Statistical Methods for RNA Sequencing Data Analysis

Anchored-fusion enables targeted fusion search in bulk and single-cell RNA sequencing data

BackgroundIt was previously reported that the production of exerkines is positively associated with the beneficial effects of exercise in lung adenocarcinoma (LUAD) patients. This study proposes a novel scoring system based on muscle failure-related genes, to assist in clinical decision making.MethodsA comprehensive analysis of bulk and single cell RNA sequencing (scRNA-seq) of early, advanced and brain metastatic LUAD tissues and normal lung tissues was performed to identify muscle failure-related genes in LUAD and to determine the distribution of muscle failure-related genes in different cell populations. A novel scoring system, named MFI (Muscle failure index), was developed and validated. The differences in biological functions, immune infiltration, genomic alterations, and clinical significance of different subtypes were also investigated.ResultsFirst, we conducted single cell analysis on the dataset GSE131907 and identified eight cell subpopulations. We found that four muscle failure-related genes (BDNF, FNDC5, IL15, MSTN) were significantly increased in tumor cells. In addition, IL15 was widely distributed in the immune cell population. And we have validated it in our own clinical cohort. Then we created the MFI model based on 10 muscle failure-related genes using the LASSO algorithm, and MFI remained an independent prognostic factor of OS in both the training and validation cohorts. Moreover, we generated MFI in the single-cell dataset, in which cells with high MFI received and sent more signals compared to those with low MFI. Biological function analysis of both subtypes revealed stronger anti-tumor immune activity in the low MFI group, while tumor cells with high MFI had stronger metabolic and proliferative activity. Finally, we systematically assessed the immune cell activity and immunotherapy responses in LUAD patients, finding that the low MFI group was more sensitive to immunotherapy.ConclusionOverall, our study can improve the understanding of the role of muscle failure-related genes in tumorigenesis and we constructed a reliable MFI model for predicting prognosis and guiding future clinical decision making.

Chronic kidney disease (CKD) emerges as a substantial contributor to various cardiovascular disorders, including ischemic cardiomyopathy (ICM). However, the underlying molecular mechanisms linking CKD and ICM remain elusive. Our study aims to unravel these connections by integrating publicly available bulk and single-cell RNA sequencing (scRNA-seq) data. Expression profiles from two ICM datasets obtained from heart tissue and one CKD with Peripheral Blood Mononuclear Cell (CKD-PBMC) dataset were collected. We initiated by identifying shared differentially expressed genes (DEGs) between ICM and CKD. Subsequent functional enrichment analysis shed light on the mechanisms connecting CKD to ICM. Machine learning algorithms enabled the identification of 13 candidate genes, including AGRN, COL16A1, COL1A2, FAP, FRZB, GPX3, ITIH5, NFASC, PTN, SLC38A1, STARD7, THBS2, and VPS35. Their expression patterns in ICM were investigated via scRNA-seq data analysis. Notably, most of them were enriched in fibroblasts. COL16A1, COL1A2, PTN, and FAP were enriched in scar-formation fibroblasts, while GPX3 and THBS2 showed enrichment in angiogenesis fibroblasts. A Gaussian naïve Bayes model was developed for diagnosing CKD-related ICM, bolstered by SHapley Additive exPlanations interpretability and validated internally and externally. In conclusion, our investigation unveils the extracellular matrix's role in CKD and ICM interplay, identifies 13 candidate genes, and showcases their expression patterns in ICM. We also constructed a diagnostic model using 13 gene features and presented an innovative approach for managing CKD-related ICM through serum-based diagnostic strategies.

Hypertrophic scar (HS) is a common skin disorder characterized by excessive extracellular matrix (ECM) deposition. However, it is still unclear how the cellular composition, cell-cell communications, and crucial transcriptionally regulatory network were changed in HS. In the present study, we found that FB-1, which was identified a major type of fibroblast and had the characteristics of myofibroblast, was significantly expanded in HS by integrative analysis of the single-cell and bulk RNA sequencing (RNA-seq) data. Moreover, the proportion of KC-2, which might be a differentiated type of keratinocyte (KC), was reduced in HS. To decipher the intercellular signaling, we conducted the cell-cell communication analysis between the cell types, and found the autocrine signaling of HB-1 through COL1A1/2-CD44 and CD99-CD99 and the intercellular contacts between FB-1/FB-5 and KC-2 through COL1A1/COL1A2/COL6A1/COL6A2-SDC4. Almost all the ligands and receptors involved in the autocrine signaling of HB-1 were upregulated in HS by both scRNA-seq and bulk RNA-seq data. In contrast, the receptor of KC-2, SDC4, which could bind to multiple ligands, was downregulated in HS, suggesting that the reduced proportion of KC-2 and apoptotic phenotype of KC-2 might be associated with the downregulation of SDC4. Furthermore, we also investigated the transcriptionally regulatory network involved in HS formation. The integrative analysis of the scRNA-seq and bulk RNA-seq data identified CREB3L1 and TWIST2 as the critical TFs involved in the myofibroblast of HS. In summary, the integrative analysis of the single-cell RNA sequencing (scRNA-seq) and bulk RNA-seq data greatly improved our understanding of the biological characteristics during the HS formation.

BackgroundAs a predominant form of cancer affecting male populations, prostate cancer (PCa) demonstrates notably high incidence rates globally. The significant heterogeneity in tumor microenvironment (TME) composition (including epithelial and diverse cell populations) hinders clear interpretation of gene and biomarker roles in disease advancement and immune response modulation. Through combined analysis of bulk and single-cell RNA sequencing data, this investigation evaluates prostate cancer-related genes’ clinical relevance and prognostic potential.MethodsWe applied approaches like FindAllMarkers, Dseq2 R package, ssGSEA, and WGCNA with both single-cell and bulk transcriptome scales to uncover genes associated with prognosis. Furthermore, we developed a machine learning approach integrating 14 algorithms and 162 algorithmic combinations to support the formation of consensus immune and prognostic-related signatures (IPRS). The IPRS underwent systematic validation in both training and test cohorts, and a multivariate nomogram was constructed to demonstrate its potential utility in prognosis quantification. Through comprehensive multi-omics analyses, which included genomic, single-cell transcriptomic, and bulk transcriptomic data, we sought to achieve a thorough understanding of prognostic characteristics. Furthermore, we evaluated the clinical applicability of the IPRS in the context of immunotherapy and personalized drug selection. Additionally, we examined and confirmed both the gene expression associated with IPRS and the expression level and function of B-cell adhesion molecule (BCAM) within prostate cancer cells and tissue.ResultWe discovered 91 genes associated with prognosis in the TME, with 15 of these genes connected to biochemical recurrence. The consensus IPRS constructed based on a machine learning computational framework demonstrates potential value in prognosis prediction and clinical relevance. Multivariate analysis further supports the possibility of IPRS serving as an independent prognostic marker for prostate cancer disease progression. Significant differences in biological functions, immune infiltration, and genomic mutations were also observed among different risk groups. Significantly, the submap method revealed enhanced immunotherapy responsiveness in high-risk patients while highlighting potential pharmacological targets for certain risk subgroups.ConclusionWe selected a collection of genes relevant to PCa prognosis and immune characteristics, which may serve as potential biomarkers with certain clinical translational value.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40001-025-03363-z.

The tumor microenvironment (TME) is crucial in cancer development and therapeutic response. Immunotherapy is increasingly recognized as a critical component of cancer treatment. While immunotherapies have shown efficacy in various cancers, including breast cancer, patient responses vary widely. Some patients receive significant benefits, while others experience minimal or no improvement. This disparity underscores the complexity and diversity of the immune system. In this study, we investigated the immune landscape and cell-cell communication within the TME of breast cancer through integrated analysis of bulk and single-cell RNA sequencing data. We established profiles of tumor immune infiltration that span across a broad spectrum of adaptive and innate immune cells. Our clustering analysis of immune infiltration identified three distinct patient groups: high T cell abundance, moderate infiltration, and low infiltration. Patients with low immune infiltration exhibited the poorest survival rates, while those in the moderate infiltration group showed better outcomes than those with high T cell abundance. Moreover, the high cell abundance group was associated with a greater tumor burden and higher rates of TP53 mutations, whereas the moderate infiltration group was characterized by a lower tumor burden and elevated PIK3CA mutations. Analysis of an independent single-cell RNA-seq breast cancer dataset confirmed the presence of similar infiltration patterns. Further investigation into ligand-receptor interactions within the TME unveiled significant variations in cell-cell communication patterns among these groups. Notably, we found that the signaling pathways SPP1 and EGF were exclusively active in the low immune infiltration group, suggesting their involvement in immune suppression. This work comprehensively characterizes the composition and dynamic interplay in the breast cancer TME. Our findings reveal associations between the extent of immune infiltration and clinical outcomes, providing valuable prognostic information for patient stratification. The unique mutations and signaling pathways associated with different patient groups offer insights into the mechanisms underlying diverse tumor immune infiltration and the formation of an immunosuppressive tumor microenvironment.

Introduction: Analyses of bulk RNA sequencing (RNA-seq) data are central to most large-scale tumour sequencing studies. Solid tumors often present a dynamic, heterogeneous environment consisting of both cancer (sub)clones and normal cells. Bulk expression data represent population averages and its interpretation is confounded by both normal cell contamination and somatic copy number alterations. Several computational methods to deconvolve tumour and normal expression profiles from bulk RNA-seq have been developed recently. However, these methods often rely on a set of cell-type-specific reference signatures and ignore the effect of copy number changes. Methods: To address these issues, we have developed a method that formalizes the relationship between allele-specific copy number, expression and sample purity to deconvolve the expression profiles of tumor and normal cells from bulk RNA-seq data in an unbiased manner. Our method was applied to sequencing data produced by the TRACERx consortium, a longitudinal study with multi-region whole-exome and RNA-seq of non-small-cell lung cancers. A total of 414 primary tumor regions and 140 adjacent normal tissue samples from 140 TRACERx patients with matched DNA and RNA sequencing data were processed. Results: Here, we were able to directly deconvolve a median of ~2,000 genes per sample and indirectly infer tumor and normal expression profiles of ~10,000 genes. The accuracy of the deconvolution was validated using in-silico mixtures of patient-derived tumour and normal cells and in regions with loss of heterogeneity (LOH) directly on the bulk sequencing data, where the total fraction of expression attributed to tumor cells can be computed directly using somatic mutations. Our method revealed a strong and constitutive genome-wide overexpression in cancer cells compared to admixed normal cells, this overexpression was more pronounced in lung squamous cell carcinoma than lung adenocarcinoma (p&lt0.001). Multidimensional projection of the purified tumor and normal expression profiles together with the adjacent normal tissue showed clear separation between the purified expression profiles and notably, the deconvolved normal expression was more similar to the normal adjacent profiles than the purified tumor profiles. Conclusion: Overall, these results suggest that our method is able to accurately disentangle the expression of tumour and normal cells from bulk RNA-seq without any previous knowledge. It has potential applications in many studies that include matched RNA-seq and copy number data and can provide new insights functional characterization, the taxonomy of cancer, and tumor evolution. Citation Format: Carla Castignani, Jonas Demeulemeester, Elizabeth Larose Cadieux, Robert E. Hynds, David R. Pearce, Stefan C. Dentro, Peter Van Loo, Charles Swanton, TRACERx Consortium. Allele-specific copy-number based deconvolution of bulk tumour RNA sequencing data from the TRACERx study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1211.

Back pain caused by intervertebral disc (IVD) degeneration has a major socio-economic impact in humans, yet historically has received minimal attention in species other than humans, mice and dogs. However, a general growing interest in this unique organ prompted the expansion of IVD research in rats, rabbits, cats, horses, monkeys, and cows, further illuminating the complex nature of the organ in both healthy and degenerative states. Application of recent biotechnological advancements, including single cell RNA sequencing and complex data analysis methods has begun to explain the shifting inflammatory signaling, variation in cellular subpopulations, differential gene expression, mechanical loading, and metabolic stresses which contribute to age and stress related degeneration of the IVD. This increase in IVD research across species introduces a need for chronicling IVD advancements and tissue biomarkers both within and between species. Here we provide a comprehensive review of recent single cell RNA sequencing data alongside existing case reports and histo/morphological data to highlight the cellular complexity and metabolic challenges of this unique organ that is of structural importance for all vertebrates.

Synovitis is one of the key pathological feature driving osteoarthritis (OA) development. Diverse programmed cell death (PCD) pathways are closely linked to the pathogenesis of OA, but few studies have explored the relationship between PCD-related genes and synovitis. The transcriptome expression profiles of OA synovial samples were obtained from the Gene Expression Omnibus (GEO) database. Using machine learning algorithms, Hub PCD-related differentially expressed genes (Hub PCD-DEGs) were identified. The expression of Hub PCD-DEGs was validated in human OA samples by qRT-PCR. A diagnostic model for OA was constructed based on the expression levels of Hub PCD-DEGs. Unsupervised consensus clustering analysis and weighted correlation network analysis (WGCNA) were employed to identify differential clustering patterns of PCD-related genes in OA patients. The molecular characteristics of Hub PCD-DEGs, their role in synovial immune inflammation, and their association with the immune microenvironment were investigated through functional enrichment analysis and ssGSEA immune infiltration analysis. Single-cell RNA sequencing analysis provided insights into the characteristics of distinct cell clusters in OA synovial tissues and their interactions with Hub PCD-DEGs. We identified five Hub PCD-DEGs: TNFAIP3, JUN, PPP1R15A, INHBB, and DDIT4. qRT-PCR analysis confirmed that all five genes were significantly downregulated in OA synovial tissue. The diagnostic model constructed based on these Hub PCD-DEGs demonstrated diagnostic efficiency in distinguishing OA tissues as well as progression of OA. Additionally, a correlation was observed between the expression levels of Hub PCD-DEGs, immune cell infiltration, and inflammatory cytokine levels. We identified two distinct PCD clusters, each exhibiting unique molecular and immunological characteristics. Single-cell RNA sequencing further revealed dynamic and complex cellular changes in OA synovial tissue, with differential expression of Hub PCD-DEGs across various immune cell types. Our study suggests that PCD-related genes may be involved in development of OA synovitis. The five screened Hub PCD-DEGs (TNFAIP3, JUN, PPP1R15A, INHBB and DDIT4) could be explored as candidate biomarkers or therapeutic targets for OA.

Background Cancer stem cells (CSCs) play a crucial role in the progression of lung adenocarcinoma (LUAD).This study aimed to explore the gene signatures of tumor stem cells in LUAD through Single-cell RNA sequencing (scRNA-seq) data and bulk RNA sequencing (RNA-seq) data, and establish a tumor stem cell marker signature(TSCMS)prognostic risk model. Methods The LUAD scRNA-seq data and bulk RNA-seq data from the GEO and TCGA databases were collected. CytoTRACE software was used to quantify the stemness score of tumor-derived epithelial cell clusters. Gene Set Variation Analysis (GSVA) was performed to identify potential biological functions in different clusters. The TSCMS prognostic risk model was constructed using Lasso-Cox regression analysis, and its prognostic value was assessed through Kaplan-Meier, Cox regression, and receiver-operating characteristic (ROC) curve analyses. The Cibersortx algorithm was used to evaluate immune infiltration, and drug response prediction was conducted using the pRRophetic package. Functional investigations of TAF10 in LUAD cells were performed using bioinformatics analysis, qRT-PCR, Western blot, Immunohistochemistry, cell proliferation and clone formation assay. Results Seven distinct cell clusters were identified by CytoTRACE (Epi C1 to C7), with Epi C1 demonstrating the highest stemness potential. The TSCMS prognostic risk model incorporated 49 tumor stemness-related genes, and high-risk patients exhibited reduced immune scores, lower ESTIMATE scores, and increased tumor purity. Furthermore, significant differences in immune landscapes and chemotherapy sensitivity were observed between high and low risk groups. TAF10 was found to be positively correlated with the RNA expression-based stemness score (RNAss) in various tumors, including LUAD. And we demonstrated that TAF10 was over-expressed in LUAD cell lines and tumor tissues of clinical patients, and high TAF10 expression was correlated with poor prognosis in LUAD patients. Silencing TAF10 inhibited LUAD cell proliferation and clone formation. Conclusions Our investigation highlights the prognostic utility of the TSCMS model for evaluating the clinical outcomes of LUAD patients, uncovering critical insights into immune cell infiltration and therapeutic response, and positions TAF10 as a novel therapeutic target for LUAD.

Background: Glioblastoma is a highly malignant brain tumor with limited treatment options and a poor prognosis, largely driven by its complex immune microenvironment. This study aimed to identify and characterize an immune-related gene signature associated with prognosis and immune regulation in glioblastoma. Methods: We performed integrative analyses using bulk and single-cell RNA sequencing data to identify prognostically significant immune-related genes. A five-gene signature (THEMIS2, SIGLEC9, CSTA, LILRB3, and MS4A6A) was derived and its expression patterns were analyzed in association with immune cell infiltration and macrophage subtypes. Functional enrichment and pathway analyses were conducted, followed by drug sensitivity profiling to explore potential therapeutic implications. Results: The five-gene signature was significantly associated with worse survival outcomes and increased immune cell infiltration. Functional analyses revealed involvement in key immune pathways, including antigen presentation, cytokine signaling, and immune cell activation. Single-cell RNA sequencing demonstrated high expression of the signature in tumor-associated macrophages, particularly immune-suppressive and proliferation-associated subtypes. The high expression in proliferation TAMs suggests a role in promoting tumor angiogenesis and growth. Drug sensitivity analysis revealed distinct vulnerabilities between high- and low-risk groups based on signature expression. Conclusions: This Macrophage-Associated Prognostic Signature (MAPS) provides new insights into glioblastoma immunobiology and identifies potential biomarkers and therapeutic targets. It may serve as a valuable tool to guide personalized immunotherapy-based strategies for glioblastoma patients.

Immunotherapy represents a pivotal therapeutic modality in endometrial cancer (EC). Nevertheless, the efficacy of this treatment is limited to a subset of patients. The present investigation endeavors to amalgamate multi-omics data in order to elucidate the determinants impacting individual immune responsiveness and enhance the optimization of immunotherapy for EC. To differentiate EC patients into non-response (NR) and response (R), multi-omics data from publicly available databases were employed in conjunction with the TIDE computational framework. The validity of these findings was further confirmed through the utilization of the EaSIeR and ImmunoPhenoScore algorithms. The study employed functional enrichment and gene set variant analysis to discern noteworthy disparities in biological pathways across various groups. Moreover, three deconvolution algorithms (ESTIMATE, TIMER, and EPIC) were employed to quantify the tumor microenvironment (TME). Somatic mutation and copy number variant (CNV) analyses were also conducted to identify genomic alterations that impact immunotherapy. Integrated bulk and single-cell RNA sequencing (scRNA-seq) data were employed to identify cell populations linked to efficacy and deduce cell-cell interactions. The immunotherapy response rates were found to be greater in elderly EC patients aged 65 years and above. The NR group of patients displayed notable enrichment in cellular differentiation, angiogenesis, and tumor proliferation characteristics, as evidenced by higher tumor purity and lower expression of immune checkpoints. Conversely, the R group exhibited a stronger correlation with immunity, as indicated by pathway enrichment and composition of TME. Patients in the NR group demonstrated higher frequencies of somatic mutations, with a 2- to 6-fold disparity between the groups in genes such as RPRD1B and CTNNB1. Patients in the R group exhibited elevated mutation scores and higher mutation frequencies at the same mutation loci compared to those in the NR group. Moreover, the incidence of mutations was more prevalent among patients in the R group. In independent cohorts, the Scissor algorithm suggests that macrophages may exert a substantial impact on immunotherapy response in patients with EC. Subsequent analysis unveiled an enrichment of M2-like tumor-associated macrophages (TAMs) within the TME of patients in the NR group. These macrophages facilitate angiogenesis and cell proliferation through intercellular communication with subpopulations such as endothelial and epithelial cells. TME of patients in the R group exhibited an enrichment of M1-like TAMs, which primarily engaged with immune cells via diverse immune-activating factors. Furthermore, immunohistochemistry and flow cytometry demonstrated that responders to immunotherapy had significantly increased infiltration of M1-like TAMs. M1-like TAMs were shown to inhibit proliferation and migration of Ishikawa cells in co-culture assays. This research offers a comprehensive insight into the multi-omics level factors influencing the immunotherapy response of EC patients, emphasizing the influence of genomic variants and TAMs on said response. This contributes to an enhanced comprehension of the biological mechanisms underlying EC immunotherapy response and aids in the optimization of EC immunotherapy.

Cell type prediction is one of the most challenging goals in single-cell RNA sequencing (scRNA-seq) data. Existing methods use unsupervised learning to identify signature genes in each cluster, followed by a literature survey to look up those genes for assigning cell types. However, finding potential marker genes in each cluster is cumbersome, which impedes the systematic analysis of single-cell RNA sequencing data. To address this challenge, we proposed a framework based on regularized multi-task learning (RMTL) that enables us to simultaneously learn the subpopulation associated with a particular cell type. Learning the structure of subpopulations is treated as a separate task in the multi-task learner. Regularization is used to modulate the multi-task model (e.g., W1, W2, … Wt) jointly, according to the specific prior. For validating our model, we trained it with reference data constructed from a single-cell RNA sequencing experiment and applied it to a query dataset. We also predicted completely independent data (the query dataset) from the reference data which are used for training. We have checked the efficacy of the proposed method by comparing it with other state-of-the-art techniques well known for cell type detection. Results revealed that the proposed method performed accurately in detecting the cell type in scRNA-seq data and thus can be utilized as a useful tool in the scRNA-seq pipeline.

BackgroundGastric cancer (GC) is still imposing a severe threat to human health. An increasing number of studies have found that neural activity plays an important role in the tumor microenvironment. However, the clinical implications of nerve-related genes (NRGs) remain largely unexplored.MethodsMatched GC and adjacent normal tissue from 8 patients were collected for single-cell RNA sequencing (scRNA-seq) analysis. Gene expression and patient information were downloaded from TCGA and GEO databases. A total of 441 NRGs were collected from the KEGG database, and LASSO regression analysis was used to construct the nerve-related risk score (NRRS). The survival, immune microenvironment, and mutation analyses were then carried out. Finally, scRNA-seq analysis was used to analyze the distribution of NRGs in GC patients.ResultsWe enrolled 441 NRGs and analyzed the association between NRGs expression and overall survival (OS). Finally, 8 NRGs highly associated with OS were identified to construct the NRRS model. Patients with low NRRS had significantly longer OS compared with high NRRS patients. We then analyzed the distribution of gene mutation landscape, enrichment annotation and immune infiltration in different NRRS subtypes. It was found that high NRRS patients displayed a significantly higher infiltration abundance of immune cell subtypes and immune checkpoint molecules. In addition, scRNA-seq was used to analyze the distribution of NRGs in 8 GC patients. We obtained 55, 052 cells in the scRNA-seq data, and the NRRS signature was significantly higher in GC tissues. EPHB3 and LPAR2 were highly expressed in epithelial cells, while NRP1, GNAI1, and SEMA6A were highly expressed in endothelial cells.ConclusionsTaken together, NRRS could serve as a stable and powerful model for survival prediction, and can help to identify GC patients who may benefit from chemotherapy and immunotherapy.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12885-025-15202-9.

Chronic obstructive pulmonary disease (COPD) is closely related to innate and adaptive inflammatory immune responses. It is increasingly becoming evident that metabolic syndrome (MetS) affects a significant portion of COPD patients. Through this investigation, we identify shared immune-related candidate biological markers. The Weighted Gene Co-Expression Network Analysis (WGCNA) was utilized to reveal the co-expression modules linked to COPD and MetS. The commonly expressed genes in the COPD and MetS were utilized to conduct an enrichment analysis. We adopted machine-learning to screen and validate hub genes. We also assessed the relationship between hub genes and immune cell infiltration in COPD and MetS, respectively. Moreover, associations across hub genes and metabolic pathways were also explored. Finally, we chose a single-cell RNA sequencing (scRNA-seq) dataset to investigate the hub genes and shared mechanisms at the level of the cells. We also applied cell trajectory analysis and cell-cell communication analysis to focus on the vital immune cell we were interested in. As a result, we selected and validated 13 shared hub genes for COPD and MetS. The enrichment analysis and immune infiltration analysis illustrated strong associations between hub genes and immunology. Additionally, we applied metabolic pathway enrichment analysis, indicating the significant role of reactive oxygen species (ROS) in COPD with MetS. Through scRNA-seq analysis, we found that ROS might accumulate the most in the alveolar macrophages. In conclusion, the 13 hub genes related to the immune response and metabolism may serve as diagnostic biomarkers and treatment targets of COPD with MetS.