagnostic genes was subsequently constructed and validated to support clinical decision-making. Additionally, single-sample gene set enrichment analysis (ssGSEA) was conducted to explore the relationship between diagnostic genes and immune cell infiltration. Finally, the findings were validated in an independent dataset. This study aims to identify potential LMRGs with diagnostic value for RA, providing new therapeutic insights for RA management.\u003c/p\u003e"},{"header":"2 Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Data acquisition and processing\u003c/h2\u003e\u003cp\u003eFive datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were retrieved from the GEO database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Among them, GSE206848 and GSE77298 were used as the training set, while GSE55235, GSE55584, and GSE55235 served as the validation set. To ensure analytical accuracy, batch effect removal was performed before merging the datasets, and datasets with substantial batch differences were excluded to maintain data consistency and compatibility. The \u0026ldquo;sva\u0026rdquo; R package was used to integrate and normalize the datasets. Additionally, lipid metabolism-related gene sets were obtained from the MigSB database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/index.jsp\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/index.jsp\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Differentially expressed genes\u003c/h2\u003e\u003cp\u003eTo identify DEGs between RA and healthy samples, the \u0026ldquo;limma\u0026rdquo; R package was used. The selection criteria were set to P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log fold change (FC)| \u0026gt;1.5. The \u0026ldquo;ggplot2\u0026rdquo; R package was then employed to generate a volcano plot for visualization.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e\u003cb\u003e2.3 WGCNA and module Identification\u003c/b\u003e\u003c/h2\u003e\u003cp\u003eData from GSE206848 (control\u0026thinsp;=\u0026thinsp;7, RA\u0026thinsp;=\u0026thinsp;2) and GSE77298 (control\u0026thinsp;=\u0026thinsp;7, RA\u0026thinsp;=\u0026thinsp;16) are merged and batch-corrected. WGCNA is performed to identify trait-associated modules. A topological overlap matrix is constructed based on expression profiles. The soft-threshold power is set to 5, and the minimum module size is set to 50 to filter core modules. A height cutoff of 0.25 is applied as a guideline for module merging. Pearson correlation analysis is then conducted to assess module significance, with a significance threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Functional enrichment analyses\u003c/h2\u003e\u003cp\u003eGene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses are performed using the \u0026ldquo;clusterProfiler\u0026rdquo; R package. GO annotation includes biological process (BP), cellular component (CC), and molecular function (MF) categories. A P-value threshold of 0.05 is applied for filtering, with P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 considered statistically significant. The enrichment analysis results are visualized using bubble plots.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Machine learning‑based hub gene screening\u003c/h2\u003e\u003cp\u003eMachine learning is an artificial intelligence technique that identifies patterns from data using models and algorithms, optimizing predictions and improving performance. In the medical field, machine learning models have widespread applications. In this study, DEGs identified from differential analysis are intersected with RA-associated gene modules obtained from WGCNA to generate a set of candidate genes. These candidate genes are then further intersected with lipid metabolism-related gene sets extracted from the MigSB database to identify key genes. For the key genes obtained from this intersection, the \u0026ldquo;glmnet\u0026rdquo; R package in R is used to perform LASSO-Cox regression, integrating variables and conducting survival analysis to derive the first gene module. Simultaneously, the randomForest function is employed to construct a random forest model, yielding the second gene module. Finally, the VennDiagram package in R is used to intersect the two modules, identifying four hub genes.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Screening of hub genes for its\u0026rsquo;diagnostic ability\u003c/h2\u003e\u003cp\u003eFor the four hub genes identified through machine learning algorithms, the \u0026ldquo;pROC\u0026rdquo; R package is used to generate ROC curves. The area under the curve (AUC) is calculated to evaluate the diagnostic performance of these genes.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Immune infiltration analysis and Gene Set Enrichment Analysis (GSEA)\u003c/h2\u003e\u003cp\u003eTo compare immune infiltration between RA and healthy samples, ssGSEA is performed using the \u0026ldquo;GSVA\u0026rdquo; R package to assess the relative abundance of 28 immune cell subsets. The \u0026ldquo;ggpubr\u0026rdquo; R package is used to visualize differences in immune infiltration, while the \u0026ldquo;corrplot\u0026rdquo; R package is applied to calculate Pearson correlation coefficients for each immune cell type and to evaluate the correlation between identified hub genes and immune cells. Additionally, ssGSEA analysis is conducted specifically for the hub genes.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.8 Establishment of Nomogram model\u003c/h2\u003e\u003cp\u003eTo assess the relationship between diagnostic genes and RA risk estimation, a nomogram model is constructed using the rms package in R and visualized with the plot(nom) function. A calibration curve is used to evaluate the predictive accuracy of the nomogram model, while decision curve analysis (DCA) is performed to assess its clinical applicability.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.9 Validation results of the validation set\u003c/h2\u003e\u003cp\u003eThe expression matrices of GSE55235, GSE55584, and GSE55235 are extracted using R, and the datasets are merged after excluding outlier samples. The Wilcoxon rank-sum test is performed to validate the differential expression of these biomarker candidates. The results are visualized using box plots. Hub genes that exhibit statistically significant differential expression are identified as key genes.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Identification of DEGs and WGCNA to Identify Lipid Metabolism-Related Cross Genes\u003c/h2\u003e\u003cp\u003eResearch Flowchart (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), After integrating, standardizing, and normalizing the datasets, a total of 1,938 DEGs were identified across 14 healthy controls and 18 RA samples, including 1,250 upregulated and 688 downregulated genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). WGCNA was performed on the integrated dataset, selecting a soft threshold of 5 for the scale-free network (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Based on this optimal threshold, a co-expression network was constructed, and a gene clustering heatmap was generated (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Notably, the module correlation heatmap revealed that the MEturquoise module was identified as the hub module (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), which contains 1,312 genes. The gene distribution results of the MEturquoise module showed a high correlation between RA and the module members, indicating that genes in this module are significantly associated with RA (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). The top three correlated module genes were intersected with the DEGs, and a Venn diagram was generated to display the overlap: 933 cross genes were identified between the DEGs and the WGCNA dataset. LMRGs were obtained from the MigSB database and intersected with the above cross genes, resulting in the identification of 29 intersections genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.2 GO and KEGG Enrichment of LMRGs\u003c/h2\u003e\u003cp\u003eTo explore the potential biological functions and signaling pathways of the identified intersections genes, GO and KEGG enrichment analyses were performed on the 29 intersections genes. The GO analysis results revealed that the top 10 enriched BP were primarily associated with lipid catabolic processes, fatty acid metabolism, phospholipid metabolism, glycerolipid biosynthesis, metabolism, glycerophospholipid metabolism, long-chain fatty acid metabolism, sphingolipid catabolic processes, and membrane lipid degradation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The top 7 enriched CC were located in lysosomal lumens, vascular lumens, major lysosomes, azurophilic granules, phosphoinositide 3-kinase complex, peroxisomal matrix, and microtubule lumens. The top 10 MF were mainly related to phosphatase activity, lipase activity, hydrolase activity, phosphatidylinositol 3-kinase activity, acylglycerol O-acyltransferase activity, O-acyltransferase activity, hydrogenase activity, 1-phosphatidylinositol-3-kinase activity, phosphoinositide phosphatase activity, and phosphoinositide 3-kinase activity. KEGG pathway enrichment analysis showed that the top 15 enriched pathways were primarily involved in sphingolipid metabolism, inositol phosphate metabolism, lysosomes, phosphoinositide signaling system, glycerophospholipid metabolism, platelet activation, biosynthesis of unsaturated fatty acids, ether lipid metabolism, ovarian steroidogenesis, fatty acid metabolism, regulation of adipocyte lipolysis, arachidonic acid metabolism, GnRH secretion, FcεRI signaling pathway, and prolactin signaling pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). These results provide valuable insights into the potential biological roles and pathways associated with lipid metabolism in RA and may offer new therapeutic targets.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Identification of Hub Genes\u003c/h2\u003e\u003cp\u003eUsing the LASSO Cox regression algorithm, five genes were selected from the intersection genes: INPP5D, PLD3, PIK3CD, NEU1, and CYP2C19 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-B). The RF method identified 10 feature genes: HSD17B4, FDXR, NR1H3, CYP2C19, PLBD1, TNFAIP8L2, PLD3, PIK3CD, NEU1, and ABCC3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC-D). The overlap of feature genes from both methods resulted in four hub genes: PLD3, PIK3CD, NEU1, and CYP2C19 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). These four overlapping genes are considered key candidates for further investigation as potential biomarkers for RA.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.4 ROC Curve Analysis\u003c/h2\u003e\u003cp\u003eTo validate the diagnostic potential of the four feature genes, we plotted ROC curves and compared their AUC values to assess their diagnostic value. The AUC values for PLD3, PIK3CD, NEU1, and CYP2C19 were 0.813, 0.857, 0.830, and 0.790, respectively, indicating moderate diagnostic value, with PIK3CD showing the highest diagnostic potential (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-D). Additionally, a combined ROC curve analysis of the hub genes revealed an AUC value of 0.913, significantly higher than the individual gene AUCs, suggesting that the model as a whole is more effective for disease prediction (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). Furthermore, we also validated the hub genes associated with ferroptosis, as identified by Yihua Fan et al\u003csup\u003e[18]\u003c/sup\u003e, The AUC values for VEGFA, PTGS2, and JUN were 0.591, 0.567, and 0.631, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). These results indicate that our identified hub genes provide higher diagnostic value compared to previously identified ferroptosis-related genes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Nomogram and Immune Infiltration Correlation Analysis\u003c/h2\u003e\u003cp\u003eA Nomogram model was constructed based on the identified hub genes to evaluate the contribution of each gene to the risk of developing RA (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Calibration curves were employed to assess the model, and the results demonstrated that the four hub genes closely aligned with the ideal model (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB-C). Subsequently, the correlation of immune cell infiltration was further analyzed using the Timer CellMarker database. The findings indicated that the four key genes were likely negatively correlated with the activation of several immune cell types, including B cells, CD4\u0026thinsp;+\u0026thinsp;T cells, CD8\u0026thinsp;+\u0026thinsp;T cells, dendritic cells, central memory CD8\u0026thinsp;+\u0026thinsp;T cells, myeloid-derived suppressor cells, regulatory T cells, follicular T cells, Th1 cells, memory B cells, macrophages, and γδT cells. In contrast, positive correlations were observed with effector memory CD4\u0026thinsp;+\u0026thinsp;T cells and immature dendritic cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). Furthermore, the correlation between the four hub genes and various immune checkpoints was assessed (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE). The analysis revealed that: PLD3 was predominantly upregulated in immature dendritic cells and CD56bright NK cells. PIK3CD was highly expressed in Th1 cells, regulatory T cells, monocytes, myeloid-derived suppressor cells, macrophages, immature B cells, γδT cells, eosinophils, central memory CD4\u0026thinsp;+\u0026thinsp;T cells, and activated CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cells, as well as dendritic cells. NEU1 was predominantly expressed in immature dendritic cells, and its expression was upregulated in memory B cells, myeloid-derived suppressor cells, immature B cells, γδT cells, central memory CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cells, and NK cells. CYP2C19 exhibited downregulation in most of the immune checkpoints where the expression of the other three genes was upregulated. These results suggest that these key genes play a significant role in modulating the immune landscape of RA and may contribute to immune cell modulation in the disease.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.6 GSEA Analysis\u003c/h2\u003e\u003cp\u003eTo further investigate the role of the hub genes in disease, single-gene GSEA analysis was performed for the four hub genes. In-depth analysis revealed that genes such as PLD3 were upregulated in pathways related to glutathione metabolism, lysosome, and viral myocarditis (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). PIK3CD exhibited high expression in pathways associated with allograft rejection, IgA in the intestinal immune network, and systemic lupus erythematosus (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB). In contrast, NEU1 was downregulated in pathways related to the hematopoietic cell lineage, lysosome, and Vibrio cholerae infection (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eC). CYP2C19 was upregulated in pathways involving DNA replication, proteasome, and systemic lupus erythematosus (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e3.7 Validation of Results in the Validation Set\u003c/h2\u003e\u003cp\u003eThe expression of the four RA lipid metabolism hub genes in the validation set was visualized. Ultimately, the expression difference of PIK3CD was statistically significant and consistent with the training set, showing upregulation in RA (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eA-D).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eRA is a chronic autoimmune disease that involves multiple bilateral joints. Its clinical characteristics include tenosynovitis, synovial tissue proliferation, vascular proliferation, cartilage destruction, and bone erosion[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The hallmark of the disease is the presence of various autoantibodies, such as rheumatoid factor (RF), anti-citrullinated protein antibodies (ACPA), and antibodies against post-translationally modified proteins, such as carbamylated proteins (anti-CarP antibodies). These autoantibodies can form immune complexes within the joints, activating the immune system and triggering aggressive synovial inflammation and cartilage destruction, ultimately leading to irreversible joint deformities or disabilities[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Therefore, exploring novel diagnostic biomarkers and targeted therapeutic strategies holds significant clinical value for early intervention in RA. Recent studies have revealed that lipid metabolism dysregulation plays a role in the pathological progression of RA by regulating the inflammatory microenvironment and oxidative stress responses. Notably, lipid peroxidation, as a key inducer of ferroptosis, may exacerbate joint damage by activating the ferroptosis pathway in synovial fibroblasts[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. However, the specific molecular mechanisms underlying lipid metabolism dysregulation and RA progression remain unclear, which serves as a critical focus of this study.\u003c/p\u003e\u003cp\u003eBased on the above background, this study identified 1,938 DEGs by comparing synovial tissue samples from the RA group and the control group. Furthermore, a gene co-expression network was constructed using the WGCNA approach to identify the top three gene modules most strongly associated with the RA phenotype. By integrating DEGs, lipid metabolism-related gene sets, and core gene modules, 29 candidate risk genes associated with lipid metabolism in RA were ultimately identified.\u003c/p\u003e\u003cp\u003eTo analyze the biological functions of the candidate genes, we performed KEGG and GO enrichment analyses on the 29 identified genes. These genes were found to be associated with lipid catabolic processes, fatty acid metabolism, phosphate metabolism, phospholipid metabolism, and triglyceride biosynthesis in biological processes. In terms of cellular components, these genes are primarily linked to the lysosomal lumen, blood vessel lumen, major lysosomes, azurophilic granules, and phosphoinositide 3-kinase complex. In molecular function, they are mainly involved in phosphatase activity, lipase activity, hydrolase activity, phosphatidylinositol kinase activity, and acylglycerol O-acyltransferase activity. These findings suggest that these genes may influence the progression of RA by regulating the energy metabolism imbalance in synovial cells. Notably, significant heterogeneity in the fatty acid profile has been observed in the synovial fluid (SF) of RA patients[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. The ratio of arachidonic acid (AA) to docosahexaenoic acid (DHA) is higher in the shoulder joint SF, while oleic acid predominates in the knee joint SF[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], This spatially specific lipid distribution may influence disease progression by altering joint lubrication, regulating synovial inflammation, and affecting cartilage degradation enzyme activity.\u003c/p\u003e\u003cp\u003eTo further identify the core regulatory factors, we integrated Lasso regression and random forest machine learning algorithms, ultimately determining PLD3, PIK3CD, NEU1, and CYP2C19 as key RA lipid metabolism-related hub genes. The nomogram model based on these four genes demonstrated excellent predictive performance in the training set (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.8), confirming their potential as RA diagnostic biomarkers. PLD3, a member of the phospholipase D family, encodes a 5'-3' nucleic acid exonuclease. It participates in phospholipid metabolism, affecting cell membrane lipid composition and signal transduction, and can contribute to inflammatory responses via lipid metabolism-related pathways, such as NF-κB\u003csup\u003e[26]\u003c/sup\u003e. PIK3CD encodes a subunit of phosphoinositide 3-kinase and plays a role in converting phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP3), thereby activating downstream signaling pathways like AKT/mTOR\u003csup\u003e[27]\u003c/sup\u003e, Previous studies have shown that inhibiting the PIK3CD pathway reduces autoreactive B cells and autoantibody levels, significantly improving inflammation in autoimmune diseases\u003csup\u003e[28]\u003c/sup\u003e. NEU1, an enzyme that hydrolyzes sialic acid on the cell surface, has been shown to activate AMPKα and subsequently SIRT3, regulating fibrosis, inflammation, apoptosis, and oxidative stress in diabetic cardiac tissue\u003csup\u003e[29]\u003c/sup\u003e. CYP2C19, a member of the CYP450 enzyme family, is closely related to drug metabolism in the body\u003csup\u003e[30]\u003c/sup\u003e. Research indicates that upregulation of CYP2C19 is associated with decreased C-reactive protein levels in patients with invasive aspergillosis treated with voriconazole\u003csup\u003e[31]\u003c/sup\u003e. To further assess the clinical value of lipid metabolism hub genes, we compared their ROC curves with those of the ferroptosis-related hub genes reported by the Yihua Fan team\u003csup\u003e[18]\u003c/sup\u003e. The results indicated that the lipid metabolism hub genes identified in this study had higher diagnostic efficacy than the ferroptosis-related genes. This finding suggests that lipid metabolism, as a chronic mechanism continuously regulating the energy homeostasis of the synovial microenvironment, may be more deeply involved in the pathological progression of RA than acute cell death pathways like ferroptosis. We speculate that lipid metabolism reprogramming, through long-term effects on immune cell activation, synovial cell proliferation, and inflammatory mediator release, plays a more central regulatory role in the development and progression of RA.\u003c/p\u003e\u003cp\u003eGiven that immune dysregulation is a core pathological feature of RA\u003csup\u003e[32]\u003c/sup\u003e, we further analyzed the relationship between hub genes and immune cell infiltration. The results indicated a significant increase in the proportion of macrophages, CD8⁺ T cells, and Treg cells in RA synovial tissue. Additionally, PLD3, PIK3CD, and NEU1 showed positive correlations with immune cell infiltration, whereas CYP2C19 exhibited a negative correlation. Notably, the exonuclease activity of PLD3 plays a crucial role in regulating immune cell inflammatory responses. When PLD3 function is impaired, the accumulation of ssDNA in the lysosome triggers downstream inflammatory signaling pathways (such as TLR9 and cGAS-STING), which in turn activate pro-inflammatory factors (e.g., IFN-α and TNF-α)\u003csup\u003e[26]\u003c/sup\u003e. Furthermore, activation of PIK3CD leads to abnormal activation and differentiation of CD4⁺ T cells, upregulating the expression of Th1 cells and suppressing the generation of Treg cells\u003csup\u003e[33]\u003c/sup\u003e. Further research revealed that gain-of-function mutations in PIK3CD resulted in an increase in memory T cells and follicular helper T cells (TFH), but with dysfunction, as indicated by increased expression of PD-1, CXCR3, and IFN-γ\u003csup\u003e[34]\u003c/sup\u003e. On the other hand, NEU1 deficiency leads to macrophage polarization towards the M2 phenotype, disrupting the balance between M1 and M2 macrophages\u003csup\u003e[35]\u003c/sup\u003e. Finally, CYP2C19 is highly expressed in M2 macrophages and, through metabolism, generates 11,12- and 14,15-epoxyeicosatrienoic acids (EETs). These metabolites act as PPARγ receptor agonists, promoting M2 macrophage polarization. This polarization helps regulate inflammatory responses and reduces the secretion of pro-inflammatory factors, thus exerting anti-inflammatory effects\u003csup\u003e[36]\u003c/sup\u003e. The above results demonstrate the significant role of these four hub genes in RA progression. Importantly, these genes are closely associated with immune cell dysfunction in immune infiltration, suggesting their potential as adjuncts for RA diagnosis and to mitigate immune responses during adjunctive therapy.\u003c/p\u003e\u003cp\u003eFinally, we integrated three external datasets for external validation, and the results demonstrated that PIK3CD is involved in the lipid metabolism process of RA, indicating its potential as a valuable diagnostic biomarker. Additionally, it is closely associated with lipid metabolism-related pathways, showing further potential as a therapeutic target. However, more experiments are needed to validate the detailed mechanisms and functions of PIK3CD.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study demonstrated through bioinformatics analysis that the central gene PIK3CD exhibits differential expression between RA patients and healthy individuals, suggesting its potential as an immune-related biomarker and therapeutic target for RA. This provides a new perspective and direction for future in-depth research and treatment strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthors\u003c/strong\u003e\u003cstrong\u003e'\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eContributions\u003c/strong\u003e: Z.F, X-PL and L-YT were responsible for writing first drafts and revising manuscripts. W.L, L-HH, W.W, K.Z, H-XX were responsible for finding literature. D-BL and B.W provided ideas for the work and revised it critically.\u0026nbsp;All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eFunding: This work was supported by the Key Fund Project of Hunan Provincial Department of Education (Grant No.23A0340), the Natural Science Foundation of Hunan Province (Grant No.2025JJ80606 and Grant No. 2023JJ50098), the Scientific Research Project of Health Commission of Hunan Province (Grant No. D202309016796) and Students’\u0026nbsp;platform for innovation and entrepreneurship training program (Grant No D202405241030541188)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of Interest Statement:\u003c/strong\u003e The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics statement\u003c/strong\u003e\u003cstrong\u003e：\u003c/strong\u003e\u003cstrong\u003eNot applicable.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number: not applicable.\u003c/strong\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eGravallese EM, Firestein GS: Rheumatoid Arthritis - Common Origins, Divergent Mechanisms. \u003cem\u003eN Engl J Med\u003c/em\u003e 2023, 388:529\u0026ndash;542.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1056/NEJMra2103726\u003c/span\u003e\u003cspan address=\"10.1056/NEJMra2103726\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang F, Palmer N, Fox K, Liao KP, Yu KH, Kou SC: Large-scale real-world data analyses of cancer risks among patients with rheumatoid arthritis. \u003cem\u003eInt J Cancer\u003c/em\u003e 2023, 153:1139\u0026ndash;1150.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ijc.34606\u003c/span\u003e\u003cspan address=\"10.1002/ijc.34606\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrown P, Pratt AG, Hyrich KL: Therapeutic advances in rheumatoid arthritis. \u003cem\u003eBmj\u003c/em\u003e 2024, 384:e070856\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1136/bmj-2022-070856\u003c/span\u003e\u003cspan address=\".10.1136/bmj-2022-070856\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSimon D, Minopoulou I, Kemenes S, Bayat S, Tascilar K, Mutlu MY, Valor-M\u0026eacute;ndez L, Kr\u0026ouml;nke G, Hueber AJ, Schett G, Kleyer A: Baricitinib Improves Bone Properties and Biomechanics in Patients With Rheumatoid Arthritis: Results of the Prospective Interventional BARE BONE Trial. \u003cem\u003eArthritis Rheumatol\u003c/em\u003e 2023, 75:1923\u0026ndash;1934.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/art.42617\u003c/span\u003e\u003cspan address=\"10.1002/art.42617\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Chen L, Chen Z, Liu M, Li X, Kou Y, Hou M, Wang H, Li X, Tian B, Dong J: Multifunctional Janus Nanoplatform for Efficiently Synergistic Theranostics of Rheumatoid Arthritis. \u003cem\u003eACS Nano\u003c/em\u003e 2023, 17:8167\u0026ndash;8182.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acsnano.2c11777\u003c/span\u003e\u003cspan address=\"10.1021/acsnano.2c11777\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIglesias-Artola JM, Nadler A: The Road to Quantitative Lipid Biochemistry in Living Cells. \u003cem\u003eAcc Chem Res\u003c/em\u003e 2023, 56:810\u0026ndash;820.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acs.accounts.2c00804\u003c/span\u003e\u003cspan address=\"10.1021/acs.accounts.2c00804\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan Y, Li P, Li J, Zhao Q, Chang Y, He X: Protein lipidation in health and disease: molecular basis, physiological function and pathological implication. \u003cem\u003eSignal Transduct Target Ther\u003c/em\u003e 2024, 9:60.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41392-024-01759-7\u003c/span\u003e\u003cspan address=\"10.1038/s41392-024-01759-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMartini T, Gobet C, Salati A, Blanc J, Mookhoek A, Reinehr M, Knott G, Sordet-Dessimoz J, Naef F: A sexually dimorphic hepatic cycle of periportal VLDL generation and subsequent pericentral VLDLR-mediated re-uptake. \u003cem\u003eNat Commun\u003c/em\u003e 2024, 15:8422.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-024-52751-2\u003c/span\u003e\u003cspan address=\"10.1038/s41467-024-52751-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSinturel F, Chera S, Brulhart-Meynet MC, Montoya JP, Stenvers DJ, Bisschop PH, Kalsbeek A, Guessous I, Jornayvaz FR, Philippe J, et al: Circadian organization of lipid landscape is perturbed in type 2 diabetic patients. \u003cem\u003eCell Rep Med\u003c/em\u003e 2023, 4:101299.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.xcrm.2023.101299\u003c/span\u003e\u003cspan address=\"10.1016/j.xcrm.2023.101299\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLuo YX, Yang LL, Yao XQ: Gut microbiota-host lipid crosstalk in Alzheimer's disease: implications for disease progression and therapeutics. \u003cem\u003eMol Neurodegener\u003c/em\u003e 2024, 19:35.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13024-024-00720-0\u003c/span\u003e\u003cspan address=\"10.1186/s13024-024-00720-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang Y, Luo D, Shao Y, Shan Z, Liu Q, Weng J, He W, Zhang R, Li Q, Wang Z, Li X: circCAPRIN1 interacts with STAT2 to promote tumor progression and lipid synthesis via upregulating ACC1 expression in colorectal cancer. \u003cem\u003eCancer Commun (Lond)\u003c/em\u003e 2023, 43:100\u0026ndash;122.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/cac2.12380\u003c/span\u003e\u003cspan address=\"10.1002/cac2.12380\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLei Q, Yang J, Li L, Zhao N, Lu C, Lu A, He X: Lipid metabolism and rheumatoid arthritis. \u003cem\u003eFront Immunol\u003c/em\u003e 2023, 14:1190607.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fimmu.2023.1190607\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2023.1190607\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFrommer KW, Sch\u0026auml;ffler A, Rehart S, Lehr A, M\u0026uuml;ller-Ladner U, Neumann E: Free fatty acids: potential proinflammatory mediators in rheumatic diseases. \u003cem\u003eAnn Rheum Dis\u003c/em\u003e 2015, 74:303\u0026ndash;310.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/annrheumdis-2013-203755\u003c/span\u003e\u003cspan address=\"10.1136/annrheumdis-2013-203755\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWei J, Huang X, Zhang X, Chen G, Zhang C, Zhou X, Qi J, Zhang Y, Li X: Elevated fatty acid β-oxidation by leptin contributes to the proinflammatory characteristics of fibroblast-like synoviocytes from RA patients via LKB1-AMPK pathway. \u003cem\u003eCell Death Dis\u003c/em\u003e 2023, 14:97.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41419-023-05641-2\u003c/span\u003e\u003cspan address=\"10.1038/s41419-023-05641-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen K, Yan Y, Li C, Yuan J, Wang F, Huang P, Qian N, Qi J, Zhou H, Zhou Q, et al: Increased 15-lipoxygenase-1 expression in chondrocytes contributes to the pathogenesis of osteoarthritis. \u003cem\u003eCell Death Dis\u003c/em\u003e 2017, 8:e3109\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1038/cddis.2017.511\u003c/span\u003e\u003cspan address=\".10.1038/cddis.2017.511\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiyabe Y, Miyabe C, Luster AD: LTB(4) and BLT1 in inflammatory arthritis. \u003cem\u003eSemin Immunol\u003c/em\u003e 2017, 33:52\u0026ndash;57.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.smim.2017.09.009\u003c/span\u003e\u003cspan address=\"10.1016/j.smim.2017.09.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu S, Lu H, Lin J, Chen Z, Jiang D: Regulation of TNFalpha and IL1beta in rheumatoid arthritis synovial fibroblasts by leukotriene B4. \u003cem\u003eRheumatol Int\u003c/em\u003e 2010, 30:1183\u0026ndash;1189.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00296-009-1125-y\u003c/span\u003e\u003cspan address=\"10.1007/s00296-009-1125-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan Y, Li Y, Fu X, Peng J, Chen Y, Chen T, Zhang D: Identification of potential ferroptosis key genes and immune infiltration in rheumatoid arthritis by integrated bioinformatics analysis. \u003cem\u003eHeliyon\u003c/em\u003e 2023, 9:e21167\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1016/j.heliyon.2023.e21167\u003c/span\u003e\u003cspan address=\".10.1016/j.heliyon.2023.e21167\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLin YJ, Anzaghe M, Sch\u0026uuml;lke S: Update on the Pathomechanism, Diagnosis, and Treatment Options for Rheumatoid Arthritis. \u003cem\u003eCells\u003c/em\u003e 2020, 9.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/cells9040880\u003c/span\u003e\u003cspan address=\"10.3390/cells9040880\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJang S, Kwon EJ, Lee JJ: Rheumatoid Arthritis: Pathogenic Roles of Diverse Immune Cells. \u003cem\u003eInt J Mol Sci\u003c/em\u003e 2022, 23.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijms23020905\u003c/span\u003e\u003cspan address=\"10.3390/ijms23020905\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePertsinidou E, Saevarsdottir S, Manivel VA, Klareskog L, Alfredsson L, Mathsson-Alm L, Hansson M, Cornillet M, Serre G, Holmdahl R, et al: In early rheumatoid arthritis, anticitrullinated peptide antibodies associate with low number of affected joints and rheumatoid factor associates with systemic inflammation. \u003cem\u003eAnn Rheum Dis\u003c/em\u003e 2024, 83:277\u0026ndash;287.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/ard-2023-224728\u003c/span\u003e\u003cspan address=\"10.1136/ard-2023-224728\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheng Q, Chen M, Liu M, Chen X, Zhu L, Xu J, Xue J, Wu H, Du Y: Semaphorin 5A suppresses ferroptosis through activation of PI3K-AKT-mTOR signaling in rheumatoid arthritis. \u003cem\u003eCell Death Dis\u003c/em\u003e 2022, 13:608.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41419-022-05065-4\u003c/span\u003e\u003cspan address=\"10.1038/s41419-022-05065-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMustonen AM, K\u0026auml;kel\u0026auml; R, Lehenkari P, Huhtakangas J, Turunen S, Joukainen A, K\u0026auml;\u0026auml;ri\u0026auml;inen T, Paakkonen T, Kr\u0026ouml;ger H, Nieminen P: Distinct fatty acid signatures in infrapatellar fat pad and synovial fluid of patients with osteoarthritis versus rheumatoid arthritis. \u003cem\u003eArthritis Res Ther\u003c/em\u003e 2019, 21:124.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13075-019-1914-y\u003c/span\u003e\u003cspan address=\"10.1186/s13075-019-1914-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMustonen AM, K\u0026auml;kel\u0026auml; R, Joukainen A, Lehenkari P, Jaroma A, K\u0026auml;\u0026auml;ri\u0026auml;inen T, Kr\u0026ouml;ger H, Paakkonen T, Sihvo SP, Nieminen P: Synovial Fluid Fatty Acid Profiles Are Differently Altered by Inflammatory Joint Pathologies in the Shoulder and Knee Joints. \u003cem\u003eBiology (Basel)\u003c/em\u003e 2021, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.https://doi.org/10.3390/biology10050401\u003c/span\u003e\u003cspan address=\"10.10.3390/biology10050401\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang N, Zhao X, Huai J, Li Y, Cheng C, Bi K, Dai R: Arachidonic acid metabonomics study for understanding therapeutic mechanism of Huo Luo Xiao Ling Dan on rat model of rheumatoid arthritis. \u003cem\u003eJ Ethnopharmacol\u003c/em\u003e 2018, 217:205\u0026ndash;211.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jep.2018.02.027\u003c/span\u003e\u003cspan address=\"10.1016/j.jep.2018.02.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGavin AL, Huang D, Huber C, M\u0026aring;rtensson A, Tardif V, Skog PD, Blane TR, Thinnes TC, Osborn K, Chong HS, et al: PLD3 and PLD4 are single-stranded acid exonucleases that regulate endosomal nucleic-acid sensing. \u003cem\u003eNat Immunol\u003c/em\u003e 2018, 19:942\u0026ndash;953.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41590-018-0179-y\u003c/span\u003e\u003cspan address=\"10.1038/s41590-018-0179-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang X, Han H, Shi X, Nie X, Zhu R, Jin J, Zhou H: Genetic insights of blood lipid metabolites on polycystic ovary syndrome risk: a bidirectional two-sample Mendelian randomization study. \u003cem\u003eFront Endocrinol (Lausanne)\u003c/em\u003e 2024, 15:1391826.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fendo.2024.1391826\u003c/span\u003e\u003cspan address=\"10.3389/fendo.2024.1391826\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScuron MD, Fay BL, Connell AJ, Oliver J, Smith PA: The PI3Kδ inhibitor parsaclisib ameliorates pathology and reduces autoantibody formation in preclinical models of systemic lupus erythematosus and Sjӧgren's syndrome. \u003cem\u003eInt Immunopharmacol\u003c/em\u003e 2021, 98:107904.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.intimp.2021.107904\u003c/span\u003e\u003cspan address=\"10.1016/j.intimp.2021.107904\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo Z, Tuo H, Tang N, Liu FY, Ma SQ, An P, Yang D, Wang MY, Fan D, Yang Z, Tang QZ: Neuraminidase 1 deficiency attenuates cardiac dysfunction, oxidative stress, fibrosis, inflammatory via AMPK-SIRT3 pathway in diabetic cardiomyopathy mice. \u003cem\u003eInt J Biol Sci\u003c/em\u003e 2022, 18:826\u0026ndash;840.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7150/ijbs.65938\u003c/span\u003e\u003cspan address=\"10.7150/ijbs.65938\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWiese MD, Schnabl M, O'Doherty C, Spargo LD, Sorich MJ, Cleland LG, Proudman SM: Polymorphisms in cytochrome P450 2C19 enzyme and cessation of leflunomide in patients with rheumatoid arthritis. \u003cem\u003eArthritis Res Ther\u003c/em\u003e 2012, 14:R163\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1186/ar3911\u003c/span\u003e\u003cspan address=\".10.1186/ar3911\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKlomp SD, Veringa A, Alffenaar JC, de Boer MGJ, Span LFR, Guchelaar HJ, Swen JJ: Inflammation altered correlation between CYP2C19 genotype and CYP2C19 activity in patients receiving voriconazole. \u003cem\u003eClin Transl Sci\u003c/em\u003e 2024, 17:e13887\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e.https://doi.org/10.1111/cts.13887\u003c/span\u003e\u003cspan address=\".10.1111/cts.13887\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang J, Wang C, Wu X, Shen Q, Du Y: Nanozyme-based therapeutic strategies for rheumatoid arthritis. \u003cem\u003eJ Control Release\u003c/em\u003e 2025, 377:716\u0026ndash;734.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jconrel.2024.11.072\u003c/span\u003e\u003cspan address=\"10.1016/j.jconrel.2024.11.072\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHanna BS, Roessner PM, Scheffold A, Jebaraj BMC, Demerdash Y, \u0026Ouml;zt\u0026uuml;rk S, Lichter P, Stilgenbauer S, Seiffert M: PI3Kδ inhibition modulates regulatory and effector T-cell differentiation and function in chronic lymphocytic leukemia. \u003cem\u003eLeukemia\u003c/em\u003e 2019, 33:1427\u0026ndash;1438.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41375-018-0318-3\u003c/span\u003e\u003cspan address=\"10.1038/s41375-018-0318-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBier J, Rao G, Payne K, Brigden H, French E, Pelham SJ, Lau A, Lenthall H, Edwards ESJ, Smart JM, et al: Activating mutations in PIK3CD disrupt the differentiation and function of human and murine CD4(+) T cells. \u003cem\u003eJ Allergy Clin Immunol\u003c/em\u003e 2019, 144:236\u0026ndash;253.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jaci.2019.01.033\u003c/span\u003e\u003cspan address=\"10.1016/j.jaci.2019.01.033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEscalona E, Olate-Briones A, Albornoz-Mu\u0026ntilde;oz S, Bonacic-Doric E, Rodr\u0026iacute;guez-Arriaza F, Herrada AA, Escobedo N: Neu1 deficiency and fibrotic lymph node microenvironment lead to imbalance in M1/M2 macrophage polarization. \u003cem\u003eFront Immunol\u003c/em\u003e 2024, 15:1462853.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fimmu.2024.1462853\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2024.1462853\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCho HY, Ahn S, Cho YS, Seo SK, Kim DH, Shin JG, Lee SJ: CYP2C19 Contributes to THP-1-Cell-Derived M2 Macrophage Polarization by Producing 11,12- and 14,15-Epoxyeicosatrienoic Acid, Agonists of the PPARγ Receptor. \u003cem\u003ePharmaceuticals (Basel)\u003c/em\u003e 2023, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e16.https://doi.org/10.3390/ph16040593\u003c/span\u003e\u003cspan address=\"16.10.3390/ph16040593\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Rheumatoid Arthritis, Lipid metabolism, Machine learning, Immune infiltration","lastPublishedDoi":"10.21203/rs.3.rs-7913208/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7913208/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cb\u003eBackground\u003c/b\u003e\u003c/p\u003e\u003cp\u003eRheumatoid arthritis (RA) is a common autoimmune inflammatory joint disease. Recent studies suggest that lipid metabolism dysregulation plays a crucial role in RA pathogenesis; however, its precise mechanisms remain unclear. This study aims to identify lipid metabolism-related diagnostic biomarkers in RA and analyze their potential pathogenic mechanisms and clinical significance.\u003c/p\u003e\u003cp\u003e\u003cb\u003eMethods\u003c/b\u003e\u003c/p\u003e\u003cp\u003eMultiple RA-related microarray datasets (GSE206848, GSE77298, GSE55235, GSE55584, and GSE55235) were obtained from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified using the \u0026ldquo;limma\u0026rdquo; R package. Weighted gene co-expression network analysis (WGCNA) was performed to identify key module genes associated with RA. Two machine learning algorithms, least absolute shrinkage and selection operator (LASSO) and random forest (RF), were used to screen hub genes closely related to synovial lipid metabolism in RA. A nomogram and receiver operating characteristic (ROC) curve were constructed to predict RA risk. Additionally, immune cell infiltration was analyzed, followed by single-sample gene set enrichment analysis (ssGSEA) and validation in an independent dataset.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResults\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePIK3CD was identified as a key gene with strong diagnostic value for RA. Abnormal immune cell infiltration was observed in RA patients and was positively correlated with PIK3CD expression.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConclusion\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThese findings suggest that PIK3CD is involved in RA-related lipid metabolism and may serve as a reliable diagnostic biomarker and potential therapeutic target for RA.\u003c/p\u003e","manuscriptTitle":"Identification of Lipid Metabolism-Related Genes in Rheumatoid Arthritis Using Bioinformatics and Machine Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-06 05:21:52","doi":"10.21203/rs.3.rs-7913208/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"aa8e03d5-20f6-4407-b324-e5d1f9de0aec","owner":[],"postedDate":"November 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-26T23:23:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-06 05:21:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7913208","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7913208","identity":"rs-7913208","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}