{"paper_id":"36d4bc72-8ba9-4b5a-b6ea-a34d2d3d51fe","body_text":"Exploring biomarkers related to exosome in primary immune thrombocytopenia based on transcriptomics and experimental verification | 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 Article Exploring biomarkers related to exosome in primary immune thrombocytopenia based on transcriptomics and experimental verification Fangfang Lou, Zhiyue Chen, Zihao Yuan, Jie Peng, Jiangyu Sun, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8092905/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Mar, 2026 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Background Primary Immune Thrombocytopenia (ITP) is an autoimmune disease with thrombocytopenia and bleeding tendency. Exosomes mediate abnormal crosstalk between immune cells and megakaryocytes in ITP, making exosome-related biomarkers crucial for the disease’s diagnosis and treatment. Methods ITP transcriptome data and exosome-related genes (ERGs) were retrieved from public databases. Candidate genes were identified by intersecting ITP’s DEGs with exosome-related key module genes, followed by biomarker screening via machine learning and nomogram construction. Multi-dimensional analyses (enrichment, immune infiltration, drug prediction) and RT-qPCR validation were performed. Results Four biomarkers (GABARAPL1, SLC39A14, HIBADH, GSR) were confirmed, involved in spliceosome and other pathways (P < 0.05, |NES| >1). GABARAPL1, SLC39A14, HIBADH negatively correlated with activated NK cells (|cor| >0.3, P < 0.05). SLC39A14/nortriptyline and GSR/oxiglutatione showed strong binding affinity (binding free energy < -5 kcal/mol). RT-qPCR verified dysregulation of GABARAPL1, SLC39A14, GSR in ITP patients (P < 0.05). Conclusion In this study, GABARAPL1, SLC39A14, HIBADH, and GSR were successfully screened as biomarkers, and their related regulatory mechanisms were also revealed, providing innovative theoretical support for the precise diagnosis and treatment of ITP. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Biological sciences/Immunology Primary immune thrombocytopenia Exosome Transcriptomics Biomarkers Immunity Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Primary immune thrombocytopenia (ITP), an acquired autoimmune hematologic condition, is defined by two key features: a significant decline in platelet levels and an increased propensity for bleeding[ 1 – 2 ]. Epidemiological studies report an annual incidence rate of 2–10 cases per 100,000 population[ 3 ]. ITP can occur at any age, with pediatric cases typically presenting as acute forms often secondary to infections, while adult cases predominantly manifest as chronic forms, showing a slightly higher incidence in females than males[ 4 – 5 ]. The pathogenesis of ITP involves complex mechanisms, with three primary contributing factors: autoantibody production against platelets, impaired megakaryocyte maturation, and T-lymphocyte dysfunction[ 6 – 7 ]. Current therapeutic strategies primarily employ corticosteroids and intravenous immunoglobulin to rapidly elevate platelet counts, though these interventions demonstrate limited long-term efficacy[ 8 – 9 ]. The diagnosis of ITP remains exclusion-based due to the absence of specific biomarkers, potentially leading to misdiagnosis in some cases and consequent exacerbation of bleeding risks. Therefore, the identification of novel biomarkers holds significant value for improving ITP diagnosis and treatment. Exosomes are nanosized extracellular vesicles containing nucleic acids, proteins, and lipid components that mediate intercellular communication and participate in diverse biological processes[ 10 – 12 ]. Emerging evidence demonstrates their pivotal role in autoimmune diseases including multiple sclerosis and systemic lupus erythematosus, where their expression levels serve as valuable biomarkers for disease progression while also exhibiting potential as drug delivery vehicles[ 13 – 14 ]. A seminal study revealed that exosomes derived from ITP patients carry miR-363-3p, which disrupts immune homeostasis by significantly impairing the immunosuppressive function of regulatory T cells through modulation of the TBX21/ARID3A/SPI1 signaling axis[ 15 ]. Furthermore, specific proteins and miRNAs within exosomes have been implicated in ITP pathogenesis[ 16 ]. These molecular components are important drivers of disease progression, yet their regulatory mechanisms are not fully understood and need more research. We utilized a multi-omics framework that combined transcriptome profiling and bioinformatics to discover ITP-related exosomal genetic signatures. Machine learning was subsequently employed to screen and validate diagnostic biomarkers. The comprehensive analytical framework encompasses clinical prediction modeling through nomogram construction, functional enrichment analysis of biological pathways, molecular network mapping to elucidate regulatory mechanisms, immune microenvironment characterization via infiltration profiling, and computational drug prediction with molecular docking simulations for therapeutic target exploration. Experimental validation was subsequently performed using reverse transcription quantitative polymerase chain reaction (RT-qPCR) on peripheral blood samples from ITP patients. This translational research paradigm aims to discover novel exosome-derived biomarkers with enhanced diagnostic precision, ultimately advancing the scientific foundation for personalized ITP management strategies. 2. Materials and methods 2.1 Data source In this study, the primary ITP-related dataset GSE43179 (sequencing type: microarray) was obtained from the GEO database ( https://www.ncbi.nlm.nih.gov/geo/ ). This dataset contains molecular expression data based on two different sequencing platforms. Among them, mRNA expression data were derived from the GPL570 platform, including 9 peripheral blood T-cell samples from ITP patients (ITP group) and 10 peripheral blood T-cell samples from normal controls (control group). miRNA expression data were obtained from the GPL14613 platform, comprising 9 peripheral blood T-cell samples from ITP patients (ITP group) and 9 peripheral blood T-cell samples from normal controls (control group). In addition, a total of 121 exosome-related genes (ERGs) were retrieved from the ExoBCD ( https://exobcd.liumwei.org/ ) and used for subsequent analyses ( Supplementary Table S1 )[ 17 ]. 2.2 Differential expression analysis For screening ITP-related differentially expressed genes (DEGs), differential expression analysis was initially performed on samples of the ITP and control groups in dataset GSE43179 (Sequencing platform: GPL570) via the R package \"limma\" (v 3.54.0)[ 18 ], identifying DEGs (P < 0.05, |log 2 fold change (FC)| >0.25). enerated for the DEGs using \"ggplot2\" (v 3.4.1)[ 20 ]. Visualization of the differential analysis included a heatmap and a volcano plot. The heatmap, created with \"pheatmap\" (v 1.0.12)[ 19 ], depicted DEG expression between ITP and control groups, with the top 10 upregulated and downregulated genes (by |log 2 FC|) highlighted. The volcano plot was generated using \"ggplot2\" (v 3.4.1)[ 20 ]. Genes were sorted identically by |log 2 FC| (descending), and the top 10 DEGs mentioned above were labeled in the plot. 2.3 WGCNA Subsequently, the ssGSEA and WGCNA were comprehensively employed to expand exosome-related genes. For visualization of the significant inter-group disparities (P < 0.05) in DE-ERG scores identified by the Wilcoxon rank-sum test, box plots were generated using the R package “ggplot2” (v 3.4.1)[ 20 ]. To further explore co-expression modules highly associated with exosomes, WGCNA was performed based on the aforementioned DE-ERG enrichment scores. The \"hclust\" function was employed for hierarchical clustering of all samples in dataset GSE43179 (sequencing platform: GPL570), aiming to identify and exclude outlier samples and guarantee data quality for subsequent co-expression network building. A sample clustering tree was also generated to visualize inter-sample relationships. Next, the optimal soft threshold (power value) was selected within the 1–20 parameter range. The best soft threshold that satisfied the scale-free network fitting index R² >0.85 was selected. A co-expression network was built based on the optimal soft threshold, and the dynamic tree cutting algorithm in the R package \"WGCNA\" (v 1.73)[ 21 ] was employed with the following parameter settings: minimum number of module genes set to 200, method = \"tree\", and cutHeight = 0.99, to divide genes into modules labeled with different colors. To identify modules highly correlated with exosome traits, the Spearman correlation coefficient between module eigengenes and DE-ERG scores was calculated (|correlation coefficient (cor)| >0.3 and P < 0.05). Meanwhile, a module-trait correlation heatmap was plotted using the labeledHeatmap function in the R package \"WGCNA\" (v 1.73)[ 21 ] for visualization. Finally, we selected the modules most strongly correlated with DE-ERG scores. Using the \"WGCNA\" R package (v 1.73)[ 21 ], Module Membership (MM) and Gene Significance (GS) were calculated for these module genes. Genes meeting the thresholds of |MM| >0.6 and |GS| >0.4 were retained. The GS-MM correlation was visualized in scatter plots via \"ggplot2\" (v 3.4.1)[ 21 ], and genes fulfilling these criteria were defined as exosome-related key module genes. 2.4 Acquisition of candidate genes and functional enrichment analysis We next sought to identify genes linked to both ITP and exosomes. Using the R package \"ggvenn\" (v 1.7.1), we found the intersection of DEGs and exosome-related key module genes, thereby acquiring candidate genes for follow-up functional verification. The R package \"clusterProfiler\" (v 4.2.2)[ 22 ] was employed to perform KEGG and GO enrichment analyses on the candidate genes (P < 0.05), aiming to explore their potential biological functions and underlying mechanisms. GO consisted of three parts, namely biological process (BP), cellular component (CC) and molecular function (MF). Additionally, STRING ( https://string-db.org/ ) was utilized to examine interactions of proteins encoded by candidate genes (confidence score > 0.15). Finally, the results were visualized by constructing a PPI network diagram using Cytoscape software (v 3.8.2)[ 23 ]. 2.5 Acquisition of biomarkers For the refined screening of diagnostically relevant ITP biomarkers, a two-step algorithm approach was utilized: Boruta followed by LASSO regression. Feature selection via the \"Boruta\" package (v 8.0.0)[ 24 ]on the GSE43179 dataset (GPL570) identified genes with importance scores exceeding all shadow feature benchmarks (shadowMin, shadowMean, shadowMax), yielding Feature Gene Set 1. Subsequent analysis involved applying LASSO regression to the same dataset with the \"glmnet\" package (v 1.1.10)[ 25 ]. Via 5-fold cross-validation, the optimal regularization parameter (lambda) (minimum mean squared error) was determined. Genes with non-zero regression coefficients were screened to generate Feature Gene Set 2, and regression coefficient plots plus cross-validation error curves were generated for visualization. Finally, the two feature gene sets’ intersection was calculated using \"ggvenn\" (v 1.7.1) to acquire the final biomarkers. 2.6 Construction and evaluation of nomogram To evaluate biomarkers’ ITP predictive reliability, we established a nomogram prediction model based on biomarkers using the R package \"rms\" (v 6.8.1)[ 26 ] and all samples from GSE43179 (Sequencing platform: GPL570). The scale range of the line corresponding to each biomarker indicates its predicted score interval (Points). The total score (Total Points) results from summing the scores of each biomarker, and a higher value of it corresponds to a higher probability of ITP developing. Predictive accuracy of the model was verified through two methods: calibration curves plotted using the \"rms\" R package (v 6.8.1)[ 26 ]and the Hosmer-Lemeshow test for assessing prediction-actuality consistency. In these curves, prediction accuracy was considered higher when the plot more closely followed the reference line (slope = 1). The nomogram demonstrated good calibration accuracy for ITP risk prediction, as evidenced by an HL test P value > 0.05, which denotes no significant difference between predicted and actual values. The predictive accuracy was further assessed by constructing an ROC curve with the \"pROC\" R package (v 1.18.5)[ 27 ]. AUC (range: 0–1) served as a discriminatory metric, with values above 0.7 deemed to indicate good predictive performance. 2.7 GSEA of biomarkers To investigate the regulatory pathways and biological functions linked to biomarkers, GSEA was initially performed for each biomarker on ITP vs. control samples from dataset GSE43179 (Sequencing platform: GPL570). Employing the R package \"psych\" (v 2.1.6)[ 28 ], we computed and sorted Spearman correlation coefficients between biomarkers and other genes. GSEA was subsequently performed with the R package \"clusterProfiler\" (v 4.2.2)[ 29 ] using the MsigDB ( https://www.gsea-msigdb.org/gsea/msigdb/ ) gene set \"c2.cp.kegg_legacy.v2024.1.Hs.symbols.gmt\" as reference, under the criteria |NES| >1 and adj.P < 0.05. Visualization of the top 5 significant pathways was finally achieved with the R package \"enrichplot\" (v 1.18.3)[ 30 ]. 2.8 GeneMANIA analysis and of biomarkers Subsequently, related genes interacting with biomarkers and their involved functions were predicted via the GeneMANIA database ( http://genemania.org/ ), with \"Homo sapiens\" set as the analysis species. From the outcomes, biomarkers and the top 20 related genes were selected as key display nodes, and the top 5 most significant enriched pathways (P < 0.05) and functions (FDR < 0.05) were presented. A gene interaction network diagram was ultimately constructed to visualize results, clearly showing complex correlations between biomarkers and their functionally similar genes. 2.9 Immune infiltration analysis ITP is an immune-mediated autoimmune disease, primarily marked by impaired immune system function. This impairment prompts autoantibodies or cytotoxic T cells to attack platelets, thereby boosting platelet destruction and hindering platelet production[ 31 – 32 ]. To evaluate differences in immune cell infiltration levels between the ITP group and the control group, based on all samples from the dataset GSE43179 (Sequencing platform: GPL570), the relative infiltration abundances of 22 immune cells were calculated using the R package \"CIBERSORT\" (v2.24.0)[ 33 ] in this study. Following sample filtration (P > 0.05) to ensure reliability, immune cell infiltration percentages were visualized in stacked bar charts. Differences in infiltration between groups were further identified using the Wilcoxon rank-sum test (P < 0.05) and illustrated in box plots, with both visualizations created using the R package \"ggplot2\" (v 3.4.1)[ 20 ]. Subsequent correlation analysis, conducted with the R package \"psych\" (v 2.1.6)[ 28 ], examined relationships between biomarkers and these differential immune cells, considering correlations with |cor| >0.3 and P < 0.05 as significant. 2.10 Molecular regulatory network analysis of biomarkers Transcription factors (TFs) are key protein molecules that can specifically recognize and bind to specific DNA sequences in regulatory regions such as gene promoters or enhancers, thereby regulating the transcriptional activity of genes. First, the TRRUST database ( https://link.zhihu.com/?target=http%3A//www.grnpedia.org/trrust/ ) was used to predict TFs that interact with biomarkers. MicroRNAs (miRNAs) primarily mediate the regulation of protein expression by binding to one or more target sites on mRNA transcripts and inhibiting translation. They are involved in various developmental pathways and multiple gene regulatory mechanisms, and play a role in a series of disease processes and phenotypic determination. Next, based on the GSE43179 dataset (sequencing platform: GPL14613), the R package \"limma\" (v 3.54.0)[ 18 ] was used to identify differentially expressed microRNAs (DE-miRNAs) between the ITP group and the control group. MiRNAs in the heatmap were sorted by |log 2 FC| values in descending order, showing the top 10 upregulated and 10 downregulated miRNAs with the most significant expression differences. A miRNA volcano plot was also created using the R package \"ggplot2\" (v 3.4.1)[ 20 ]. Similarly, genes were sorted by descending |log 2 FC|, with the plot labeling the top 10 up- and downregulated miRNAs (most significant expression differences). Subsequently, miRNet ( https://www.mirnet.ca/ ) predicted miRNAs targeting biomarkers. The final integrative visualization of the regulatory network, built from the screened TFs, key miRNAs, and biomarkers, was achieved using Cytoscape software (v 3.8.2)[ 23 ]. 2.11 Drug prediction analysis of biomarkers and molecular docking To identify potential ITP drugs, drug prediction for each biomarker was done via DGIdb ( https://dgidb.org/ ). Construction and visualization of the drug-mRNA interaction network were performed with Cytoscape software (v 3.8.2)[ 23 ]. To validate targeted drug-biomarker binding, we selected drugs per biomarker (highest interaction scores, PubChem-retrievable 2D structures) for molecular docking. Drug 2D structures (SDF) were downloaded from PubChem ( https://pubchem.ncbi.nlm.nih.gov/ ), and biomarker-encoded protein 3D structures (PDB) from RCSB PDB ( https://www.rcsb.org/ ). These were uploaded to CB-Dock ( https://cadd.labshare.cn/cb-dock2/php/index.php ) for docking, with binding affinity evaluated by Vina score (binding free energy). Generally, Vina score < -5 kcal/mol indicates high ligand-receptor affinity and stable complex conformation. 2.12 Expression level validation of biomarkers Differences in biomarker expression within the GSE43179 dataset (GPL570) were assessed using the Wilcoxon rank-sum test (P < 0.05) and subsequently visualized in box plots constructed with the R package \"ggplot2\" (v 3.4.1)[ 20 ]. 2.13 RT-qPCR Validation of biomarker expression in clinical samples was performed by RT-qPCR. Ten blood samples (5 from ITP patients and 5 from controls) were obtained from Fengdu County Hospital of Traditional Chinese Medicine. Total RNA was extracted using TRIzol reagent (Ambion, USA) per the manufacturer's protocol, and its concentration was determined with a NanoPhotometer N50. cDNA was synthesized by reverse transcription employing the SureScript First-Strand cDNA Synthesis Kit on a Bio-Rad S1000™ Thermal Cycler (USA). Using primers listed in Supplementary Table S2 , qPCR was run on a Bio-Rad CFX Connect Real-Time PCR Instrument (USA) with the cycling profile: 95°C for 1 min; 40 cycles of 95°C for 20s, 55°C for 20s, and 72°C for 30s. The 2-ΔΔCT method was applied for relative quantification, and data were analyzed and visualized in GraphPad Prism 5 ( https://www.graphpad.com/ ) after initial export to Excel. 2.14 Statistical analysis Bioinformatics analyses utilized R software (v 4.2.2) for statistical processing. Statistical significance was set at P < 0.05. The Wilcoxon rank-sum test was used for between-group comparisons, while the t-test was employed for between-group analyses of RT-qPCR results. 2.15 Ethics approval and consent to participate The human subjects research component of this study was reviewed and approved by the Ethics Committee of Fengdu County Hospital of Traditional Chinese Medicine, Chongqing, China (Approval No.: FDYY−2024−0729), with documented informed consent obtained from all participants. All procedures were conducted in strict compliance with the ethical principles set forth in the Declaration of Helsinki (1964) and its subsequent amendments. 3. Results 3.1 Acquisition and functional enrichment analysis of 23 candidate genes A total of 105 DEGs were identified from the GSE43179 dataset (GPL570) through a comparative analysis between ITP and control groups, using thresholds of P < 0.05 and |log 2 FC| >0.25. The ITP-associated DEGs comprised 48 upregulated and 57 downregulated genes (Figs. 1 A-B, Supplementary Table S3 ). Expression levels of 121 ERGs were then compared between ITP and control groups in the GSE43179 dataset (GPL570) via Wilcoxon rank-sum test. This screening identified seven DE-ERGs: POLR2K, IGF1R, NANOG, RAB13, ALDH9A1, HDC, and TLR5 (P < 0.05) (Fig. 1 C). Further, DE-ERG scores were calculated based on all samples from the dataset GSE43179 (Sequencing platform: GPL570). The Wilcoxon rank-sum test showed that there was a significant difference in DE-ERG scores between the ITP group and the control group (P = 0.011) (Fig. 1 D). No obvious outlier samples were identified by hierarchical clustering analysis, indicating that the data quality was reliable (Fig. 1 E). A scale-free co-expression network was successfully constructed by WGCNA with a soft-thresholding power of 19. This parameter selection resulted in a topological fit (R² = 0.85) and a mean connectivity approaching zero (Figs. 1 F-G). After constructing the weighted co-expression network based on the selected power value, the dynamic tree cutting method was used to identify gene modules. The minimum number of genes in each module was set to 200, and after merging highly similar modules (cutHeight = 0.99), a total of 6 co-expression modules were identified (excluding the grey module) (Fig. 1 H). Next, the correlations between each module and DE-ERG scores were analyzed, and 4 modules were found to have a significant correlation with DE-ERG scores (|cor| >0.3, P < 0.05). Among the modules, the brown module demonstrated the strongest positive correlation with DE-ERG scores (cor = 0.67, P < 0.05), whereas the red and blue modules were most negatively correlated (cor = -0.64, P < 0.05) (Fig. 1 I). In addition, it was further observed that there were highly significant correlations between GS and MM in the brown module (cor = 0.581, P < 0.05), red module (cor = 0.784, P < 0.05), and blue module (cor = -0.723, P < 0.05) (Fig. 1 J). This indicated that the genes within the modules had good consistency with the phenotypic characteristics. Finally, genes were screened according to the criteria of |MM| >0.6 and |GS| >0.4. A total of 3788, 767, and 159 genes were obtained from the blue, brown, and red modules respectively, resulting in a total of 4714 exosome-related key module genes (Fig. 1 K, Supplementary Table S4 ). The intersection of DEGs and exosome-related key module genes yielded 23 candidate genes (Fig. 1 K, Supplementary Table S5 ). To investigate their potential functions, these candidate genes were subjected to GO and KEGG enrichment analyses, which identified 212 significant GO terms (P < 0.05), including 151 BP terms such as response to starvation and transcription by RNA polymerase III (Fig. 1 L, Supplementary Table S6 ); 11 CC terms such as transferase complex, transferring phosphorus-containing groups and RNA polymerase II, holoenzyme (Fig. 1 L, Supplementary Table S7 ); and 50 MF terms such as NADP binding and Tat protein binding (Fig. 1 L, Supplementary Table S8 ). KEGG enrichment analysis revealed that the candidate genes were significantly enriched in 7 signaling pathways (P < 0.05), including FoxO signaling pathway, mTOR signaling pathway, and autophagy - animal, and other pathways (Fig. 1 M, Supplementary Table S9 ). Subsequently, the interaction relationships between proteins encoded by the candidate genes were analyzed via the STRING database ( https://www.string-db.org ) with a confidence score > 0.15. From the analysis results, a tightly interacting network containing 18 core protein nodes was identified. Among these nodes, high-strength connections were observed between SIMCI and POLR2K, TCF20, as well as EIF4ENIF1. This suggested that these proteins may have a tendency toward functional synergy or complex formation, indicating that these core proteins might collectively participate in key biological processes during the disease progression of ITP (Fig. 1 N). The candidate genes and enriched pathways identified in this part can provide a theoretical basis for potential therapeutic targets and mechanistic studies of ITP. 3.2 Identification of 4 biomarkers and construction of nomogram To further screen reliable biomarkers from 23 candidate genes, two machine learning algorithms—Boruta analysis and LASSO regression analysis—were applied in this study for biomarkers screening. Based on all samples from the dataset GSE43179 (Sequencing platform: GPL570), first, Feature Gene Set 1—comprising 5 genes (GABARAPL1, SLC39A14, HIBADH, FUT11, and GSR)—was obtained via Boruta analysis (Fig. 2 A). Next, for the LASSO regression analysis, the optimal regularization parameter was determined through 5-fold cross-validation. When lambda.min = 0.175, the model achieved the minimum Mean Squared Error (MSE); subsequently, feature gene set 2—consisting of 5 genes (GABARAPL1, SLC39A14, HIBADH, IGF1R, and GSR)—was obtained (Figs. 2 B-C). By taking the intersection of these two feature gene sets, 4 biomarkers were finally identified for subsequent analysis, namely GABARAPL1, SLC39A14, HIBADH, and GSR (Fig. 2 D). Subsequently, the nomogram model constructed based on the 4 biomarkers showed good predictive performance. For example, when the total points reached 229, the predicted probability of ITP was 89% (Fig. 2 E). The nomogram's performance was assessed using multiple indicators, demonstrating strong predictive ability. Its good calibration was reflected in a non-significant Hosmer-Lemeshow test result (P = 0.168) (Fig. 2 F). And an AUC of 0.878 in ROC analysis confirming its high discriminatory power (Fig. 2 G). This study successfully identified diagnostic biomarkers and constructed a reliable nomogram, thereby providing a theoretical foundation for the early diagnosis and personalized management of ITP. 3.3 GSEA and GeneMANIA analysis of biomarkers GSEA of the biomarkers was performed to identify relevant signaling pathways and biological mechanisms underlying ITP (|NES| >1, adj.p < 0.05). It was found in the study that GABARAPL1 was mainly enriched in 30 pathways, including spliceosome and neuroactive ligand receptor interaction, among others (Fig. 3 A, Supplementary Table S10 ); SLC39A14 was mainly enriched in 29 pathways, including ubiquitin-mediated proteolysis and neuroactive ligand receptor interaction, among others (Fig. 3 B, Supplementary Table S11 ); HIBADH was mainly enriched in 35 pathways, including spliceosome and neuroactive ligand receptor interaction, among others (Fig. 3 C, Supplementary Table S12 ). GSR was mainly enriched in 26 pathways, including ribosome and spliceosome, among others (Fig. 3 D, Supplementary Table S13 ). Subsequently, genes interacting with the four biomarkers and the biological processes they co-participated in were predicted using the GeneMANIA database ( http://genemania.org/ ). The results showed that the top 20 genes interacting with the four biomarkers included HIBCH, GSTO2, ATG4D, ALDH6A1, PGD, ATG7, GLRX, ATP6V1C1, ATG4A, G6PD, DLD, FOLR1, TXN, SLC4A4, HSD17B10, GLYR1, STBD1, AIFM1, RETREG3, and SLC25A20. Among these, the biomarker HIBADH, together with genes HIBCH, ALDH6A1, and HSD17B10, was found to co-participate in the branched-chain amino acid metabolic process, branched-chain amino acid catabolic process, and alpha-amino acid catabolic process. The biomarker GSR, along with genes TXN and GLRX, was involved in oxidoreductase activity (acting on a sulfur group of donors) and disulfide oxidoreductase activity. Notably, the associations between the four biomarkers and their interacting genes were mainly based on physical interactions, suggesting that these biomarkers might function in protein complex assembly or direct binding (Fig. 3 E). Taken together, GSEA and GeneMANIA analyses implicated the four biomarkers in ITP disease mechanisms, potentially through their influence on essential biological processes like spliceosome function, amino acid metabolism, redox equilibrium, and the ubiquitin-proteasome system. These insights thereby furnish important clues for subsequent research to unravel their molecular underpinnings and to pinpoint potential therapeutic targets. 3.4 Immune infiltration analysis Analysis of the GSE43179 dataset (GPL570) revealed distinct immune cell infiltration profiles in ITP (Fig. 4 A). Wilcoxon test identified three differentially abundant immune cells: activated NK cells, plasma cells, and monocytes (Fig. 4 B). Significant biomarker-immune cell correlations were uncovered (Fig. 4 C): GABARAPL1 was positively correlated with plasma cells (cor = 0.74) and negatively with activated NK cells (cor=-0.50); SLC39A14 and HIBADH both showed negative correlations with activated NK cells (cor=-0.69 and − 0.74, respectively), with SLC39A14 also correlating negatively with monocytes (cor=-0.53) (all P < 0.05). These results underscore the immunoregulatory importance of these biomarkers, providing mechanistic insights and a theoretical foundation for targeted ITP therapy. 3.5 Molecular regulatory network analysis of biomarkers The potential regulatory mechanisms of the biomarkers were further revealed by molecular regulatory network analysis. Upstream TFs that regulate the biomarkers were first predicted, and the results showed that GABARAPL1, GSR, HIBADH, and SLC39A14 were found to be regulated by 20, 47, 19, and 35 TFs, respectively (Fig. 5 A). This finding indicated that these biomarkers are subject to complex and diverse regulatory networks at the transcriptional level. To further investigate the post-transcriptional regulatory mechanisms, differentially expressed miRNAs (DE-miRNAs) between the disease group and the control group were analyzed based on the GSE43179 dataset (sequencing platform: GPL14613). A total of 45 DE-miRNAs were identified, among which 35 miRNAs were downregulated and 10 miRNAs were upregulated in the Immune Thrombocytopenia (ITP) group, with the screening criteria of |log 2 FC| >0.25 and P < 0.05 (Figs. 5 B-C, Supplementary Table S14 ). Meanwhile, screening of the miRNet database yielded 490 miRNAs predicted to target the biomarkers (Figs. 5 D, Supplementary Table S15 ). These results were provided as important theoretical bases and resource clues for an in-depth understanding of the molecular regulatory networks (including miRNA and TF levels) of GABARAPL1, GSR, HIBADH, and SLC39A14 in the pathological process of ITP. 3.6 Drug prediction analysis of biomarkers and molecular docking To identify potential ITP therapeutics, drug prediction analysis nominated 22 and 1 candidate drugs for GSR and SLC39A14, respectively (Fig. 6 A). Subsequent molecular docking revealed favourable binding affinities between the biomarkers and their respective drugs: SLC39A14 bound to nortriptyline with a free energy of -7.9 kcal/mol (Figs. 6 B-C, Table 1 ), while GSR bound to oxiglutatione at -17.1 kcal/mol (Fig. 6 D, Table 1 ), both values being far lower than the − 5 kcal/mol threshold. This suggested that SLC39A14 and GSR may be specific targets for these drugs, offering insights into drug action and a basis for future targeted therapy development. 3.7 Expression level validation of biomarkers and RT-qPCR Analysis of the GSE43179 dataset (Sequencing platform: GPL570) revealed that all four biomarkers were significantly downregulated in ITP samples (P < 0.05) (Fig. 7 A). This consistent downregulation suggests a potential role in ITP pathogenesis. Subsequent validation in clinical samples by RT-qPCR confirmed that GABARAPL1, SLC39A14, and GSR were significantly downregulated in ITP (Figs. 7 B-E). These results verify the association between these biomarkers and ITP, indicating that their decreased expression may play a role in disease mechanisms. Discussion ITP is an autoimmune disorder characterized by both humoral and cellular immune-mediated platelet destruction coupled with impaired platelet production[ 34 ]. Emerging evidence indicates that exosomes play a pivotal role in ITP pathogenesis through immunomodulation and intercellular communication[ 35 ]. Nevertheless, the precise etiological mechanisms underlying ITP remain incompletely understood, and reliable diagnostic criteria are still lacking. This study employed bioinformatics approaches to systematically identify exosome-associated biomarkers and elucidate their functional mechanisms in ITP. Experimental validation in peripheral blood samples from ITP patients confirmed dysregulated expression patterns of three core genes - GABARAPL1, SLC39A14, and GSR - which demonstrate significant involvement in ITP pathogenesis. These findings suggest that GABARAPL1, SLC39A14, and GSR may serve as potential diagnostic biomarkers for ITP. However, the expression pattern of HIBADH was not consistently validated across clinical samples, possibly attributable to variations in sample size and patient characteristics. GABARAPL1 serves as a critical regulatory protein in autophagy processes, playing an indispensable role in maintaining intracellular homeostasis and ensuring cellular survival[ 36 ]. Research by Lei Li et al.[ 37 ] has demonstrated that GABARAPL1 modulates platelet production through its regulation of autophagy levels in megakaryocytes. Ahmad Reza Panahi Meymandi et al.[ 38 ] further established that GABARAPL1 dysfunction may promote the survival and proliferation of autoreactive lymphocytes, consequently disrupting immune tolerance to platelet antigens. The autophagy process exhibits extensive crosstalk with exosome biogenesis, which may amplify autoimmune responses against platelets[ 39 ]. Notably, studies have revealed significantly reduced GABARAPL1 expression in ITP patients[ 40 ], potentially contributing to disease progression through multiple interconnected mechanisms: impairment of ubiquitin phosphorylation, dysregulation of PPAR signaling pathways, compromised mitophagy, and suppression of ferroptosis. Our current findings corroborate these observations, demonstrating markedly lower GABARAPL1 expression in ITP patients compared to healthy controls. These collective insights suggest that GABARAPL1 likely influences ITP pathogenesis through its dual regulatory effects on megakaryocyte function and immune cell homeostasis. SLC39A14 functions as a crucial zinc transporter that plays a pivotal role in maintaining immune homeostasis[ 41 ]. Zinc deficiency or metabolic dysregulation can promote inflammatory responses and contribute to autoimmune pathogenesis. Alterations in SLC39A14 function or expression may disrupt zinc homeostasis, thereby participating in the immune dysregulation observed in ITP[ 42 ]. Zhizhao Deng et al.[ 43 ] reported a potential association between exosomes and SLC39A14, demonstrating that bone marrow mesenchymal stem cell-derived exosomes deliver miR-16-5p to hepatocytes, leading to post-transcriptional suppression of SLC39A14 expression. Exosomes released from cells with low SLC39A14 expression can modify intracellular zinc levels and exacerbate inflammatory responses[ 44 ]. Activated immune cells may secrete exosomes carrying specific miRNAs or autoantigens, which upon uptake by other immune cells, could potentially accelerate ITP progression[ 45 ]. In our current study, we observed significantly reduced SLC39A14 expression in ITP patients compared to healthy controls. These findings suggest that SLC39A14 likely contributes to ITP pathogenesis through its involvement in zinc metabolism dysregulation. HIBADH serves as a pivotal enzyme in valine catabolism, catalyzing the oxidation of 3-hydroxyisobutyrate to methylmalonate semialdehyde, thereby playing an essential role in energy metabolism and amino acid degradation[ 46 ]. Emerging evidence suggests that defects in energy production and biosynthetic pathways may lead to immune cell dysfunction. Wenwei Chen et al.[ 47 ] demonstrated that HIBADH exerts protective effects through its regulation of mitochondrial function and mitochondria-associated oxidative stress. Altered HIBADH expression can induce significant changes in metabolite concentrations, potentially exerting undefined regulatory effects on immune cells that may contribute to ITP progression[ 48 ]. Yuanlan Huang et al.[ 49 ] further proposed that decreased HIBADH expression might reduce the abundance of regulatory molecules carried by exosomes, consequently diminishing their protective effects on target cells and potentially exacerbating autoimmune responses. While our study observed lower HIBADH expression levels in the initial screening compared to healthy controls, this finding was not consistently validated in peripheral blood samples from ITP patients, possibly due to sample heterogeneity. These collective findings suggest that HIBADH may participate in ITP pathogenesis through its role in maintaining mitochondrial function and redox homeostasis. GSR serves as a pivotal enzyme in the cellular antioxidant defense system, playing a critical role in maintaining redox homeostasis[ 50 ]. Patients with ITP exhibit a pronounced state of oxidative stress. Yanxia Zhan et al.[ 51 ] demonstrated that elevated reactive oxygen species (ROS) levels can promote inflammatory cytokine release through immune cell activation while simultaneously causing direct damage to both platelets and megakaryocytes. Complementary research by Yuquan Xie et al.[ 52 ] revealed that excessive ROS directly attacks platelet membrane phospholipids, thereby triggering platelet apoptosis. Further investigations have shown that exosomes secreted by immune cells under oxidative stress conditions may carry increased loads of inflammatory non-coding RNAs[ 53 ]. When internalized by megakaryocytes, these exosomes could amplify oxidative stress and inflammatory signaling, thereby exacerbating autoimmune responses. Our current study identified significantly reduced GSR expression in ITP patients compared to healthy controls. These collective findings strongly suggest that GSR contributes to ITP pathogenesis through its dual roles in antioxidant defense and potential modulation of exosome-mediated intercellular communication pathways that sustain autoimmune responses. The spliceosome, a macromolecular complex composed of small nuclear ribonucleoproteins and auxiliary proteins, plays a critical role in mRNA processing, with its dysfunction being implicated in various hematological disorders[ 54 ]. Emerging evidence suggests that aberrant spliceosome regulation can lead to abnormal alternative splicing of platelet production-related genes in megakaryocytes[ 55 ]. Furthermore, spliceosome components may contribute to autoimmune responses through dysregulated activation of T and B lymphocytes, resulting in immune-mediated platelet destruction[ 56 ]. Andrea Pellagatti et al.[ 57 ] demonstrated that exosomes can transport spliceosome-associated non-coding RNAs, thereby modulating gene expression in recipient cells through intercellular communication. Our GSEA revealed that GABARAPL1, SLC39A14, HIBADH, and GSR are collectively enriched in the spliceosome signaling pathway. These findings suggest that exosomes derived from aberrant immune cells may transmit defective splicing signals, potentially impairing megakaryocyte differentiation and platelet maturation. Notably, these four biomarkers also show significant enrichment in the neuroactive ligand-receptor interaction pathway and ubiquitin-mediated proteolysis signaling pathway, indicating their potential involvement in multiple regulatory mechanisms in ITP pathogenesis. Monocytes, as essential components of the innate immune system, play a pivotal role in inflammatory responses[ 58 ]. In recent years, their involvement in ITP pathogenesis has garnered increasing attention. Yajing Zhao et al.[ 59 ] demonstrated that elevated TNF-α mRNA expression in peripheral blood monocytes of ITP patients exacerbates immune dysregulation through activation of the NF-κB signaling pathway. Complementary research by Wei Wang et al.[ 60 ] revealed that GSK-3β inhibitors can attenuate monocyte phagocytic capacity by blocking TNF-α signaling, thereby ameliorating thrombocytopenia. Our immune infiltration analysis yielded several significant findings: First, SLC39A14 expression showed a strong negative correlation with monocyte infiltration levels. Second, GABARAPL1, SLC39A14, and HIBADH all demonstrated significant negative correlations with activated NK cells - a finding consistent with Farida Hussein El-Rashedi et al.'s[ 61 ] observations of NK cell activity in pediatric ITP patients. These collective results suggest that dysregulated monocyte and NK cell activities contribute substantially to immune homeostasis disruption and subsequent platelet destruction in ITP. miR-484 has emerged as a critically important microRNA in various cancers and metabolic disorders, where it plays a regulatory role in apoptosis and mitochondrial function signaling pathways[ 62 ]. Existing research has established that miR-484 modulates apoptotic processes in immune cells[ 63 ], leading us to hypothesize that its dysregulated expression may contribute to T-cell homeostasis imbalance in ITP. Further mechanistic studies have demonstrated that miR-484 interferes with megakaryocyte differentiation and maturation by suppressing the Wnt/MAPK pathway and downregulating β-catenin expression[ 64 ]. Our bioinformatic analyses provide compelling evidence that miR-484 directly targets and regulates SLC39A14 expression. This regulatory interaction appears to exert dual pathological effects: impairing T-cell function and disrupting megakaryocyte development, ultimately leading to enhanced platelet destruction in ITP. Our study conducted comprehensive drug prediction analyses for the four candidate biomarkers, revealing that only SLC39A14 and GSR possess target-specific therapeutic agents. Nortriptyline, a tricyclic antidepressant with immunomodulatory properties, has been shown to suppress pro-inflammatory cytokine production and modulate T-cell function, thereby potentially reducing immune-mediated platelet destruction[ 65 ]. Mechanistically, this drug may interact with SLC39A14 to influence zinc ion homeostasis, consequently regulating both immune cell activity and megakaryocyte differentiation. Oxiglutathione (GSSG), as a substrate for GSR-catalyzed reactions, plays a pivotal role in maintaining cellular antioxidant capacity[ 66 ]. In ITP patients, enhancing GSR activity could elevate intracellular reduced glutathione (GSH) levels, thereby mitigating oxidative damage to platelets. These findings suggest that nortriptyline and oxiglutathione represent promising therapeutic candidates targeting SLC39A14 and GSR respectively, with complementary mechanisms of action: nortriptyline primarily through immunomodulation and oxiglutathione via antioxidant pathways. Their combined use may offer a synergistic approach to address the multifactorial pathogenesis of ITP. In conclusion, this study provides compelling evidence that GABARAPL1, SLC39A14, HIBADH and GSR are functionally involved in the pathogenesis of ITP. Importantly, GABARAPL1, SLC39A14 and GSR emerge as particularly promising candidates for novel diagnostic biomarkers in ITP. While these findings advance our understanding of ITP pathology, certain limitations must be acknowledged regarding the current validation using peripheral blood samples with relatively small sample sizes and incomplete mechanistic exploration of disease pathogenesis. Future investigations employing expanded clinical cohorts will be essential to further validate the diagnostic potential of these biomarkers and more comprehensively characterize their functional roles in ITP progression. Declarations The authors gratefully acknowledge the technical support and facilities provided by the Department of Hematology at the First Affiliated Hospital of Chongqing Medical University and Fengdu County Traditional Chinese Medicine Hospital. Author Contributions F.F.L.completed the bioinformatics analysis component of this study, drafted the initial manuscript, and participated in subsequent revisions. Z .Y.C. and Z.H.Y. systematically organized the research data and refined all figures and tables. J .Y.S. and J .P. collected clinical specimens and performed analytical processing of the research outco mes. P. H. and Z .S.Y. were responsible for the conceptual framework and methodological rigor of this study, additionally conducting critical reviews of the revised manuscript for academic coherence. Funding This study was supported by grants from the Chongqing Medical Scientific Research Project (Grant No. 2025MSXM065) and the Chongqing Natural Science Foundation (Grant No. CSTB2025NSCQ-GPX1218). Data Availability Statement The dataset (GSE43179) supporting the conclusions of this article is available in the [GEO] repository, [https://www.ncbi.nlm.nih.gov/geo/]. The ERGs data are available in the [ExoBCD] repository, [https://exobcd.liumwei.org/]. Competing Interests The authors declare no competing interests. Corresponding Authors Correspondence to Ping Huang and Zesong Yang. References Schramm, T. et al. Fibrinolysis is impaired in patients with primary immune thrombocytopenia. J. Thromb. Haemost . 22 (11), 3209–3220 (2024). Cao, J. et al. MST4 kinase regulates immune thrombocytopenia by phosphorylating STAT1-mediated M1 polarization of macrophages. Cell. Mol. Immunol. 20 (12), 1413–1427 (2023). Schifferli, A. et al. Adolescents and young adults with newly diagnosed primary immune thrombocytopenia. Haematologica 108 (10), 2783–2793 (2023). Published 2023 Oct 1. Author, R. M. et al. Erratum to: 36th International Symposium on Intensive Care and Emergency Medicine: Brussels, Belgium. 15–18 March 2016. Crit Care. ;20:347. Published 2016 Oct 24. (2016). Hu, Y. et al. Efficacy and safety of sovleplenib (HMPL–523) in adult patients with chronic primary immune thrombocytopenia in China (ESLIM–01): a randomised, double-blind, placebo-controlled, phase 3 study. Lancet Haematol. 11 (8), e567–e579 (2024). Zhuang, X. et al. Decreased cyclooxygenase–2 associated with impaired megakaryopoiesis and thrombopoiesis in primary immune thrombocytopenia. J. Transl Med. 21 (1), 540 (2023). Published 2023 Aug 12. Huang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21 + Regulatory T Cell-Mediated Immune Suppression via MiR–363–3p. Inflamm. Published online March 4 , (2025). Larue, M. et al. Efficacy and safety of dapsone in adult primary immune thrombocytopenia. Blood Adv. 9 (8), 1976–1983 (2025). Yan, S. et al. A phase 2 trial of orelabrutinib showing promising efficacy and safety in patients with persistent or chronic primary immune thrombocytopenia. Am. J. Hematol. 99 (7), 1392–1395 (2024). Goodman, S. R. The exosome collection. Cell. Mol. Biol. Lett. 30 (1), 33 (2025). Published 2025 Mar 24. Arya, S. B., Collie, S. P. & Parent, C. A. The ins-and-outs of exosome biogenesis, secretion, and internalization. Trends Cell. Biol. 34 (2), 90–108 (2024). Isaac, R., Reis, F. C. G., Ying, W. & Olefsky, J. M. Exosomes as mediators of intercellular crosstalk in metabolism. Cell. Metab. 33 (9), 1744–1762 (2021). Fang, Y. et al. Exosomes as biomarkers and therapeutic delivery for autoimmune diseases: Opportunities and challenges. Autoimmun. Rev. 22 (3), 103260 (2023). Shen, Z. et al. Effects of Mesenchymal Stem Cell-Derived Exosomes on Autoimmune Diseases. Front. Immunol. 12 , 749192 (2021). Published 2021 Sep 27. Huang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21 + Regulatory T Cell-Mediated Immune Suppression via MiR–363–3p. Inflamm. Published online March 4 , (2025). Sun, Y. et al. Proteomic analysis and microRNA expression profiling of plasma-derived exosomes in primary immune thrombocytopenia. Br. J. Haematol. 194 (6), 1045–1052 (2021). Wang, J., Qing, M., Gui, J., Zhong, P. & Hua, H. Identification of exosome-related genes associated with prognosis and immune infiltration features in pancreatic cancer. Discov Oncol. 16 (1), 192 (2025). Published 2025 Feb 17. Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43 (7), e47 (2015). Shi, Y. et al. Crosstalk of ferroptosis regulators and tumor immunity in pancreatic adenocarcinoma: novel perspective to mRNA vaccines and personalized immunotherapy. Apoptosis 28 (9–10), 1423–1435 (2023). Gustavsson, E. K., Zhang, D., Reynolds, R. H., Garcia-Ruiz, S. & Ryten, M. ggtranscript: an R package for the visualization and interpretation of transcript isoforms using ggplot2. Bioinformatics 38 (15), 3844–3846 (2022). Horvath, S. & Dong, J. Geometric interpretation of gene coexpression network analysis. PLoS Comput. Biol. 4 (8), e1000117 (2008). Published 2008 Aug 15. Wu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innov. (Camb) . 2 (3), 100141 (2021). Published 2021 Jul 1. Smoot, M. E., Ono, K., Ruscheinski, J., Wang, P. L. & Ideker, T. Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics 27 (3), 431–432 (2011). Mahmoudian, M., Venäläinen, M. S., Klén, R. & Elo, L. L. Stable Iterative Variable Selection Bioinf. ; 37 (24):4810–4817. (2021). Friedman, J., Hastie, T. & Tibshirani, R. Regularization Paths for Generalized Linear Models via Coordinate Descent. J. Stat. Softw. 33 (1), 1–22 (2010). Li, M. et al. Recognition of refractory Mycoplasma pneumoniae pneumonia among Myocoplasma pneumoniae pneumonia in hospitalized children: development and validation of a predictive nomogram model. BMC Pulm Med. 23 (1), 383 (2023). Published 2023 Oct 10. Robin, X. et al. pROC: an open-source package for R and S + to analyze and compare ROC curves. BMC Bioinform. 12 , 77 (2011). Published 2011 Mar 17. Nazari-Ghadikolaei, A., Fikse, F., Gelinder Viklund, Å. & Eriksson, S. Factor analysis of evaluated and linearly scored traits in Swedish Warmblood horses. J. Anim. Breed. Genet. 140 (4), 366–375 (2023). Wu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innov. (Camb) . 2 (3), 100141 (2021). Published 2021 Jul 1. Qi, W. Y. et al. Immune cells in metabolic associated fatty liver disease: Global trends and hotspots (2004–2024). World J. Hepatol. 17 (3), 103327 (2025). Li, Y., Jiang, H., Luo, L. & Mei, H. Immunotherapy-associated hemostatic abnormalities: bleeding and thrombotic complications. Ann. Hematol. 104 (7), 3537–3551 (2025). Hillier, K. et al. Immunoglobulins act as predictors of chronicity in pediatric immune thrombocytopenia. Blood Adv. 8 (19), 5118–5121 (2024). Aran, D., Hu, Z. & Butte, A. J. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 18 (1), 220 (2017). Published 2017 Nov 15. González-López, T. J. & Provan, D. The new era of primary immune thrombocytopenia management in adults: A narrative review of current and emerging treatments. Blood Rev. 73 , 101300 (2025). Li, G. et al. The therapeutic potential of exosomes in immunotherapy. Front. Immunol. 15 , 1424081 (2024). Published 2024 Jul 8. Guo, H. et al. Dysfunction of astrocytic glycophagy exacerbates reperfusion injury in ischemic stroke. Redox Biol. 74 , 103234 (2024). Li, L. et al. Beclin 1 of megakaryocytic lineage cells is locally dispensable for platelet hemostasis but functions distally in bone homeostasis. Bone Res. 13 (1), 32 (2025). Published 2025 Mar 3. Panahi Meymandi, A. R. et al. Crosstalk between autophagy and metabolic regulation of (CAR) T cells: therapeutic implications. Front. Immunol. 14 , 1212695 (2023). Published 2023 Aug 22. Yan, J. et al. Platelet Pharmacytes for the Hierarchical Amplification of Antitumor Immunity in Response to Self-Generated Immune Signals. Adv. Mater. 34 (23), e2109517 (2022). Mo, J. et al. Comprehensive analysis and prediction model of mitophagy and ferroptosis in primary immune thrombocytopenia. Br. J. Haematol. 204 (6), 2429–2441 (2024). Yang, Y. et al. SOX4-ZIP14-zinc metabolism mediates oncogenesis and suppresses T cell immunity in nasopharyngeal carcinoma. Cell. Rep. Med. 6 (9), 102300 (2025). Shao, Y. et al. SLC39A10 is a key zinc transporter in T cells and its loss mitigates autoimmune disease. Sci. China Life Sci. 68 (7), 1855–1870 (2025). Deng, Z. et al. Mesenchymal Stem Cells Prevent SLC39A14-Dependent Hepatocyte Ferroptosis through Exosomal miR–16–5p in Liver Graft. Adv. Sci. (Weinh) . 12 (6), e2411380 (2025). Sauer, A. K. et al. Zinc is a key regulator of gastrointestinal development, microbiota composition and inflammation with relevance for autism spectrum disorders. Cell. Mol. Life Sci. 79 (1), 46 (2021). Published 2021 Dec 22. Ma, L. et al. lncRNA, miRNA, and mRNA of plasma and tumor-derived exosomes of cardiac myxoma-related ischaemic stroke. Sci Data. ;12(1):91. Published 2025 Jan 16. (2025). Tukacs, V. et al. Chronic Cerebral Hypoperfusion-Induced Disturbed Proteostasis of Mitochondria and MAM Is Reflected in the CSF of Rats by Proteomic Analysis. Mol. Neurobiol. 60 (6), 3158–3174 (2023). Chen, W. et al. Mitochondrial enzyme HIBADH protects against calcium oxalate nephrolithiasis by modulating oxidative stress and apoptosis. Arch. Biochem. Biophys. 770 , 110452 (2025). Bishop, C. A. et al. Detrimental effects of branched-chain amino acids in glucose tolerance can be attributed to valine induced glucotoxicity in skeletal muscle. Nutr. Diabetes . 12 (1), 20 (2022). Published 2022 Apr 13. Huang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21 + Regulatory T Cell-Mediated Immune Suppression via MiR–363–3p. Inflamm. Published online March 4 , (2025). Zhang, R. et al. Pleiotropic effects of a mitochondrion-targeted glutathione reductase inhibitor on restraining tumor cells. Eur. J. Med. Chem. 248 , 115069 (2023). Zhan, Y. et al. Impaired mitochondria of Tregs decreases OXPHOS-derived ATP in primary immune thrombocytopenia with positive plasma pathogens detected by metagenomic sequencing. Exp. Hematol. Oncol. 11 (1), 48 (2022). Published 2022 Sep 1. Xie, Y. et al. PM2.5 promotes platelet activation and thrombosis via ROS/MAPKs pathway-mediated mitochondrial dysfunction. Environ. Res. 283 , 122116 (2025). Wang, C., Xu, M., Fan, Q., Li, C. & Zhou, X. Therapeutic potential of exosome-based personalized delivery platform in chronic inflammatory diseases. Asian J. Pharm. Sci. 18 (1), 100772 (2023). Wang, J. et al. Second-generation antipsychotics induce cardiotoxicity by disrupting spliceosome signaling: Implications from proteomic and transcriptomic analyses. Pharmacol. Res. 170 , 105714 (2021). Sun, T. et al. Proteomics landscape and machine learning prediction of long-term response to splenectomy in primary immune thrombocytopenia. Br. J. Haematol. 204 (6), 2418–2428 (2024). Karginov, T. A., Ménoret, A. & Vella, A. T. Optimal CD8 + T cell effector function requires costimulation-induced RNA-binding proteins that reprogram the transcript isoform landscape. Nat. Commun. 13 (1), 3540 (2022). Published 2022 Jun 20. Pellagatti, A. et al. Impact of spliceosome mutations on RNA splicing in myelodysplasia: dysregulated genes/pathways and clinical associations. Blood 132 (12), 1225–1240 (2018). Mildner, A. & Yona, S. Monocytes and their doppelgängers: An immunological crossroads. Sci. Immunol. 9 (101), eadr6672 (2024). Zhao, Y. et al. Tumor Necrosis Factor-α Blockade Corrects Monocyte/Macrophage Imbalance in Primary Immune Thrombocytopenia. Thromb. Haemost . 121 (6), 767–781 (2021). Wang, W. et al. Quercetin, a key active ingredient of Jianpi Zishen Xiehuo Formula, suppresses M1 macrophage polarization and platelet phagocytosis by inhibiting STAT3 activation based on network pharmacology. Naunyn Schmiedebergs Arch. Pharmacol. 397 (6), 4219–4233 (2024). El-Rashedi, F. H., El-Hawy, M. A., Helwa, M. A. & Abd-Allah, S. S. Study of CD4+, CD8+, and natural killer cells (CD16+, CD56+) in children with immune thrombocytopenic purpura. Hematol. Oncol. Stem Cell. Ther. 10 (1), 8–14 (2017). Zhao, Z. et al. Targeted delivery of exosomal miR–484 reprograms tumor vasculature for chemotherapy sensitization. Cancer Lett. 530 , 45–58 (2022). Li, H. et al. Mir–484 contributes to diminished ovarian reserve by regulating granulosa cell function via YAP1-mediated mitochondrial function and apoptosis. Int. J. Biol. Sci. 18 (3), 1008–1021 (2022). Published 2022 Jan 1. Lin, T. et al. Identification of exosomal miR–484 role in reprogramming mitochondrial metabolism in pancreatic cancer through Wnt/MAPK axis control. Pharmacol. Res. 197 , 106980 (2023). Guo, Y. et al. Rbms1 promotes pulmonary fibrosis by stabilising Sumo2 mRNA to facilitate Smad4-SUMOylation and fibroblast activation. Eur. Respir J. 66 (3), 2401667 (2025). Published 2025 Sep 25. De Luca, B. et al. Efficacy and tolerability of antidepressants in individuals suffering from physical conditions and depressive disorders: network meta-analysis. Br. J. Psychiatry . 227 (2), 553–566 (2025). Tables Table 1 is available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files floatimage1.png Table1.xls SupplementaryTableS1.csv SupplementaryTableS2.xls SupplementaryTableS3.csv SupplementaryTableS4.csv SupplementaryTableS5.csv SupplementaryTableS6.csv SupplementaryTableS7.csv SupplementaryTableS8.csv SupplementaryTableS9.csv SupplementaryTableS10.csv SupplementaryTableS11.csv SupplementaryTableS12.csv SupplementaryTableS13.csv SupplementaryTableS14.csv SupplementaryTableS15.csv Cite Share Download PDF Status: Published Journal Publication published 20 Mar, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 05 Jan, 2026 Reviews received at journal 31 Dec, 2025 Reviews received at journal 15 Dec, 2025 Reviewers agreed at journal 15 Dec, 2025 Reviewers agreed at journal 07 Dec, 2025 Reviewers invited by journal 04 Dec, 2025 Editor assigned by journal 24 Nov, 2025 Editor invited by journal 24 Nov, 2025 Submission checks completed at journal 20 Nov, 2025 First submitted to journal 20 Nov, 2025 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-8092905\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":556572149,\"identity\":\"27420dca-985f-46bb-91bd-e910c4643e02\",\"order_by\":0,\"name\":\"Fangfang Lou\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Chongqing Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Fangfang\",\"middleName\":\"\",\"lastName\":\"Lou\",\"suffix\":\"\"},{\"id\":556572150,\"identity\":\"0a77809d-5037-4fdb-a747-21ff46c36bdc\",\"order_by\":1,\"name\":\"Zhiyue Chen\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Chongqing Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zhiyue\",\"middleName\":\"\",\"lastName\":\"Chen\",\"suffix\":\"\"},{\"id\":556572151,\"identity\":\"e5051603-5a65-4722-b5b0-2deb345d8f68\",\"order_by\":2,\"name\":\"Zihao Yuan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Chongqing Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zihao\",\"middleName\":\"\",\"lastName\":\"Yuan\",\"suffix\":\"\"},{\"id\":556572152,\"identity\":\"03fb7cfa-5db8-4627-be1f-bb07005184ec\",\"order_by\":3,\"name\":\"Jie Peng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Fengdu County Traditional Chinese Medicine Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jie\",\"middleName\":\"\",\"lastName\":\"Peng\",\"suffix\":\"\"},{\"id\":556572153,\"identity\":\"b5c54a9c-0a79-492e-981d-39c4a70b9b73\",\"order_by\":4,\"name\":\"Jiangyu Sun\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Fengdu County Traditional Chinese Medicine Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jiangyu\",\"middleName\":\"\",\"lastName\":\"Sun\",\"suffix\":\"\"},{\"id\":556572154,\"identity\":\"5e0f69b9-419d-4cef-9973-374f188af2ba\",\"order_by\":5,\"name\":\"Ping Huang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Fengdu County Traditional Chinese Medicine Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Ping\",\"middleName\":\"\",\"lastName\":\"Huang\",\"suffix\":\"\"},{\"id\":556572155,\"identity\":\"9f196ccc-4034-4e6e-98c2-d238ee88b87f\",\"order_by\":6,\"name\":\"Zesong Yang\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIiWNgGAWjYHAD5gMHPlSQpoUt8eCMM6Rp4TE+zNtChDr59vaHjwt+HZY351/z4QBvA4M8v9gB/FoYe84YG8/sO2y4c8bbDQckdzAYzpydgF8Ls0QOmzRvz2HGDTfObjhgeIYhweA2AS1s8s+fgbTYb7hx5sGBxDYitPBIMJhJ8/w4nLjhfA/DgYPEaJHgyTE25m1IT95wg83gYMMZCcJ+kW8//vAxzx9r2w3nDz/+/KfCRp5fmoAWMGBsawbaB1YpQYRyMPhTx8DAf4BY1aNgFIyCUTDSAAAK6kzt7jIDEgAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Chongqing Medical University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Zesong\",\"middleName\":\"\",\"lastName\":\"Yang\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-11-12 06:53:19\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-8092905/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-8092905/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1038/s41598-026-43618-1\",\"type\":\"published\",\"date\":\"2026-03-20T15:57:50+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":97897108,\"identity\":\"b76cbd08-1690-40ae-a367-f3f8bd6b7212\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:27\",\"extension\":\"doc\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":6213349,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"2.doc\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/d293092d2095d9c59c5806dc.doc\"},{\"id\":97744928,\"identity\":\"06e0718d-041e-496a-9414-4c2ad51c3bd7\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"json\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":8280,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"89530134661347c5919190df0683d150.json\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/6d3828f17ff2a51774ea99cf.json\"},{\"id\":97896931,\"identity\":\"f74312fa-d871-4722-8906-0c668dcaa674\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:13\",\"extension\":\"csv\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":716,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS1.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/a33eccb6ec6b63e5109cfa1d.csv\"},{\"id\":97744931,\"identity\":\"feab46c8-a43b-4176-bed1-81e960119856\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":14909,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS10.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/ca3073b610623db7b60683db.csv\"},{\"id\":97744956,\"identity\":\"77b540fe-14f4-4a9e-a81a-2d6fb6cebc8a\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"csv\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":15144,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS11.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/a0716ef810bd45d6405f93b1.csv\"},{\"id\":97745032,\"identity\":\"eff76f30-52bb-4f45-a172-02848d30ea28\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"csv\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":15288,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS12.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/f431ab3b741b1ee5ff36e909.csv\"},{\"id\":97744934,\"identity\":\"16ae0d2c-60ce-4448-b1a0-c67c7841259c\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":6,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":12127,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS13.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/d2b83b80a2fa824f8a039f3a.csv\"},{\"id\":97744955,\"identity\":\"240858a3-ccb5-4cb3-b6e9-b13a392be48a\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":136260,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS14.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/13b0e5c9cb2fdc1da798ab7b.csv\"},{\"id\":97745023,\"identity\":\"7973bc8f-61f5-4ad7-838d-44fd66f295ff\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"csv\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":65503,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS15.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/1ff170f3e4e4ba8e2936aac5.csv\"},{\"id\":97744944,\"identity\":\"95ddb605-1be7-4e01-a222-069a51d6368e\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"xls\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":20992,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS2.xls\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/ca77353ffc4dbd51527bbdcb.xls\"},{\"id\":97896269,\"identity\":\"04e5d0e9-06ad-432f-9c36-165918798189\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:36:16\",\"extension\":\"csv\",\"order_by\":10,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":2757822,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS3.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/387bbbd80b8e89fd054b0ecc.csv\"},{\"id\":97745042,\"identity\":\"574ad313-751a-43fc-9f0f-014b439d7e01\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":11,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":41078,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS4.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/09e0815b3f7d90f8fe2af591.csv\"},{\"id\":97745034,\"identity\":\"777fae09-1c2d-4258-b667-fad20c7395c7\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":12,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":195,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS5.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/6908ad9298e3531c004e9dc0.csv\"},{\"id\":97744998,\"identity\":\"3265a152-a142-4785-8d0f-ade679ffe97d\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"csv\",\"order_by\":13,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":23455,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS6.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/583730c7171477f666f8acdf.csv\"},{\"id\":97744941,\"identity\":\"ac47e374-16de-4915-81f1-bb637b2b02dd\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":14,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":1768,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS7.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/6ce1ac29ae61cde629a445d0.csv\"},{\"id\":97745035,\"identity\":\"b21808e9-9ddc-4dd4-9ae0-31bffcd3ce89\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":15,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":7787,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS8.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/3b463c28caf2d75950f7b76f.csv\"},{\"id\":97745039,\"identity\":\"42ddc8bf-a44d-46bc-9662-345d4c4f962f\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":16,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":1395,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS9.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/97b696f2c81538f88eb6db64.csv\"},{\"id\":97745030,\"identity\":\"bf67aa00-9fcf-4e29-a6cd-3d55f609b9b1\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"xls\",\"order_by\":17,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":19968,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Table1.xls\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/aca7548a87d05c314cf1cc7b.xls\"},{\"id\":97745018,\"identity\":\"b38915fd-24b1-4277-8fdd-c1eae99d2e40\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"xml\",\"order_by\":18,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":146721,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"89530134661347c5919190df0683d1501enriched.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/fd5c1b8b7b933328af9e7f1f.xml\"},{\"id\":97744945,\"identity\":\"69d853e1-82f0-422b-86f7-e7ac36baeb40\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":19,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":281503,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/87059e9a3bd1240de0fd5393.png\"},{\"id\":97744951,\"identity\":\"590a559f-d562-4efb-9391-ffdfa71a1f0b\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":20,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":373655,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/0ae7dcd60d2294a9abcfeb33.png\"},{\"id\":97744950,\"identity\":\"48210def-1dd6-4dee-b5b1-a0b965235e4a\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":21,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":229991,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/5c2fa5d063ad753721efd1f4.png\"},{\"id\":97896961,\"identity\":\"a719d771-3134-4fad-a010-0398e4fb4308\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:17\",\"extension\":\"png\",\"order_by\":22,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":487738,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/c67371277749af7c5ce07945.png\"},{\"id\":97744948,\"identity\":\"acb39a53-5f27-46b1-aac2-3584f6a29b7f\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":23,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":255304,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/31a3c860ac1ba6d14cf4eb85.png\"},{\"id\":97895828,\"identity\":\"212c416e-eff0-4fc8-bdf6-fda1d24ac2e0\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:35:08\",\"extension\":\"png\",\"order_by\":24,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":756691,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/76f5fd7be1df1b52cef83305.png\"},{\"id\":97745083,\"identity\":\"fd8cd81a-2d45-4be6-955b-99f1b4574b8c\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:37\",\"extension\":\"png\",\"order_by\":25,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":734931,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/73ef1266b691598a85871214.png\"},{\"id\":97745067,\"identity\":\"5df35af0-a951-46f9-8f92-ce72eaf65788\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"png\",\"order_by\":26,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":182726,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage8.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/d3eeb0f1bb23af63eda28722.png\"},{\"id\":97897330,\"identity\":\"c52f1fb0-41db-453e-a116-15f029be9f3b\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:44\",\"extension\":\"png\",\"order_by\":27,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":54579,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/5392ba8665b903fa272125c9.png\"},{\"id\":97745024,\"identity\":\"24e1e6fe-b8d7-4a25-8bcf-04ba566aa354\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"png\",\"order_by\":28,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":109358,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/06de9571f38058a58fe90e9e.png\"},{\"id\":97896357,\"identity\":\"a5c7c696-b095-4dd8-a135-188c46704358\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:36:24\",\"extension\":\"png\",\"order_by\":29,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":68249,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/59b27061df500e2c6a398e81.png\"},{\"id\":97744952,\"identity\":\"40cf8af1-14e2-4aba-8c68-be66f3c6c2d1\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":30,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":107541,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/b350f31a4e4c3b8b49619323.png\"},{\"id\":97896271,\"identity\":\"53362372-dda3-49e1-8478-d984ab0d64e6\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:36:16\",\"extension\":\"png\",\"order_by\":31,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":100117,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/5a8abc921968deeb1a7cf3a6.png\"},{\"id\":97894753,\"identity\":\"e50fe3a7-48bd-4dc3-a4e5-907ecaa7b76c\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:33:00\",\"extension\":\"png\",\"order_by\":32,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":161592,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/d08543c05469a41c457a3239.png\"},{\"id\":97744954,\"identity\":\"e3419c9e-f2dd-4357-abad-db486e2b9666\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":33,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":123785,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/3cf449144ee33db006a2b735.png\"},{\"id\":97894699,\"identity\":\"a9d6546b-6449-468c-9c20-02815890ab11\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:32:54\",\"extension\":\"png\",\"order_by\":34,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":53151,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage8.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/b7206a034e684eff33a23451.png\"},{\"id\":97744959,\"identity\":\"ab2d3033-19d8-4050-a117-e26f7e8004cc\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"xml\",\"order_by\":35,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":145417,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"89530134661347c5919190df0683d1501structuring.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/77aa7bb4493ff8228d45e688.xml\"},{\"id\":97895815,\"identity\":\"58494058-120e-4633-acc2-d98f43e8b34d\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:35:06\",\"extension\":\"html\",\"order_by\":36,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":160211,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"earlyproof.html\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/c0ebdc501c0929b852e4b08c.html\"},{\"id\":97745025,\"identity\":\"c68e0ea9-0014-46f1-aa4c-71a34ffdd238\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":914477,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) Volcano plot of differentially expressed genes; (B) Heatmap displaying the expression levels of the top 10 up- and down-regulated genes, with higher expression levels shown in red and lower levels in blue; (C) Box plot of seven significantly differentially expressed ERGs; (D) Distribution of ssGSEA scores for gene sets with differential directionality in disease samples versus healthy controls in the training set; (E) Cluster analysis dendrogram of samples from the training set GSE43179; (F–G) Screening of the soft-thresholding power for the WGCNA co-expression network (threshold β value set to 19); (H–J) Module assignment of the gene co-expression network, distinguished by different colors, identifying a total of six co-expression modules; (K) Venn diagram showing the intersection of differentially expressed genes (DEGs) from the transcriptome and ERG module genes, resulting in 23 differentially expressed ERGs; (L–M) GO and KEGG enrichment analysis of the 23 candidate genes; (N) Protein–protein interaction (PPI) network.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/442001aff14aee3d566bf276.png\"},{\"id\":97896836,\"identity\":\"d080051f-27a8-45ce-b4d8-0bd900f19ec5\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:06\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":611472,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) The Boruta algorithm was employed to perform feature selection on the 23 differentially expressed ERGs, ultimately identifying 5 feature genes (highlighted in green). (B–C) Screening of the 23 differentially expressed ERGs by LASSO regression selected 5 feature genes. (D) The intersection of two machine learning algorithms revealed 4 shared genes (Venn diagram). (E–G) The resulting nomogram was validated with calibration curves and ROC analysis, confirming its diagnostic efficacy.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/8f014f71635babea43887a42.png\"},{\"id\":97894743,\"identity\":\"b8dcc603-79d7-4cc4-aac1-f88347bbda23\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:32:58\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":869017,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A–D) Panels display the top 5 enriched gene sets from the Gene Set Enrichment Analysis (GSEA) results, ranked by their correlation with four biomarkers. (E) Protein–protein interaction network between candidate genes and their predicted functional partners.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/0cb5360b5409274bc5e762a2.png\"},{\"id\":97744933,\"identity\":\"fe25d5ba-7839-455a-8ba9-1116f51f11a2\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":960469,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) The composition of immune cell infiltration in ITP and control samples. (B) Comparison of infiltration levels for differentially abundant immune cells between the ITP (orange) and control (blue) groups. (C) Correlation analysis between the candidate biomarkers and the differentially infiltrated immune cells.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/1309997b0bab70c7d1deb602.png\"},{\"id\":97895719,\"identity\":\"a362d3bf-4bec-44a9-abb6-8fbafc019867\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:34:48\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1073144,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) TF–mRNA regulatory network (genes: blue; TFs: red). (B) Volcano plot of DE-miRNAs (up: red; down: blue). (C) Heatmap of the top 10 DE-miRNAs. (D) miRNA-mRNA regulatory network.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/3a808d04e167fe7f01253463.png\"},{\"id\":97745020,\"identity\":\"854855db-da47-4c42-9560-d8de9dbd0d89\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:34\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":766876,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) Network of differentially expressed miRNAs targeting candidate genes. (B–D) Molecular docking models.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/79f7333f816803911b94f2c0.png\"},{\"id\":97744936,\"identity\":\"bad7ab5c-2a11-4b76-b187-d4c249307d4f\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"png\",\"order_by\":7,\"title\":\"Figure 7\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":592676,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e(A) Validation of biomarker expression levels in disease versus control samples from the GSE43179 dataset.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage8.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/f76998bb35a2ce7f516808e7.png\"},{\"id\":105223339,\"identity\":\"03f1e4ff-0c3d-484b-b0d7-f945c1cb4c29\",\"added_by\":\"auto\",\"created_at\":\"2026-03-23 16:04:24\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":7124448,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/0391a96e-2a18-4421-ba96-8356059547fd.pdf\"},{\"id\":97896846,\"identity\":\"c3c7dfcc-c044-42c4-a2e7-4afb624a2a9f\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:07\",\"extension\":\"png\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":440308,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/d282020d20f22fa73cd8d024.png\"},{\"id\":97745036,\"identity\":\"adeb33ce-40c0-4550-a65e-1c0713988564\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"xls\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":19968,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Table1.xls\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/5003064682f12d8b1a00e939.xls\"},{\"id\":97745026,\"identity\":\"a2799c44-da23-4681-9bf2-c160254194e4\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"csv\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":716,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS1.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/6cf26bdf620738bd2d27c50d.csv\"},{\"id\":97744935,\"identity\":\"3a6f46f6-2124-4026-98b9-bfb4cd063b26\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"xls\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":20992,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS2.xls\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/18f68c0fca01777085f94e2c.xls\"},{\"id\":97895819,\"identity\":\"3b30cc12-8061-4061-b493-f50be932ce34\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:35:06\",\"extension\":\"csv\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":2757822,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS3.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/7911fdb8fc9fb29aa786e1ff.csv\"},{\"id\":97896465,\"identity\":\"f34e6ef1-c4da-4acb-88cb-d554d539dc23\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:36:35\",\"extension\":\"csv\",\"order_by\":6,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":41078,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS4.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/a8a253cb94bcb4addf1a22b5.csv\"},{\"id\":97894692,\"identity\":\"91c94fc1-f5e2-4d46-b9ec-bc03cfd626e1\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:32:53\",\"extension\":\"csv\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":195,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS5.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/58efa82cb5da2b6bd2b5a7c1.csv\"},{\"id\":97897117,\"identity\":\"a4d6b1dd-5231-4e12-9a67-eab27d61d397\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:27\",\"extension\":\"csv\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":23455,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS6.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/dfbea5bd61fccc6293fce73a.csv\"},{\"id\":97744938,\"identity\":\"995f8733-e1f7-473b-ae0d-93207392c0db\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":1768,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS7.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/dcffa28d5e78d07434aae54a.csv\"},{\"id\":97896924,\"identity\":\"c2fc58d5-02a7-42a9-868b-d64bd1e4cfae\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:37:12\",\"extension\":\"csv\",\"order_by\":10,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":7787,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS8.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/f5964733acdb16c3474239e1.csv\"},{\"id\":97744946,\"identity\":\"eda91195-2e54-4d44-9a7b-a95df834b957\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:33\",\"extension\":\"csv\",\"order_by\":11,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":1395,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS9.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/e5c2cce579a0963bc38f126a.csv\"},{\"id\":97745028,\"identity\":\"816188ad-d34c-4a36-9652-083333a4b3ce\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"csv\",\"order_by\":12,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":14909,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS10.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/da1ca2a17fdd400dce2693f4.csv\"},{\"id\":97745031,\"identity\":\"c4815a72-bdbd-4526-ad6d-8d804dcd17f3\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:35\",\"extension\":\"csv\",\"order_by\":13,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":15144,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS11.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/0398665c13055805d25ea6de.csv\"},{\"id\":97895397,\"identity\":\"be70b964-08a9-40d5-9a4f-6dfdb801d598\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:34:07\",\"extension\":\"csv\",\"order_by\":14,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":15288,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS12.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/f3d23de05e988b2077f70c6b.csv\"},{\"id\":97895564,\"identity\":\"694beb11-35e0-4331-938c-90fa01fc29ea\",\"added_by\":\"auto\",\"created_at\":\"2025-12-10 15:34:27\",\"extension\":\"csv\",\"order_by\":15,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":12127,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS13.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/81a89e357983c6e9148f2075.csv\"},{\"id\":97745081,\"identity\":\"680b401d-343e-461c-8ab8-ffaec672e2ef\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":16,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":136260,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS14.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/0df62d33b6096b10afe9d040.csv\"},{\"id\":97745037,\"identity\":\"d70438f5-57ff-4865-9c48-233c0c06efc7\",\"added_by\":\"auto\",\"created_at\":\"2025-12-09 00:23:36\",\"extension\":\"csv\",\"order_by\":17,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":65503,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryTableS15.csv\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-8092905/v1/6445d5300eaa0ebdc7d08bee.csv\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Exploring biomarkers related to exosome in primary immune thrombocytopenia based on transcriptomics and experimental verification\",\"fulltext\":[{\"header\":\"1. Introduction\",\"content\":\"\\u003cp\\u003ePrimary immune thrombocytopenia (ITP), an acquired autoimmune hematologic condition, is defined by two key features: a significant decline in platelet levels and an increased propensity for bleeding[\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e]. Epidemiological studies report an annual incidence rate of 2\\u0026ndash;10 cases per 100,000 population[\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e]. ITP can occur at any age, with pediatric cases typically presenting as acute forms often secondary to infections, while adult cases predominantly manifest as chronic forms, showing a slightly higher incidence in females than males[\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e]. The pathogenesis of ITP involves complex mechanisms, with three primary contributing factors: autoantibody production against platelets, impaired megakaryocyte maturation, and T-lymphocyte dysfunction[\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e]. Current therapeutic strategies primarily employ corticosteroids and intravenous immunoglobulin to rapidly elevate platelet counts, though these interventions demonstrate limited long-term efficacy[\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e]. The diagnosis of ITP remains exclusion-based due to the absence of specific biomarkers, potentially leading to misdiagnosis in some cases and consequent exacerbation of bleeding risks. Therefore, the identification of novel biomarkers holds significant value for improving ITP diagnosis and treatment.\\u003c/p\\u003e\\u003cp\\u003eExosomes are nanosized extracellular vesicles containing nucleic acids, proteins, and lipid components that mediate intercellular communication and participate in diverse biological processes[\\u003cspan additionalcitationids=\\\"CR11\\\" citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e]. Emerging evidence demonstrates their pivotal role in autoimmune diseases including multiple sclerosis and systemic lupus erythematosus, where their expression levels serve as valuable biomarkers for disease progression while also exhibiting potential as drug delivery vehicles[\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e]. A seminal study revealed that exosomes derived from ITP patients carry miR-363-3p, which disrupts immune homeostasis by significantly impairing the immunosuppressive function of regulatory T cells through modulation of the TBX21/ARID3A/SPI1 signaling axis[\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e]. Furthermore, specific proteins and miRNAs within exosomes have been implicated in ITP pathogenesis[\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e]. These molecular components are important drivers of disease progression, yet their regulatory mechanisms are not fully understood and need more research.\\u003c/p\\u003e\\u003cp\\u003eWe utilized a multi-omics framework that combined transcriptome profiling and bioinformatics to discover ITP-related exosomal genetic signatures. Machine learning was subsequently employed to screen and validate diagnostic biomarkers. The comprehensive analytical framework encompasses clinical prediction modeling through nomogram construction, functional enrichment analysis of biological pathways, molecular network mapping to elucidate regulatory mechanisms, immune microenvironment characterization via infiltration profiling, and computational drug prediction with molecular docking simulations for therapeutic target exploration. Experimental validation was subsequently performed using reverse transcription quantitative polymerase chain reaction (RT-qPCR) on peripheral blood samples from ITP patients. This translational research paradigm aims to discover novel exosome-derived biomarkers with enhanced diagnostic precision, ultimately advancing the scientific foundation for personalized ITP management strategies.\\u003c/p\\u003e\"},{\"header\":\"2. Materials and methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.1 Data source\\u003c/h2\\u003e\\u003cp\\u003eIn this study, the primary ITP-related dataset GSE43179 (sequencing type: microarray) was obtained 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). This dataset contains molecular expression data based on two different sequencing platforms. Among them, mRNA expression data were derived from the GPL570 platform, including 9 peripheral blood T-cell samples from ITP patients (ITP group) and 10 peripheral blood T-cell samples from normal controls (control group). miRNA expression data were obtained from the GPL14613 platform, comprising 9 peripheral blood T-cell samples from ITP patients (ITP group) and 9 peripheral blood T-cell samples from normal controls (control group). In addition, a total of 121 exosome-related genes (ERGs) were retrieved from the ExoBCD (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://exobcd.liumwei.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://exobcd.liumwei.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) and used for subsequent analyses (\\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003e\\u003c/b\\u003e)[\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e].\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.2 Differential expression analysis\\u003c/h2\\u003e\\u003cp\\u003eFor screening ITP-related differentially expressed genes (DEGs), differential expression analysis was initially performed on samples of the ITP and control groups in dataset GSE43179 (Sequencing platform: GPL570) via the R package \\\"limma\\\" (v 3.54.0)[\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e], identifying DEGs (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05, |log\\u003csub\\u003e2\\u003c/sub\\u003efold change (FC)| \\u0026gt;0.25). enerated for the DEGs using \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e].\\u003c/p\\u003e\\u003cp\\u003eVisualization of the differential analysis included a heatmap and a volcano plot. The heatmap, created with \\\"pheatmap\\\" (v 1.0.12)[\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e], depicted DEG expression between ITP and control groups, with the top 10 upregulated and downregulated genes (by |log\\u003csub\\u003e2\\u003c/sub\\u003eFC|) highlighted. The volcano plot was generated using \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. Genes were sorted identically by |log\\u003csub\\u003e2\\u003c/sub\\u003eFC| (descending), and the top 10 DEGs mentioned above were labeled in the plot.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.3 WGCNA\\u003c/h2\\u003e\\u003cp\\u003eSubsequently, the ssGSEA and WGCNA were comprehensively employed to expand exosome-related genes. For visualization of the significant inter-group disparities (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) in DE-ERG scores identified by the Wilcoxon rank-sum test, box plots were generated using the R package \\u0026ldquo;ggplot2\\u0026rdquo; (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. To further explore co-expression modules highly associated with exosomes, WGCNA was performed based on the aforementioned DE-ERG enrichment scores. The \\\"hclust\\\" function was employed for hierarchical clustering of all samples in dataset GSE43179 (sequencing platform: GPL570), aiming to identify and exclude outlier samples and guarantee data quality for subsequent co-expression network building. A sample clustering tree was also generated to visualize inter-sample relationships. Next, the optimal soft threshold (power value) was selected within the 1\\u0026ndash;20 parameter range. The best soft threshold that satisfied the scale-free network fitting index R\\u0026sup2; \\u0026gt;0.85 was selected. A co-expression network was built based on the optimal soft threshold, and the dynamic tree cutting algorithm in the R package \\\"WGCNA\\\" (v 1.73)[\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e] was employed with the following parameter settings: minimum number of module genes set to 200, method = \\\"tree\\\", and cutHeight\\u0026thinsp;=\\u0026thinsp;0.99, to divide genes into modules labeled with different colors. To identify modules highly correlated with exosome traits, the Spearman correlation coefficient between module eigengenes and DE-ERG scores was calculated (|correlation coefficient (cor)| \\u0026gt;0.3 and P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05). Meanwhile, a module-trait correlation heatmap was plotted using the labeledHeatmap function in the R package \\\"WGCNA\\\" (v 1.73)[\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e] for visualization. Finally, we selected the modules most strongly correlated with DE-ERG scores. Using the \\\"WGCNA\\\" R package (v 1.73)[\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e], Module Membership (MM) and Gene Significance (GS) were calculated for these module genes. Genes meeting the thresholds of |MM| \\u0026gt;0.6 and |GS| \\u0026gt;0.4 were retained. The GS-MM correlation was visualized in scatter plots via \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e], and genes fulfilling these criteria were defined as exosome-related key module genes.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec6\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.4 Acquisition of candidate genes and functional enrichment analysis\\u003c/h2\\u003e\\u003cp\\u003eWe next sought to identify genes linked to both ITP and exosomes. Using the R package \\\"ggvenn\\\" (v 1.7.1), we found the intersection of DEGs and exosome-related key module genes, thereby acquiring candidate genes for follow-up functional verification. The R package \\\"clusterProfiler\\\" (v 4.2.2)[\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e] was employed to perform KEGG and GO enrichment analyses on the candidate genes (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05), aiming to explore their potential biological functions and underlying mechanisms. GO consisted of three parts, namely biological process (BP), cellular component (CC) and molecular function (MF). Additionally, STRING (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://string-db.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://string-db.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) was utilized to examine interactions of proteins encoded by candidate genes (confidence score\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.15). Finally, the results were visualized by constructing a PPI network diagram using Cytoscape software (v 3.8.2)[\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e].\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.5 Acquisition of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eFor the refined screening of diagnostically relevant ITP biomarkers, a two-step algorithm approach was utilized: Boruta followed by LASSO regression. Feature selection via the \\\"Boruta\\\" package (v 8.0.0)[\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e]on the GSE43179 dataset (GPL570) identified genes with importance scores exceeding all shadow feature benchmarks (shadowMin, shadowMean, shadowMax), yielding Feature Gene Set 1. Subsequent analysis involved applying LASSO regression to the same dataset with the \\\"glmnet\\\" package (v 1.1.10)[\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e]. Via 5-fold cross-validation, the optimal regularization parameter (lambda) (minimum mean squared error) was determined. Genes with non-zero regression coefficients were screened to generate Feature Gene Set 2, and regression coefficient plots plus cross-validation error curves were generated for visualization. Finally, the two feature gene sets\\u0026rsquo; intersection was calculated using \\\"ggvenn\\\" (v 1.7.1) to acquire the final biomarkers.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.6 Construction and evaluation of nomogram\\u003c/h2\\u003e\\u003cp\\u003eTo evaluate biomarkers\\u0026rsquo; ITP predictive reliability, we established a nomogram prediction model based on biomarkers using the R package \\\"rms\\\" (v 6.8.1)[\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e] and all samples from GSE43179 (Sequencing platform: GPL570). The scale range of the line corresponding to each biomarker indicates its predicted score interval (Points). The total score (Total Points) results from summing the scores of each biomarker, and a higher value of it corresponds to a higher probability of ITP developing. Predictive accuracy of the model was verified through two methods: calibration curves plotted using the \\\"rms\\\" R package (v 6.8.1)[\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e]and the Hosmer-Lemeshow test for assessing prediction-actuality consistency. In these curves, prediction accuracy was considered higher when the plot more closely followed the reference line (slope\\u0026thinsp;=\\u0026thinsp;1). The nomogram demonstrated good calibration accuracy for ITP risk prediction, as evidenced by an HL test P value\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.05, which denotes no significant difference between predicted and actual values. The predictive accuracy was further assessed by constructing an ROC curve with the \\\"pROC\\\" R package (v 1.18.5)[\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e]. AUC (range: 0\\u0026ndash;1) served as a discriminatory metric, with values above 0.7 deemed to indicate good predictive performance.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.7 GSEA of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eTo investigate the regulatory pathways and biological functions linked to biomarkers, GSEA was initially performed for each biomarker on ITP vs. control samples from dataset GSE43179 (Sequencing platform: GPL570). Employing the R package \\\"psych\\\" (v 2.1.6)[\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e], we computed and sorted Spearman correlation coefficients between biomarkers and other genes. GSEA was subsequently performed with the R package \\\"clusterProfiler\\\" (v 4.2.2)[\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e] using the MsigDB (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.gsea-msigdb.org/gsea/msigdb/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.gsea-msigdb.org/gsea/msigdb/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) gene set \\\"c2.cp.kegg_legacy.v2024.1.Hs.symbols.gmt\\\" as reference, under the criteria |NES| \\u0026gt;1 and adj.P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05. Visualization of the top 5 significant pathways was finally achieved with the R package \\\"enrichplot\\\" (v 1.18.3)[\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e].\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec10\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.8 GeneMANIA analysis and of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eSubsequently, related genes interacting with biomarkers and their involved functions were predicted via the GeneMANIA database (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://genemania.org/\\u003c/span\\u003e\\u003cspan address=\\\"http://genemania.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), with \\\"Homo sapiens\\\" set as the analysis species. From the outcomes, biomarkers and the top 20 related genes were selected as key display nodes, and the top 5 most significant enriched pathways (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) and functions (FDR\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) were presented. A gene interaction network diagram was ultimately constructed to visualize results, clearly showing complex correlations between biomarkers and their functionally similar genes.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.9 Immune infiltration analysis\\u003c/h2\\u003e\\u003cp\\u003eITP is an immune-mediated autoimmune disease, primarily marked by impaired immune system function. This impairment prompts autoantibodies or cytotoxic T cells to attack platelets, thereby boosting platelet destruction and hindering platelet production[\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e]. To evaluate differences in immune cell infiltration levels between the ITP group and the control group, based on all samples from the dataset GSE43179 (Sequencing platform: GPL570), the relative infiltration abundances of 22 immune cells were calculated using the R package \\\"CIBERSORT\\\" (v2.24.0)[\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e] in this study. Following sample filtration (P\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.05) to ensure reliability, immune cell infiltration percentages were visualized in stacked bar charts. Differences in infiltration between groups were further identified using the Wilcoxon rank-sum test (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) and illustrated in box plots, with both visualizations created using the R package \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. Subsequent correlation analysis, conducted with the R package \\\"psych\\\" (v 2.1.6)[\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e], examined relationships between biomarkers and these differential immune cells, considering correlations with |cor| \\u0026gt;0.3 and P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05 as significant.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.10 Molecular regulatory network analysis of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eTranscription factors (TFs) are key protein molecules that can specifically recognize and bind to specific DNA sequences in regulatory regions such as gene promoters or enhancers, thereby regulating the transcriptional activity of genes. First, the TRRUST database (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://link.zhihu.com/?target=http%3A//www.grnpedia.org/trrust/\\u003c/span\\u003e\\u003cspan address=\\\"https://link.zhihu.com/?target=http%3A//www.grnpedia.org/trrust/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) was used to predict TFs that interact with biomarkers. MicroRNAs (miRNAs) primarily mediate the regulation of protein expression by binding to one or more target sites on mRNA transcripts and inhibiting translation. They are involved in various developmental pathways and multiple gene regulatory mechanisms, and play a role in a series of disease processes and phenotypic determination. Next, based on the GSE43179 dataset (sequencing platform: GPL14613), the R package \\\"limma\\\" (v 3.54.0)[\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e] was used to identify differentially expressed microRNAs (DE-miRNAs) between the ITP group and the control group. MiRNAs in the heatmap were sorted by |log\\u003csub\\u003e2\\u003c/sub\\u003eFC| values in descending order, showing the top 10 upregulated and 10 downregulated miRNAs with the most significant expression differences. A miRNA volcano plot was also created using the R package \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. Similarly, genes were sorted by descending |log\\u003csub\\u003e2\\u003c/sub\\u003eFC|, with the plot labeling the top 10 up- and downregulated miRNAs (most significant expression differences). Subsequently, miRNet (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.mirnet.ca/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.mirnet.ca/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) predicted miRNAs targeting biomarkers. The final integrative visualization of the regulatory network, built from the screened TFs, key miRNAs, and biomarkers, was achieved using Cytoscape software (v 3.8.2)[\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e].\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec13\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.11 Drug prediction analysis of biomarkers and molecular docking\\u003c/h2\\u003e\\u003cp\\u003eTo identify potential ITP drugs, drug prediction for each biomarker was done via DGIdb (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://dgidb.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://dgidb.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). Construction and visualization of the drug-mRNA interaction network were performed with Cytoscape software (v 3.8.2)[\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e]. To validate targeted drug-biomarker binding, we selected drugs per biomarker (highest interaction scores, PubChem-retrievable 2D structures) for molecular docking. Drug 2D structures (SDF) were downloaded from PubChem (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://pubchem.ncbi.nlm.nih.gov/\\u003c/span\\u003e\\u003cspan address=\\\"https://pubchem.ncbi.nlm.nih.gov/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), and biomarker-encoded protein 3D structures (PDB) from RCSB PDB (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.rcsb.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.rcsb.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). These were uploaded to CB-Dock (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://cadd.labshare.cn/cb-dock2/php/index.php\\u003c/span\\u003e\\u003cspan address=\\\"https://cadd.labshare.cn/cb-dock2/php/index.php\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) for docking, with binding affinity evaluated by Vina score (binding free energy). Generally, Vina score \\u0026lt; -5 kcal/mol indicates high ligand-receptor affinity and stable complex conformation.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec14\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.12 Expression level validation of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eDifferences in biomarker expression within the GSE43179 dataset (GPL570) were assessed using the Wilcoxon rank-sum test (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) and subsequently visualized in box plots constructed with the R package \\\"ggplot2\\\" (v 3.4.1)[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e].\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec15\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.13 RT-qPCR\\u003c/h2\\u003e\\u003cp\\u003e\\u003cul\\u003e\\u003cli\\u003e\\u003cp\\u003eValidation of biomarker expression in clinical samples was performed by RT-qPCR. Ten blood samples (5 from ITP patients and 5 from controls) were obtained from Fengdu County Hospital of Traditional Chinese Medicine. Total RNA was extracted using TRIzol reagent (Ambion, USA) per the manufacturer's protocol, and its concentration was determined with a NanoPhotometer N50. cDNA was synthesized by reverse transcription employing the SureScript First-Strand cDNA Synthesis Kit on a Bio-Rad S1000\\u0026trade; Thermal Cycler (USA). Using primers listed in \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003e\\u003c/b\\u003e, qPCR was run on a Bio-Rad CFX Connect Real-Time PCR Instrument (USA) with the cycling profile: 95\\u0026deg;C for 1 min; 40 cycles of 95\\u0026deg;C for 20s, 55\\u0026deg;C for 20s, and 72\\u0026deg;C for 30s. The 2-ΔΔCT method was applied for relative quantification, and data were analyzed and visualized in GraphPad Prism 5 (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.graphpad.com/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.graphpad.com/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) after initial export to Excel.\\u003c/p\\u003e\\u003c/li\\u003e\\u003c/ul\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec16\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.14 Statistical analysis\\u003c/h2\\u003e\\u003cp\\u003eBioinformatics analyses utilized R software (v 4.2.2) for statistical processing. Statistical significance was set at P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05. The Wilcoxon rank-sum test was used for between-group comparisons, while the t-test was employed for between-group analyses of RT-qPCR results.\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec17\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e2.15 Ethics approval and consent to participate\\u003c/h2\\u003e\\u003cp\\u003eThe human subjects research component of this study was reviewed and approved by the Ethics Committee of Fengdu County Hospital of Traditional Chinese Medicine, Chongqing, China (Approval No.: FDYY\\u0026minus;2024\\u0026minus;0729), with documented informed consent obtained from all participants. All procedures were conducted in strict compliance with the ethical principles set forth in the Declaration of Helsinki (1964) and its subsequent amendments.\\u003c/p\\u003e\\u003c/div\\u003e\"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec19\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.1 Acquisition and functional enrichment analysis of 23 candidate genes\\u003c/h2\\u003e\\u003cp\\u003eA total of 105 DEGs were identified from the GSE43179 dataset (GPL570) through a comparative analysis between ITP and control groups, using thresholds of P \\u0026lt; 0.05 and |log\\u003csub\\u003e2\\u003c/sub\\u003eFC| \\u0026gt;0.25. The ITP-associated DEGs comprised 48 upregulated and 57 downregulated genes (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eA-B, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM3\\\" class=\\\"InternalRef\\\"\\u003eS3\\u003c/span\\u003e\\u003c/b\\u003e). Expression levels of 121 ERGs were then compared between ITP and control groups in the GSE43179 dataset (GPL570) via Wilcoxon rank-sum test. This screening identified seven DE-ERGs: POLR2K, IGF1R, NANOG, RAB13, ALDH9A1, HDC, and TLR5 (P \\u0026lt; 0.05) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eC). Further, DE-ERG scores were calculated based on all samples from the dataset GSE43179 (Sequencing platform: GPL570). The Wilcoxon rank-sum test showed that there was a significant difference in DE-ERG scores between the ITP group and the control group (P = 0.011) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eD). No obvious outlier samples were identified by hierarchical clustering analysis, indicating that the data quality was reliable (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eE). A scale-free co-expression network was successfully constructed by WGCNA with a soft-thresholding power of 19. This parameter selection resulted in a topological fit (R² = 0.85) and a mean connectivity approaching zero (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eF-G). After constructing the weighted co-expression network based on the selected power value, the dynamic tree cutting method was used to identify gene modules. The minimum number of genes in each module was set to 200, and after merging highly similar modules (cutHeight = 0.99), a total of 6 co-expression modules were identified (excluding the grey module) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eH). Next, the correlations between each module and DE-ERG scores were analyzed, and 4 modules were found to have a significant correlation with DE-ERG scores (|cor| \\u0026gt;0.3, P \\u0026lt; 0.05). Among the modules, the brown module demonstrated the strongest positive correlation with DE-ERG scores (cor = 0.67, P \\u0026lt; 0.05), whereas the red and blue modules were most negatively correlated (cor = -0.64, P \\u0026lt; 0.05) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eI). In addition, it was further observed that there were highly significant correlations between GS and MM in the brown module (cor = 0.581, P \\u0026lt; 0.05), red module (cor = 0.784, P \\u0026lt; 0.05), and blue module (cor = -0.723, P \\u0026lt; 0.05) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eJ). This indicated that the genes within the modules had good consistency with the phenotypic characteristics. Finally, genes were screened according to the criteria of |MM| \\u0026gt;0.6 and |GS| \\u0026gt;0.4. A total of 3788, 767, and 159 genes were obtained from the blue, brown, and red modules respectively, resulting in a total of 4714 exosome-related key module genes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eK, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM4\\\" class=\\\"InternalRef\\\"\\u003eS4\\u003c/span\\u003e\\u003c/b\\u003e).\\u003c/p\\u003e\\u003cp\\u003eThe intersection of DEGs and exosome-related key module genes yielded 23 candidate genes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eK, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM5\\\" class=\\\"InternalRef\\\"\\u003eS5\\u003c/span\\u003e\\u003c/b\\u003e). To investigate their potential functions, these candidate genes were subjected to GO and KEGG enrichment analyses, which identified 212 significant GO terms (P \\u0026lt; 0.05), including 151 BP terms such as response to starvation and transcription by RNA polymerase III (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eL, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM6\\\" class=\\\"InternalRef\\\"\\u003eS6\\u003c/span\\u003e\\u003c/b\\u003e); 11 CC terms such as transferase complex, transferring phosphorus-containing groups and RNA polymerase II, holoenzyme (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eL, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM7\\\" class=\\\"InternalRef\\\"\\u003eS7\\u003c/span\\u003e\\u003c/b\\u003e); and 50 MF terms such as NADP binding and Tat protein binding (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eL, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM8\\\" class=\\\"InternalRef\\\"\\u003eS8\\u003c/span\\u003e\\u003c/b\\u003e). KEGG enrichment analysis revealed that the candidate genes were significantly enriched in 7 signaling pathways (P \\u0026lt; 0.05), including FoxO signaling pathway, mTOR signaling pathway, and autophagy - animal, and other pathways (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eM, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM9\\\" class=\\\"InternalRef\\\"\\u003eS9\\u003c/span\\u003e\\u003c/b\\u003e). Subsequently, the interaction relationships between proteins encoded by the candidate genes were analyzed via the STRING database (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.string-db.org\\u003c/span\\u003e\\u003cspan address=\\\"https://www.string-db.org\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) with a confidence score \\u0026gt; 0.15. From the analysis results, a tightly interacting network containing 18 core protein nodes was identified. Among these nodes, high-strength connections were observed between SIMCI and POLR2K, TCF20, as well as EIF4ENIF1. This suggested that these proteins may have a tendency toward functional synergy or complex formation, indicating that these core proteins might collectively participate in key biological processes during the disease progression of ITP (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eN). The candidate genes and enriched pathways identified in this part can provide a theoretical basis for potential therapeutic targets and mechanistic studies of ITP.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec20\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.2 Identification of 4 biomarkers and construction of nomogram\\u003c/h2\\u003e\\u003cp\\u003eTo further screen reliable biomarkers from 23 candidate genes, two machine learning algorithms—Boruta analysis and LASSO regression analysis—were applied in this study for biomarkers screening. Based on all samples from the dataset GSE43179 (Sequencing platform: GPL570), first, Feature Gene Set 1—comprising 5 genes (GABARAPL1, SLC39A14, HIBADH, FUT11, and GSR)—was obtained via Boruta analysis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA). Next, for the LASSO regression analysis, the optimal regularization parameter was determined through 5-fold cross-validation. When lambda.min = 0.175, the model achieved the minimum Mean Squared Error (MSE); subsequently, feature gene set 2—consisting of 5 genes (GABARAPL1, SLC39A14, HIBADH, IGF1R, and GSR)—was obtained (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB-C). By taking the intersection of these two feature gene sets, 4 biomarkers were finally identified for subsequent analysis, namely GABARAPL1, SLC39A14, HIBADH, and GSR (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eD). Subsequently, the nomogram model constructed based on the 4 biomarkers showed good predictive performance. For example, when the total points reached 229, the predicted probability of ITP was 89% (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eE). The nomogram's performance was assessed using multiple indicators, demonstrating strong predictive ability. Its good calibration was reflected in a non-significant Hosmer-Lemeshow test result (P = 0.168) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eF). And an AUC of 0.878 in ROC analysis confirming its high discriminatory power (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eG). This study successfully identified diagnostic biomarkers and constructed a reliable nomogram, thereby providing a theoretical foundation for the early diagnosis and personalized management of ITP.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec21\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.3 GSEA and GeneMANIA analysis of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eGSEA of the biomarkers was performed to identify relevant signaling pathways and biological mechanisms underlying ITP (|NES| \\u0026gt;1, adj.p \\u0026lt; 0.05). It was found in the study that GABARAPL1 was mainly enriched in 30 pathways, including spliceosome and neuroactive ligand receptor interaction, among others (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM10\\\" class=\\\"InternalRef\\\"\\u003eS10\\u003c/span\\u003e\\u003c/b\\u003e); SLC39A14 was mainly enriched in 29 pathways, including ubiquitin-mediated proteolysis and neuroactive ligand receptor interaction, among others (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eB, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM11\\\" class=\\\"InternalRef\\\"\\u003eS11\\u003c/span\\u003e\\u003c/b\\u003e); HIBADH was mainly enriched in 35 pathways, including spliceosome and neuroactive ligand receptor interaction, among others (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eC, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM12\\\" class=\\\"InternalRef\\\"\\u003eS12\\u003c/span\\u003e\\u003c/b\\u003e). GSR was mainly enriched in 26 pathways, including ribosome and spliceosome, among others (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eD, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM13\\\" class=\\\"InternalRef\\\"\\u003eS13\\u003c/span\\u003e\\u003c/b\\u003e). Subsequently, genes interacting with the four biomarkers and the biological processes they co-participated in were predicted using the GeneMANIA database (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://genemania.org/\\u003c/span\\u003e\\u003cspan address=\\\"http://genemania.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). The results showed that the top 20 genes interacting with the four biomarkers included HIBCH, GSTO2, ATG4D, ALDH6A1, PGD, ATG7, GLRX, ATP6V1C1, ATG4A, G6PD, DLD, FOLR1, TXN, SLC4A4, HSD17B10, GLYR1, STBD1, AIFM1, RETREG3, and SLC25A20. Among these, the biomarker HIBADH, together with genes HIBCH, ALDH6A1, and HSD17B10, was found to co-participate in the branched-chain amino acid metabolic process, branched-chain amino acid catabolic process, and alpha-amino acid catabolic process. The biomarker GSR, along with genes TXN and GLRX, was involved in oxidoreductase activity (acting on a sulfur group of donors) and disulfide oxidoreductase activity. Notably, the associations between the four biomarkers and their interacting genes were mainly based on physical interactions, suggesting that these biomarkers might function in protein complex assembly or direct binding (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eE). Taken together, GSEA and GeneMANIA analyses implicated the four biomarkers in ITP disease mechanisms, potentially through their influence on essential biological processes like spliceosome function, amino acid metabolism, redox equilibrium, and the ubiquitin-proteasome system. These insights thereby furnish important clues for subsequent research to unravel their molecular underpinnings and to pinpoint potential therapeutic targets.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec22\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.4 Immune infiltration analysis\\u003c/h2\\u003e\\u003cp\\u003eAnalysis of the GSE43179 dataset (GPL570) revealed distinct immune cell infiltration profiles in ITP (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eA). Wilcoxon test identified three differentially abundant immune cells: activated NK cells, plasma cells, and monocytes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB). Significant biomarker-immune cell correlations were uncovered (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC): GABARAPL1 was positively correlated with plasma cells (cor = 0.74) and negatively with activated NK cells (cor=-0.50); SLC39A14 and HIBADH both showed negative correlations with activated NK cells (cor=-0.69 and − 0.74, respectively), with SLC39A14 also correlating negatively with monocytes (cor=-0.53) (all P \\u0026lt; 0.05). These results underscore the immunoregulatory importance of these biomarkers, providing mechanistic insights and a theoretical foundation for targeted ITP therapy.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec23\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.5 Molecular regulatory network analysis of biomarkers\\u003c/h2\\u003e\\u003cp\\u003eThe potential regulatory mechanisms of the biomarkers were further revealed by molecular regulatory network analysis. Upstream TFs that regulate the biomarkers were first predicted, and the results showed that GABARAPL1, GSR, HIBADH, and SLC39A14 were found to be regulated by 20, 47, 19, and 35 TFs, respectively (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eA). This finding indicated that these biomarkers are subject to complex and diverse regulatory networks at the transcriptional level. To further investigate the post-transcriptional regulatory mechanisms, differentially expressed miRNAs (DE-miRNAs) between the disease group and the control group were analyzed based on the GSE43179 dataset (sequencing platform: GPL14613). A total of 45 DE-miRNAs were identified, among which 35 miRNAs were downregulated and 10 miRNAs were upregulated in the Immune Thrombocytopenia (ITP) group, with the screening criteria of |log\\u003csub\\u003e2\\u003c/sub\\u003eFC| \\u0026gt;0.25 and P \\u0026lt; 0.05 (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eB-C, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM14\\\" class=\\\"InternalRef\\\"\\u003eS14\\u003c/span\\u003e\\u003c/b\\u003e). Meanwhile, screening of the miRNet database yielded 490 miRNAs predicted to target the biomarkers (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eD, \\u003cb\\u003eSupplementary Table \\u003cspan refid=\\\"MOESM15\\\" class=\\\"InternalRef\\\"\\u003eS15\\u003c/span\\u003e\\u003c/b\\u003e). These results were provided as important theoretical bases and resource clues for an in-depth understanding of the molecular regulatory networks (including miRNA and TF levels) of GABARAPL1, GSR, HIBADH, and SLC39A14 in the pathological process of ITP.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec24\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.6 Drug prediction analysis of biomarkers and molecular docking\\u003c/h2\\u003e\\u003cp\\u003eTo identify potential ITP therapeutics, drug prediction analysis nominated 22 and 1 candidate drugs for GSR and SLC39A14, respectively (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eA). Subsequent molecular docking revealed favourable binding affinities between the biomarkers and their respective drugs: SLC39A14 bound to nortriptyline with a free energy of -7.9 kcal/mol (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eB-C, \\u003cb\\u003eTable\\u0026nbsp;1\\u003c/b\\u003e), while GSR bound to oxiglutatione at -17.1 kcal/mol (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eD, \\u003cb\\u003eTable\\u0026nbsp;1\\u003c/b\\u003e), both values being far lower than the − 5 kcal/mol threshold. This suggested that SLC39A14 and GSR may be specific targets for these drugs, offering insights into drug action and a basis for future targeted therapy development.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\\u003cdiv id=\\\"Sec25\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003e3.7 Expression level validation of biomarkers and RT-qPCR\\u003c/h2\\u003e\\u003cp\\u003eAnalysis of the GSE43179 dataset (Sequencing platform: GPL570) revealed that all four biomarkers were significantly downregulated in ITP samples (P \\u0026lt; 0.05) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eA). This consistent downregulation suggests a potential role in ITP pathogenesis. Subsequent validation in clinical samples by RT-qPCR confirmed that GABARAPL1, SLC39A14, and GSR were significantly downregulated in ITP (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eB-E). These results verify the association between these biomarkers and ITP, indicating that their decreased expression may play a role in disease mechanisms.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003c/div\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eITP is an autoimmune disorder characterized by both humoral and cellular immune-mediated platelet destruction coupled with impaired platelet production[\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e]. Emerging evidence indicates that exosomes play a pivotal role in ITP pathogenesis through immunomodulation and intercellular communication[\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e]. Nevertheless, the precise etiological mechanisms underlying ITP remain incompletely understood, and reliable diagnostic criteria are still lacking. This study employed bioinformatics approaches to systematically identify exosome-associated biomarkers and elucidate their functional mechanisms in ITP. Experimental validation in peripheral blood samples from ITP patients confirmed dysregulated expression patterns of three core genes - GABARAPL1, SLC39A14, and GSR - which demonstrate significant involvement in ITP pathogenesis. These findings suggest that GABARAPL1, SLC39A14, and GSR may serve as potential diagnostic biomarkers for ITP. However, the expression pattern of HIBADH was not consistently validated across clinical samples, possibly attributable to variations in sample size and patient characteristics.\\u003c/p\\u003e\\u003cp\\u003eGABARAPL1 serves as a critical regulatory protein in autophagy processes, playing an indispensable role in maintaining intracellular homeostasis and ensuring cellular survival[\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e]. Research by Lei Li et al.[\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e] has demonstrated that GABARAPL1 modulates platelet production through its regulation of autophagy levels in megakaryocytes. Ahmad Reza Panahi Meymandi et al.[\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e] further established that GABARAPL1 dysfunction may promote the survival and proliferation of autoreactive lymphocytes, consequently disrupting immune tolerance to platelet antigens. The autophagy process exhibits extensive crosstalk with exosome biogenesis, which may amplify autoimmune responses against platelets[\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e]. Notably, studies have revealed significantly reduced GABARAPL1 expression in ITP patients[\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e], potentially contributing to disease progression through multiple interconnected mechanisms: impairment of ubiquitin phosphorylation, dysregulation of PPAR signaling pathways, compromised mitophagy, and suppression of ferroptosis. Our current findings corroborate these observations, demonstrating markedly lower GABARAPL1 expression in ITP patients compared to healthy controls. These collective insights suggest that GABARAPL1 likely influences ITP pathogenesis through its dual regulatory effects on megakaryocyte function and immune cell homeostasis.\\u003c/p\\u003e\\u003cp\\u003eSLC39A14 functions as a crucial zinc transporter that plays a pivotal role in maintaining immune homeostasis[\\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e]. Zinc deficiency or metabolic dysregulation can promote inflammatory responses and contribute to autoimmune pathogenesis. Alterations in SLC39A14 function or expression may disrupt zinc homeostasis, thereby participating in the immune dysregulation observed in ITP[\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e]. Zhizhao Deng et al.[\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e] reported a potential association between exosomes and SLC39A14, demonstrating that bone marrow mesenchymal stem cell-derived exosomes deliver miR-16-5p to hepatocytes, leading to post-transcriptional suppression of SLC39A14 expression. Exosomes released from cells with low SLC39A14 expression can modify intracellular zinc levels and exacerbate inflammatory responses[\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e]. Activated immune cells may secrete exosomes carrying specific miRNAs or autoantigens, which upon uptake by other immune cells, could potentially accelerate ITP progression[\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e]. In our current study, we observed significantly reduced SLC39A14 expression in ITP patients compared to healthy controls. These findings suggest that SLC39A14 likely contributes to ITP pathogenesis through its involvement in zinc metabolism dysregulation.\\u003c/p\\u003e\\u003cp\\u003eHIBADH serves as a pivotal enzyme in valine catabolism, catalyzing the oxidation of 3-hydroxyisobutyrate to methylmalonate semialdehyde, thereby playing an essential role in energy metabolism and amino acid degradation[\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e]. Emerging evidence suggests that defects in energy production and biosynthetic pathways may lead to immune cell dysfunction. Wenwei Chen et al.[\\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e] demonstrated that HIBADH exerts protective effects through its regulation of mitochondrial function and mitochondria-associated oxidative stress. Altered HIBADH expression can induce significant changes in metabolite concentrations, potentially exerting undefined regulatory effects on immune cells that may contribute to ITP progression[\\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e]. Yuanlan Huang et al.[\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e] further proposed that decreased HIBADH expression might reduce the abundance of regulatory molecules carried by exosomes, consequently diminishing their protective effects on target cells and potentially exacerbating autoimmune responses. While our study observed lower HIBADH expression levels in the initial screening compared to healthy controls, this finding was not consistently validated in peripheral blood samples from ITP patients, possibly due to sample heterogeneity. These collective findings suggest that HIBADH may participate in ITP pathogenesis through its role in maintaining mitochondrial function and redox homeostasis.\\u003c/p\\u003e\\u003cp\\u003eGSR serves as a pivotal enzyme in the cellular antioxidant defense system, playing a critical role in maintaining redox homeostasis[\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e]. Patients with ITP exhibit a pronounced state of oxidative stress. Yanxia Zhan et al.[\\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e] demonstrated that elevated reactive oxygen species (ROS) levels can promote inflammatory cytokine release through immune cell activation while simultaneously causing direct damage to both platelets and megakaryocytes. Complementary research by Yuquan Xie et al.[\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e] revealed that excessive ROS directly attacks platelet membrane phospholipids, thereby triggering platelet apoptosis. Further investigations have shown that exosomes secreted by immune cells under oxidative stress conditions may carry increased loads of inflammatory non-coding RNAs[\\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e]. When internalized by megakaryocytes, these exosomes could amplify oxidative stress and inflammatory signaling, thereby exacerbating autoimmune responses. Our current study identified significantly reduced GSR expression in ITP patients compared to healthy controls. These collective findings strongly suggest that GSR contributes to ITP pathogenesis through its dual roles in antioxidant defense and potential modulation of exosome-mediated intercellular communication pathways that sustain autoimmune responses.\\u003c/p\\u003e\\u003cp\\u003eThe spliceosome, a macromolecular complex composed of small nuclear ribonucleoproteins and auxiliary proteins, plays a critical role in mRNA processing, with its dysfunction being implicated in various hematological disorders[\\u003cspan citationid=\\\"CR54\\\" class=\\\"CitationRef\\\"\\u003e54\\u003c/span\\u003e]. Emerging evidence suggests that aberrant spliceosome regulation can lead to abnormal alternative splicing of platelet production-related genes in megakaryocytes[\\u003cspan citationid=\\\"CR55\\\" class=\\\"CitationRef\\\"\\u003e55\\u003c/span\\u003e]. Furthermore, spliceosome components may contribute to autoimmune responses through dysregulated activation of T and B lymphocytes, resulting in immune-mediated platelet destruction[\\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e]. Andrea Pellagatti et al.[\\u003cspan citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e] demonstrated that exosomes can transport spliceosome-associated non-coding RNAs, thereby modulating gene expression in recipient cells through intercellular communication. Our GSEA revealed that GABARAPL1, SLC39A14, HIBADH, and GSR are collectively enriched in the spliceosome signaling pathway. These findings suggest that exosomes derived from aberrant immune cells may transmit defective splicing signals, potentially impairing megakaryocyte differentiation and platelet maturation. Notably, these four biomarkers also show significant enrichment in the neuroactive ligand-receptor interaction pathway and ubiquitin-mediated proteolysis signaling pathway, indicating their potential involvement in multiple regulatory mechanisms in ITP pathogenesis.\\u003c/p\\u003e\\u003cp\\u003eMonocytes, as essential components of the innate immune system, play a pivotal role in inflammatory responses[\\u003cspan citationid=\\\"CR58\\\" class=\\\"CitationRef\\\"\\u003e58\\u003c/span\\u003e]. In recent years, their involvement in ITP pathogenesis has garnered increasing attention. Yajing Zhao et al.[\\u003cspan citationid=\\\"CR59\\\" class=\\\"CitationRef\\\"\\u003e59\\u003c/span\\u003e] demonstrated that elevated TNF-α mRNA expression in peripheral blood monocytes of ITP patients exacerbates immune dysregulation through activation of the NF-κB signaling pathway. Complementary research by Wei Wang et al.[\\u003cspan citationid=\\\"CR60\\\" class=\\\"CitationRef\\\"\\u003e60\\u003c/span\\u003e] revealed that GSK-3β inhibitors can attenuate monocyte phagocytic capacity by blocking TNF-α signaling, thereby ameliorating thrombocytopenia. Our immune infiltration analysis yielded several significant findings: First, SLC39A14 expression showed a strong negative correlation with monocyte infiltration levels. Second, GABARAPL1, SLC39A14, and HIBADH all demonstrated significant negative correlations with activated NK cells - a finding consistent with Farida Hussein El-Rashedi et al.'s[\\u003cspan citationid=\\\"CR61\\\" class=\\\"CitationRef\\\"\\u003e61\\u003c/span\\u003e] observations of NK cell activity in pediatric ITP patients. These collective results suggest that dysregulated monocyte and NK cell activities contribute substantially to immune homeostasis disruption and subsequent platelet destruction in ITP.\\u003c/p\\u003e\\u003cp\\u003emiR-484 has emerged as a critically important microRNA in various cancers and metabolic disorders, where it plays a regulatory role in apoptosis and mitochondrial function signaling pathways[\\u003cspan citationid=\\\"CR62\\\" class=\\\"CitationRef\\\"\\u003e62\\u003c/span\\u003e]. Existing research has established that miR-484 modulates apoptotic processes in immune cells[\\u003cspan citationid=\\\"CR63\\\" class=\\\"CitationRef\\\"\\u003e63\\u003c/span\\u003e], leading us to hypothesize that its dysregulated expression may contribute to T-cell homeostasis imbalance in ITP. Further mechanistic studies have demonstrated that miR-484 interferes with megakaryocyte differentiation and maturation by suppressing the Wnt/MAPK pathway and downregulating β-catenin expression[\\u003cspan citationid=\\\"CR64\\\" class=\\\"CitationRef\\\"\\u003e64\\u003c/span\\u003e]. Our bioinformatic analyses provide compelling evidence that miR-484 directly targets and regulates SLC39A14 expression. This regulatory interaction appears to exert dual pathological effects: impairing T-cell function and disrupting megakaryocyte development, ultimately leading to enhanced platelet destruction in ITP.\\u003c/p\\u003e\\u003cp\\u003eOur study conducted comprehensive drug prediction analyses for the four candidate biomarkers, revealing that only SLC39A14 and GSR possess target-specific therapeutic agents. Nortriptyline, a tricyclic antidepressant with immunomodulatory properties, has been shown to suppress pro-inflammatory cytokine production and modulate T-cell function, thereby potentially reducing immune-mediated platelet destruction[\\u003cspan citationid=\\\"CR65\\\" class=\\\"CitationRef\\\"\\u003e65\\u003c/span\\u003e]. Mechanistically, this drug may interact with SLC39A14 to influence zinc ion homeostasis, consequently regulating both immune cell activity and megakaryocyte differentiation. Oxiglutathione (GSSG), as a substrate for GSR-catalyzed reactions, plays a pivotal role in maintaining cellular antioxidant capacity[\\u003cspan citationid=\\\"CR66\\\" class=\\\"CitationRef\\\"\\u003e66\\u003c/span\\u003e]. In ITP patients, enhancing GSR activity could elevate intracellular reduced glutathione (GSH) levels, thereby mitigating oxidative damage to platelets. These findings suggest that nortriptyline and oxiglutathione represent promising therapeutic candidates targeting SLC39A14 and GSR respectively, with complementary mechanisms of action: nortriptyline primarily through immunomodulation and oxiglutathione via antioxidant pathways. Their combined use may offer a synergistic approach to address the multifactorial pathogenesis of ITP.\\u003c/p\\u003e\\u003cp\\u003eIn conclusion, this study provides compelling evidence that GABARAPL1, SLC39A14, HIBADH and GSR are functionally involved in the pathogenesis of ITP. Importantly, GABARAPL1, SLC39A14 and GSR emerge as particularly promising candidates for novel diagnostic biomarkers in ITP. While these findings advance our understanding of ITP pathology, certain limitations must be acknowledged regarding the current validation using peripheral blood samples with relatively small sample sizes and incomplete mechanistic exploration of disease pathogenesis. Future investigations employing expanded clinical cohorts will be essential to further validate the diagnostic potential of these biomarkers and more comprehensively characterize their functional roles in ITP progression.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003eThe authors gratefully acknowledge the technical support and facilities provided by the Department of Hematology at the First Affiliated Hospital of Chongqing Medical University and Fengdu County Traditional Chinese Medicine Hospital.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthor Contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eF.F.L.completed the bioinformatics analysis component of this study, drafted the initial manuscript, and participated in subsequent revisions. \\u003cstrong\\u003eZ\\u003c/strong\\u003e\\u003cstrong\\u003e.Y.C.\\u003c/strong\\u003eand\\u0026nbsp;\\u003cstrong\\u003eZ.H.Y.\\u003c/strong\\u003esystematically organized the research data and refined all figures and tables. \\u003cstrong\\u003eJ\\u003c/strong\\u003e\\u003cstrong\\u003e.Y.S.\\u003c/strong\\u003eand \\u003cstrong\\u003eJ\\u003c/strong\\u003e\\u003cstrong\\u003e.P.\\u003c/strong\\u003ecollected clinical specimens and performed analytical processing of the research outco mes. \\u003cstrong\\u003eP.\\u003c/strong\\u003e\\u003cstrong\\u003eH.\\u003c/strong\\u003eand \\u003cstrong\\u003eZ\\u003c/strong\\u003e\\u003cstrong\\u003e.S.Y.\\u003c/strong\\u003ewere responsible for the conceptual framework and methodological rigor of this study, additionally conducting critical reviews of the revised manuscript for academic coherence.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was supported by grants from the Chongqing Medical Scientific Research Project (Grant No. 2025MSXM065) and the Chongqing Natural Science Foundation (Grant No. CSTB2025NSCQ-GPX1218).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eData Availability Statement\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe dataset (GSE43179) supporting the conclusions of this article is available in the [GEO] repository, [https://www.ncbi.nlm.nih.gov/geo/]. The ERGs data are available in the [ExoBCD] repository, [https://exobcd.liumwei.org/].\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting Interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare no competing interests.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCorresponding Authors\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eCorrespondence to Ping Huang and Zesong Yang.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eSchramm, T. et al. Fibrinolysis is impaired in patients with primary immune thrombocytopenia. \\u003cem\\u003eJ. Thromb. Haemost\\u003c/em\\u003e. \\u003cb\\u003e22\\u003c/b\\u003e (11), 3209\\u0026ndash;3220 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eCao, J. et al. MST4 kinase regulates immune thrombocytopenia by phosphorylating STAT1-mediated M1 polarization of macrophages. \\u003cem\\u003eCell. Mol. Immunol.\\u003c/em\\u003e \\u003cb\\u003e20\\u003c/b\\u003e (12), 1413\\u0026ndash;1427 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSchifferli, A. et al. Adolescents and young adults with newly diagnosed primary immune thrombocytopenia. \\u003cem\\u003eHaematologica\\u003c/em\\u003e \\u003cb\\u003e108\\u003c/b\\u003e (10), 2783\\u0026ndash;2793 (2023). Published 2023 Oct 1.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eAuthor, R. M. et al. Erratum to: 36th International Symposium on Intensive Care and Emergency Medicine: Brussels, Belgium. 15\\u0026ndash;18 March 2016. Crit Care. ;20:347. Published 2016 Oct 24. (2016).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHu, Y. et al. Efficacy and safety of sovleplenib (HMPL\\u0026ndash;523) in adult patients with chronic primary immune thrombocytopenia in China (ESLIM\\u0026ndash;01): a randomised, double-blind, placebo-controlled, phase 3 study. \\u003cem\\u003eLancet Haematol.\\u003c/em\\u003e \\u003cb\\u003e11\\u003c/b\\u003e (8), e567\\u0026ndash;e579 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhuang, X. et al. Decreased cyclooxygenase\\u0026ndash;2 associated with impaired megakaryopoiesis and thrombopoiesis in primary immune thrombocytopenia. \\u003cem\\u003eJ. Transl Med.\\u003c/em\\u003e \\u003cb\\u003e21\\u003c/b\\u003e (1), 540 (2023). Published 2023 Aug 12.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHuang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21\\u0026thinsp;+\\u0026thinsp;Regulatory T Cell-Mediated Immune Suppression via MiR\\u0026ndash;363\\u0026ndash;3p. \\u003cem\\u003eInflamm. Published online March\\u003c/em\\u003e \\u003cb\\u003e4\\u003c/b\\u003e, (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLarue, M. et al. Efficacy and safety of dapsone in adult primary immune thrombocytopenia. \\u003cem\\u003eBlood Adv.\\u003c/em\\u003e \\u003cb\\u003e9\\u003c/b\\u003e (8), 1976\\u0026ndash;1983 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eYan, S. et al. A phase 2 trial of orelabrutinib showing promising efficacy and safety in patients with persistent or chronic primary immune thrombocytopenia. \\u003cem\\u003eAm. J. Hematol.\\u003c/em\\u003e \\u003cb\\u003e99\\u003c/b\\u003e (7), 1392\\u0026ndash;1395 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGoodman, S. R. The exosome collection. \\u003cem\\u003eCell. Mol. Biol. Lett.\\u003c/em\\u003e \\u003cb\\u003e30\\u003c/b\\u003e (1), 33 (2025). Published 2025 Mar 24.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eArya, S. B., Collie, S. P. \\u0026amp; Parent, C. A. The ins-and-outs of exosome biogenesis, secretion, and internalization. \\u003cem\\u003eTrends Cell. Biol.\\u003c/em\\u003e \\u003cb\\u003e34\\u003c/b\\u003e (2), 90\\u0026ndash;108 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eIsaac, R., Reis, F. C. G., Ying, W. \\u0026amp; Olefsky, J. M. Exosomes as mediators of intercellular crosstalk in metabolism. \\u003cem\\u003eCell. Metab.\\u003c/em\\u003e \\u003cb\\u003e33\\u003c/b\\u003e (9), 1744\\u0026ndash;1762 (2021).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eFang, Y. et al. Exosomes as biomarkers and therapeutic delivery for autoimmune diseases: Opportunities and challenges. \\u003cem\\u003eAutoimmun. Rev.\\u003c/em\\u003e \\u003cb\\u003e22\\u003c/b\\u003e (3), 103260 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eShen, Z. et al. Effects of Mesenchymal Stem Cell-Derived Exosomes on Autoimmune Diseases. \\u003cem\\u003eFront. Immunol.\\u003c/em\\u003e \\u003cb\\u003e12\\u003c/b\\u003e, 749192 (2021). Published 2021 Sep 27.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHuang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21\\u0026thinsp;+\\u0026thinsp;Regulatory T Cell-Mediated Immune Suppression via MiR\\u0026ndash;363\\u0026ndash;3p. \\u003cem\\u003eInflamm. Published online March\\u003c/em\\u003e \\u003cb\\u003e4\\u003c/b\\u003e, (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSun, Y. et al. Proteomic analysis and microRNA expression profiling of plasma-derived exosomes in primary immune thrombocytopenia. \\u003cem\\u003eBr. J. Haematol.\\u003c/em\\u003e \\u003cb\\u003e194\\u003c/b\\u003e (6), 1045\\u0026ndash;1052 (2021).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWang, J., Qing, M., Gui, J., Zhong, P. \\u0026amp; Hua, H. Identification of exosome-related genes associated with prognosis and immune infiltration features in pancreatic cancer. \\u003cem\\u003eDiscov Oncol.\\u003c/em\\u003e \\u003cb\\u003e16\\u003c/b\\u003e (1), 192 (2025). Published 2025 Feb 17.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eRitchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. \\u003cem\\u003eNucleic Acids Res.\\u003c/em\\u003e \\u003cb\\u003e43\\u003c/b\\u003e (7), e47 (2015).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eShi, Y. et al. Crosstalk of ferroptosis regulators and tumor immunity in pancreatic adenocarcinoma: novel perspective to mRNA vaccines and personalized immunotherapy. \\u003cem\\u003eApoptosis\\u003c/em\\u003e \\u003cb\\u003e28\\u003c/b\\u003e (9\\u0026ndash;10), 1423\\u0026ndash;1435 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGustavsson, E. K., Zhang, D., Reynolds, R. H., Garcia-Ruiz, S. \\u0026amp; Ryten, M. ggtranscript: an R package for the visualization and interpretation of transcript isoforms using ggplot2. \\u003cem\\u003eBioinformatics\\u003c/em\\u003e \\u003cb\\u003e38\\u003c/b\\u003e (15), 3844\\u0026ndash;3846 (2022).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHorvath, S. \\u0026amp; Dong, J. Geometric interpretation of gene coexpression network analysis. \\u003cem\\u003ePLoS Comput. Biol.\\u003c/em\\u003e \\u003cb\\u003e4\\u003c/b\\u003e (8), e1000117 (2008). Published 2008 Aug 15.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. \\u003cem\\u003eInnov. (Camb)\\u003c/em\\u003e. \\u003cb\\u003e2\\u003c/b\\u003e (3), 100141 (2021). Published 2021 Jul 1.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSmoot, M. E., Ono, K., Ruscheinski, J., Wang, P. L. \\u0026amp; Ideker, T. Cytoscape 2.8: new features for data integration and network visualization. \\u003cem\\u003eBioinformatics\\u003c/em\\u003e \\u003cb\\u003e27\\u003c/b\\u003e (3), 431\\u0026ndash;432 (2011).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eMahmoudian, M., Ven\\u0026auml;l\\u0026auml;inen, M. S., Kl\\u0026eacute;n, R. \\u0026amp; Elo, L. L. \\u003cem\\u003eStable Iterative Variable Selection Bioinf.\\u003c/em\\u003e ;\\u003cb\\u003e37\\u003c/b\\u003e(24):4810\\u0026ndash;4817. (2021).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eFriedman, J., Hastie, T. \\u0026amp; Tibshirani, R. Regularization Paths for Generalized Linear Models via Coordinate Descent. \\u003cem\\u003eJ. Stat. Softw.\\u003c/em\\u003e \\u003cb\\u003e33\\u003c/b\\u003e (1), 1\\u0026ndash;22 (2010).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLi, M. et al. Recognition of refractory Mycoplasma pneumoniae pneumonia among Myocoplasma pneumoniae pneumonia in hospitalized children: development and validation of a predictive nomogram model. \\u003cem\\u003eBMC Pulm Med.\\u003c/em\\u003e \\u003cb\\u003e23\\u003c/b\\u003e (1), 383 (2023). Published 2023 Oct 10.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eRobin, X. et al. pROC: an open-source package for R and S\\u0026thinsp;+\\u0026thinsp;to analyze and compare ROC curves. \\u003cem\\u003eBMC Bioinform.\\u003c/em\\u003e \\u003cb\\u003e12\\u003c/b\\u003e, 77 (2011). Published 2011 Mar 17.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eNazari-Ghadikolaei, A., Fikse, F., Gelinder Viklund, \\u0026Aring;. \\u0026amp; Eriksson, S. Factor analysis of evaluated and linearly scored traits in Swedish Warmblood horses. \\u003cem\\u003eJ. Anim. Breed. Genet.\\u003c/em\\u003e \\u003cb\\u003e140\\u003c/b\\u003e (4), 366\\u0026ndash;375 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWu, T. et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. \\u003cem\\u003eInnov. (Camb)\\u003c/em\\u003e. \\u003cb\\u003e2\\u003c/b\\u003e (3), 100141 (2021). Published 2021 Jul 1.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eQi, W. Y. et al. Immune cells in metabolic associated fatty liver disease: Global trends and hotspots (2004\\u0026ndash;2024). \\u003cem\\u003eWorld J. Hepatol.\\u003c/em\\u003e \\u003cb\\u003e17\\u003c/b\\u003e (3), 103327 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLi, Y., Jiang, H., Luo, L. \\u0026amp; Mei, H. Immunotherapy-associated hemostatic abnormalities: bleeding and thrombotic complications. \\u003cem\\u003eAnn. Hematol.\\u003c/em\\u003e \\u003cb\\u003e104\\u003c/b\\u003e (7), 3537\\u0026ndash;3551 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHillier, K. et al. Immunoglobulins act as predictors of chronicity in pediatric immune thrombocytopenia. \\u003cem\\u003eBlood Adv.\\u003c/em\\u003e \\u003cb\\u003e8\\u003c/b\\u003e (19), 5118\\u0026ndash;5121 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eAran, D., Hu, Z. \\u0026amp; Butte, A. J. xCell: digitally portraying the tissue cellular heterogeneity landscape. \\u003cem\\u003eGenome Biol.\\u003c/em\\u003e \\u003cb\\u003e18\\u003c/b\\u003e (1), 220 (2017). Published 2017 Nov 15.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGonz\\u0026aacute;lez-L\\u0026oacute;pez, T. J. \\u0026amp; Provan, D. The new era of primary immune thrombocytopenia management in adults: A narrative review of current and emerging treatments. \\u003cem\\u003eBlood Rev.\\u003c/em\\u003e \\u003cb\\u003e73\\u003c/b\\u003e, 101300 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLi, G. et al. The therapeutic potential of exosomes in immunotherapy. \\u003cem\\u003eFront. Immunol.\\u003c/em\\u003e \\u003cb\\u003e15\\u003c/b\\u003e, 1424081 (2024). Published 2024 Jul 8.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGuo, H. et al. Dysfunction of astrocytic glycophagy exacerbates reperfusion injury in ischemic stroke. \\u003cem\\u003eRedox Biol.\\u003c/em\\u003e \\u003cb\\u003e74\\u003c/b\\u003e, 103234 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLi, L. et al. Beclin 1 of megakaryocytic lineage cells is locally dispensable for platelet hemostasis but functions distally in bone homeostasis. \\u003cem\\u003eBone Res.\\u003c/em\\u003e \\u003cb\\u003e13\\u003c/b\\u003e (1), 32 (2025). Published 2025 Mar 3.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePanahi Meymandi, A. R. et al. Crosstalk between autophagy and metabolic regulation of (CAR) T cells: therapeutic implications. \\u003cem\\u003eFront. Immunol.\\u003c/em\\u003e \\u003cb\\u003e14\\u003c/b\\u003e, 1212695 (2023). Published 2023 Aug 22.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eYan, J. et al. Platelet Pharmacytes for the Hierarchical Amplification of Antitumor Immunity in Response to Self-Generated Immune Signals. \\u003cem\\u003eAdv. Mater.\\u003c/em\\u003e \\u003cb\\u003e34\\u003c/b\\u003e (23), e2109517 (2022).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eMo, J. et al. Comprehensive analysis and prediction model of mitophagy and ferroptosis in primary immune thrombocytopenia. \\u003cem\\u003eBr. J. Haematol.\\u003c/em\\u003e \\u003cb\\u003e204\\u003c/b\\u003e (6), 2429\\u0026ndash;2441 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eYang, Y. et al. SOX4-ZIP14-zinc metabolism mediates oncogenesis and suppresses T cell immunity in nasopharyngeal carcinoma. \\u003cem\\u003eCell. Rep. Med.\\u003c/em\\u003e \\u003cb\\u003e6\\u003c/b\\u003e (9), 102300 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eShao, Y. et al. SLC39A10 is a key zinc transporter in T cells and its loss mitigates autoimmune disease. \\u003cem\\u003eSci. China Life Sci.\\u003c/em\\u003e \\u003cb\\u003e68\\u003c/b\\u003e (7), 1855\\u0026ndash;1870 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eDeng, Z. et al. Mesenchymal Stem Cells Prevent SLC39A14-Dependent Hepatocyte Ferroptosis through Exosomal miR\\u0026ndash;16\\u0026ndash;5p in Liver Graft. \\u003cem\\u003eAdv. Sci. (Weinh)\\u003c/em\\u003e. \\u003cb\\u003e12\\u003c/b\\u003e (6), e2411380 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSauer, A. K. et al. Zinc is a key regulator of gastrointestinal development, microbiota composition and inflammation with relevance for autism spectrum disorders. \\u003cem\\u003eCell. Mol. Life Sci.\\u003c/em\\u003e \\u003cb\\u003e79\\u003c/b\\u003e (1), 46 (2021). Published 2021 Dec 22.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eMa, L. et al. lncRNA, miRNA, and mRNA of plasma and tumor-derived exosomes of cardiac myxoma-related ischaemic stroke. Sci Data. ;12(1):91. Published 2025 Jan 16. (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eTukacs, V. et al. Chronic Cerebral Hypoperfusion-Induced Disturbed Proteostasis of Mitochondria and MAM Is Reflected in the CSF of Rats by Proteomic Analysis. \\u003cem\\u003eMol. Neurobiol.\\u003c/em\\u003e \\u003cb\\u003e60\\u003c/b\\u003e (6), 3158\\u0026ndash;3174 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eChen, W. et al. Mitochondrial enzyme HIBADH protects against calcium oxalate nephrolithiasis by modulating oxidative stress and apoptosis. \\u003cem\\u003eArch. Biochem. Biophys.\\u003c/em\\u003e \\u003cb\\u003e770\\u003c/b\\u003e, 110452 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eBishop, C. A. et al. Detrimental effects of branched-chain amino acids in glucose tolerance can be attributed to valine induced glucotoxicity in skeletal muscle. \\u003cem\\u003eNutr. Diabetes\\u003c/em\\u003e. \\u003cb\\u003e12\\u003c/b\\u003e (1), 20 (2022). Published 2022 Apr 13.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eHuang, Y. et al. Plasma Exosomes Derived from Patients with Primary Immune Thrombocytopenia Attenuate TBX21\\u0026thinsp;+\\u0026thinsp;Regulatory T Cell-Mediated Immune Suppression via MiR\\u0026ndash;363\\u0026ndash;3p. \\u003cem\\u003eInflamm. Published online March\\u003c/em\\u003e \\u003cb\\u003e4\\u003c/b\\u003e, (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhang, R. et al. Pleiotropic effects of a mitochondrion-targeted glutathione reductase inhibitor on restraining tumor cells. \\u003cem\\u003eEur. J. Med. Chem.\\u003c/em\\u003e \\u003cb\\u003e248\\u003c/b\\u003e, 115069 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhan, Y. et al. Impaired mitochondria of Tregs decreases OXPHOS-derived ATP in primary immune thrombocytopenia with positive plasma pathogens detected by metagenomic sequencing. \\u003cem\\u003eExp. Hematol. Oncol.\\u003c/em\\u003e \\u003cb\\u003e11\\u003c/b\\u003e (1), 48 (2022). Published 2022 Sep 1.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eXie, Y. et al. PM2.5 promotes platelet activation and thrombosis via ROS/MAPKs pathway-mediated mitochondrial dysfunction. \\u003cem\\u003eEnviron. Res.\\u003c/em\\u003e \\u003cb\\u003e283\\u003c/b\\u003e, 122116 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWang, C., Xu, M., Fan, Q., Li, C. \\u0026amp; Zhou, X. Therapeutic potential of exosome-based personalized delivery platform in chronic inflammatory diseases. \\u003cem\\u003eAsian J. Pharm. Sci.\\u003c/em\\u003e \\u003cb\\u003e18\\u003c/b\\u003e (1), 100772 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWang, J. et al. Second-generation antipsychotics induce cardiotoxicity by disrupting spliceosome signaling: Implications from proteomic and transcriptomic analyses. \\u003cem\\u003ePharmacol. Res.\\u003c/em\\u003e \\u003cb\\u003e170\\u003c/b\\u003e, 105714 (2021).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSun, T. et al. Proteomics landscape and machine learning prediction of long-term response to splenectomy in primary immune thrombocytopenia. \\u003cem\\u003eBr. J. Haematol.\\u003c/em\\u003e \\u003cb\\u003e204\\u003c/b\\u003e (6), 2418\\u0026ndash;2428 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eKarginov, T. A., M\\u0026eacute;noret, A. \\u0026amp; Vella, A. T. Optimal CD8\\u0026thinsp;+\\u0026thinsp;T cell effector function requires costimulation-induced RNA-binding proteins that reprogram the transcript isoform landscape. \\u003cem\\u003eNat. Commun.\\u003c/em\\u003e \\u003cb\\u003e13\\u003c/b\\u003e (1), 3540 (2022). Published 2022 Jun 20.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePellagatti, A. et al. Impact of spliceosome mutations on RNA splicing in myelodysplasia: dysregulated genes/pathways and clinical associations. \\u003cem\\u003eBlood\\u003c/em\\u003e \\u003cb\\u003e132\\u003c/b\\u003e (12), 1225\\u0026ndash;1240 (2018).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eMildner, A. \\u0026amp; Yona, S. Monocytes and their doppelg\\u0026auml;ngers: An immunological crossroads. \\u003cem\\u003eSci. Immunol.\\u003c/em\\u003e \\u003cb\\u003e9\\u003c/b\\u003e (101), eadr6672 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhao, Y. et al. Tumor Necrosis Factor-α Blockade Corrects Monocyte/Macrophage Imbalance in Primary Immune Thrombocytopenia. \\u003cem\\u003eThromb. Haemost\\u003c/em\\u003e. \\u003cb\\u003e121\\u003c/b\\u003e (6), 767\\u0026ndash;781 (2021).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eWang, W. et al. Quercetin, a key active ingredient of Jianpi Zishen Xiehuo Formula, suppresses M1 macrophage polarization and platelet phagocytosis by inhibiting STAT3 activation based on network pharmacology. \\u003cem\\u003eNaunyn Schmiedebergs Arch. Pharmacol.\\u003c/em\\u003e \\u003cb\\u003e397\\u003c/b\\u003e (6), 4219\\u0026ndash;4233 (2024).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eEl-Rashedi, F. H., El-Hawy, M. A., Helwa, M. A. \\u0026amp; Abd-Allah, S. S. Study of CD4+, CD8+, and natural killer cells (CD16+, CD56+) in children with immune thrombocytopenic purpura. \\u003cem\\u003eHematol. Oncol. Stem Cell. Ther.\\u003c/em\\u003e \\u003cb\\u003e10\\u003c/b\\u003e (1), 8\\u0026ndash;14 (2017).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhao, Z. et al. Targeted delivery of exosomal miR\\u0026ndash;484 reprograms tumor vasculature for chemotherapy sensitization. \\u003cem\\u003eCancer Lett.\\u003c/em\\u003e \\u003cb\\u003e530\\u003c/b\\u003e, 45\\u0026ndash;58 (2022).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLi, H. et al. Mir\\u0026ndash;484 contributes to diminished ovarian reserve by regulating granulosa cell function via YAP1-mediated mitochondrial function and apoptosis. \\u003cem\\u003eInt. J. Biol. Sci.\\u003c/em\\u003e \\u003cb\\u003e18\\u003c/b\\u003e (3), 1008\\u0026ndash;1021 (2022). Published 2022 Jan 1.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLin, T. et al. Identification of exosomal miR\\u0026ndash;484 role in reprogramming mitochondrial metabolism in pancreatic cancer through Wnt/MAPK axis control. \\u003cem\\u003ePharmacol. Res.\\u003c/em\\u003e \\u003cb\\u003e197\\u003c/b\\u003e, 106980 (2023).\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGuo, Y. et al. Rbms1 promotes pulmonary fibrosis by stabilising Sumo2 mRNA to facilitate Smad4-SUMOylation and fibroblast activation. \\u003cem\\u003eEur. Respir J.\\u003c/em\\u003e \\u003cb\\u003e66\\u003c/b\\u003e (3), 2401667 (2025). Published 2025 Sep 25.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eDe Luca, B. et al. Efficacy and tolerability of antidepressants in individuals suffering from physical conditions and depressive disorders: network meta-analysis. \\u003cem\\u003eBr. J. Psychiatry\\u003c/em\\u003e. \\u003cb\\u003e227\\u003c/b\\u003e (2), 553\\u0026ndash;566 (2025).\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"},{\"header\":\"Tables\",\"content\":\"\\u003cp\\u003eTable 1 is available in the Supplementary Files section.\\u003c/p\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"scientific-reports\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"scirep\",\"sideBox\":\"Learn more about [Scientific Reports](http://www.nature.com/srep/)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Scientific Reports\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Scientific Reports\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Primary immune thrombocytopenia, Exosome, Transcriptomics, Biomarkers, Immunity\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-8092905/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-8092905/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003ch2\\u003eBackground\\u003c/h2\\u003e\\u003cp\\u003ePrimary Immune Thrombocytopenia (ITP) is an autoimmune disease with thrombocytopenia and bleeding tendency. Exosomes mediate abnormal crosstalk between immune cells and megakaryocytes in ITP, making exosome-related biomarkers crucial for the disease\\u0026rsquo;s diagnosis and treatment.\\u003c/p\\u003e\\u003ch2\\u003eMethods\\u003c/h2\\u003e\\u003cp\\u003eITP transcriptome data and exosome-related genes (ERGs) were retrieved from public databases. Candidate genes were identified by intersecting ITP\\u0026rsquo;s DEGs with exosome-related key module genes, followed by biomarker screening via machine learning and nomogram construction. Multi-dimensional analyses (enrichment, immune infiltration, drug prediction) and RT-qPCR validation were performed.\\u003c/p\\u003e\\u003ch2\\u003eResults\\u003c/h2\\u003e\\u003cp\\u003eFour biomarkers (GABARAPL1, SLC39A14, HIBADH, GSR) were confirmed, involved in spliceosome and other pathways (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05, |NES| \\u0026gt;1). GABARAPL1, SLC39A14, HIBADH negatively correlated with activated NK cells (|cor| \\u0026gt;0.3, P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05). SLC39A14/nortriptyline and GSR/oxiglutatione showed strong binding affinity (binding free energy \\u0026lt; -5 kcal/mol). RT-qPCR verified dysregulation of GABARAPL1, SLC39A14, GSR in ITP patients (P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05).\\u003c/p\\u003e\\u003ch2\\u003eConclusion\\u003c/h2\\u003e\\u003cp\\u003eIn this study, GABARAPL1, SLC39A14, HIBADH, and GSR were successfully screened as biomarkers, and their related regulatory mechanisms were also revealed, providing innovative theoretical support for the precise diagnosis and treatment of ITP.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Exploring biomarkers related to exosome in primary immune thrombocytopenia based on transcriptomics and experimental verification\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-12-09 00:23:24\",\"doi\":\"10.21203/rs.3.rs-8092905/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2026-01-05T17:38:06+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2025-12-31T13:13:08+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2025-12-15T11:50:13+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"50374959129282403799362647607934990599\",\"date\":\"2025-12-15T09:05:43+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"273033445826160519248339824680608798583\",\"date\":\"2025-12-07T05:19:59+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2025-12-04T23:01:33+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2025-11-25T02:48:07+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvited\",\"content\":\"\",\"date\":\"2025-11-24T12:56:44+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2025-11-20T14:38:02+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Scientific Reports\",\"date\":\"2025-11-20T14:32:24+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"scientific-reports\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"scirep\",\"sideBox\":\"Learn more about [Scientific Reports](http://www.nature.com/srep/)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Scientific Reports\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Scientific Reports\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"2a4b822c-6c3b-49dc-a4d1-f951cdf2748a\",\"owner\":[],\"postedDate\":\"December 9th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[{\"id\":59233412,\"name\":\"Health sciences/Biomarkers\"},{\"id\":59233413,\"name\":\"Biological sciences/Computational biology and bioinformatics\"},{\"id\":59233414,\"name\":\"Health sciences/Diseases\"},{\"id\":59233415,\"name\":\"Biological sciences/Immunology\"}],\"tags\":[],\"updatedAt\":\"2026-03-23T16:01:21+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-8092905\",\"link\":\"https://doi.org/10.1038/s41598-026-43618-1\",\"journal\":{\"identity\":\"scientific-reports\",\"isVorOnly\":false,\"title\":\"Scientific Reports\"},\"publishedOn\":\"2026-03-20 15:57:50\",\"publishedOnDateReadable\":\"March 20th, 2026\"},\"versionCreatedAt\":\"2025-12-09 00:23:24\",\"video\":\"\",\"vorDoi\":\"10.1038/s41598-026-43618-1\",\"vorDoiUrl\":\"https://doi.org/10.1038/s41598-026-43618-1\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-8092905\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-8092905\",\"identity\":\"rs-8092905\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}