Identification of key genes and biological pathways in chronic obstructive pulmonary disease using bioinformatics and next generation sequencing data analysis

Abstract

Identification of accurate biomarkers is still particularly urgent for improving the poor survival of chronic obstructive pulmonary disease (COPD) patients. In this investigation, we aimed to identity the potential biomarkers in COPD via bioinformatics and next generation sequencing (NGS) data analysis. In this investigation, the differentially expressed genes (DEGs) in COPD were identified using NGS dataset (GSE239897) from Gene Expression Omnibus (GEO) database. Subsequently, gene ontology (GO) and pathway enrichment analysis was conducted to evaluate the underlying molecular mechanisms involved in progression of COPD. Protein-protein interaction (PPI), modules, miRNA-hub gene regulatory network and TF-hub gene regulatory network analysis were performed to determine the hub genes, miRNAs and TFs. The receiver operating characteristic (ROC) analysis was performed to determine the diagnostic value of hub genes. A total of 956 overlapping DEGs (478 up regulated and 478 down regulated genes) were identified in the NGS dataset. DEGs were mainly associated with GO functional terms and pathways in cellular response to stimulus. response to stimulus, immune system and neutrophil degranulation. There were 10 hub genes (MYC, LMNA, VCAM1, MAPK6, DDX3X, SHMT2, PHGDH, S100A9, FKBP5 and RPS6KA2) identified by PPI, modules, miRNA-hub gene regulatory network and TF-hub gene regulatory network analysis. In conclusion, the DEGs, relative GO terms, pathways and hub genes identified in the present investigation might aid in understanding of the molecular mechanisms underlying COPD progression and provide potential molecular targets and biomarkers for COPD.

Introduction

Chronic obstructive pulmonary disease (COPD) represents the third leading cause of death in 2030 according to WHO prediction [1]. The clinical incidence of COPD is high, and its main features include expiratory airflow limitation that is not fully reversible, deregulated chronic airway inflammation, and emphysematous destruction of the lungs [2]. It might affect many other organs and cause other diseases include pneumonia [3], respiratory viral infections [4], pneumothorax [5], heart problems [6], osteoporosis [7], depression and anxiety [8], lung cancer [9], pulmonary hypertension [10], secondary polycythemia [11], idiopathic pulmonary fibrosis [12], obesity [13] and diabetes mellitus [14]. Airway inflammation is triggered by both genetic susceptibility [15] and environmental factors [16]. However, the etiology and symptoms of COPD are complex in clinical practice, making its diagnosis challenging.

Correct early diagnosis assessment of COPD is very difficult, even though much disease related genes and cellular pathways related to COPD have appeared [17]. The common treatments of COPD are bronchodilator and antiinflammatory treatments, such as phosphodiesterase (PDE)-4, p38 mitogen-activated protein kinase (MAPK), and nuclear factor (NF)-κB inhibitors [18], but there are no valid treatment tactics available to treat COPD. Therefore, it is vitally important to explore potential diagnostic and prognostic biomarkers, and therapeutic targets of COPD.

In recent years, next generation sequencing (NGS) technology has been applied to the research on various bioinformatics, which can screen and identify genes and signalizing pathways of various diseases [19–20]. NGS is an emerging molecular biology technology based on a high-throughput platform, which is widely used in COPD [21]. Aberrant expression of genes plays an essential role in the initiation and progression of COPD, so mastering the modification in the characteristics of essential genes promotes to comprehensively understand COPD progression and screen related molecular markers [22]. Recently, investigation shows expression of genes include SERPINE2 [23], TNFα, IL1β, and IL1RN [24], TGFB1 [25], ADRB2 [26] and PTX3 [27] in COPD. Indeed, some researchers found signaling pathways include HIFL1 signaling pathway [28], Wnt signal pathway [29], PI3K signaling in pathway [30], TLR4/NF-kB signaling pathway [31] and FoxO1/MuRF1/Atrogin-1 signaling pathway [32] were responsible for advancement of COPD. However, the use of bioinformatics analysis methods to identify the relevant genes and pathways of COPD has not yet been confirmed.

Our main purpose is to explore the connection between COPD and its associated complications. First, we download the GSE239897 [33] NGS dataset file in the NCBI Gene Expression Omnibus (GEO) [https://www.ncbi.nlm.nih.gov/geo/] [34] database for analysis, then use limma package in R software to draw the differentially expressed genes (DEGs) distribution map of the COPD and normal control samples in the dataset. Then, based on gene ontology (GO) terms and the REACTOME pathways related to COPD are analyzed. Immediately afterward, the protein-protein interaction (PPI) network, modules, miRNA-hub gene regulatory network and TF-hub gene regulatory network are drawn and the hub genes, miRNAs and TFs are identified. Hub genes were further validated by receiver operating characteristic (ROC) curve analysis. The final results will help us obtain novel treatment targets for COPD.

Materials and Methods

Next generation sequencing data source

The GEO database is an open source platform for the storage of NGS data. NGS dataset [GSE239897 (GPL17303 Ion Torrent Proton (Homo sapiens)] [33] was downloaded. The GSE239897 dataset includes 43 COPD samples and 39 normal control samples.

Identification of DEGs

The DEGs in the samples were identified by the limma package of R software [35]. The Benjamini & Hochberg False Discovery Rate correction method [36] was used to correct the P values. Fold change > 0.45 for up regulated genes, Fold change < - 0.407 for down regulated genes and adj. PLvalue <0.05 were considered to be statistically significant. ggplot2 packages of R software was applied to generate volcano plot. Hierarchical clustering analysis was performed and the gplot packages in R software was used for visualization of heatmaps.

GO and pathway enrichment analyses of DEGs

To better investigate the biological functions of DEGs, we performed functional enrichment analysis of COPD DEGs using the g:Profiler (http://biit.cs.ut.ee/gprofiler/) [37], which includes GO (http://www.geneontology.org) [38] terms and REACTOME (https://reactome.org/) [39] pathway enrichment analysis, with P < 0.05 being statistically significant. GO analysis includes biological processes (BP), cellular components (CC), and molecular functions (MF).

Construction of the PPI network and module analysis

We constructed PPI networks using pickle interactome (http://pickle.gr/) [40] database and Cytoscape visualization software version 3.10.1 (http://www.cytoscape.org/) [41]. The Network Analyzer plugin was used to obtain hub genes according to the node degree [42], betweenness [43], stress [44] and closeness [45] methods. The highest scoring modules were screened using the PEWCC [46] plugin of Cytoscape software, and the genes in the modules were defined as hub genes.

Construction of the miRNA-hub gene regulatory network

miRNAs can play a role in maintaining physiological stability by regulating the expression of hub genes. miRNA-hub gene regulatory network was constructed using the online tool of miRNet database (https://www.mirnet.ca/) [47]. We used TarBase, miRTarBase, miRecords, miRanda (S mansoni only), miR2Disease, HMDD, PhenomiR, SM2miR, PharmacomiR, EpimiR, starBase, TransmiR, ADmiRE and TAM 2 databases to find miRNAs regulating hub genes, taking the intersection of the results of these fourteen databases. We visualized miRNA-hub gene regulatory network with Cytoscape software [41].

Construction of the TF-hub gene regulatory network

TFs can play a role in maintaining physiological stability by regulating the expression of hub genes. TF-hub gene regulatory network was constructed using the online tool of NetworkAnalyst database (https://www.networkanalyst.ca/) [48]. We used Jasper database to find TFs regulating hub genes, taking the intersection of the results of this database. We visualized TF-hub gene regulatory network with Cytoscape software [41].

Receiver operating characteristic curve (ROC) analysis

ROC analysis was performed to predict the diagnostic effectiveness of COPD. We plotted ROC curves for each hub gene using the pROC package in R software [49]. The area under the ROC curve (AUC) value was utilized to determine the diagnostic effectiveness in discriminating COPD from normal control samples.

Results

Identification of DEGs

NGS dataset (GSE239897) including 43 COPD and 39 normal control samples. A total of 956 DEGs were obtained and included 478 up regulated and 478 down regulated genes (Table 1). The DEGs were visualized by the volcano plot (Fig.1) and heatmap (Fig.2).

Table 1 The statistical metrics for key differentially expressed genes (DEGs)

Fig. 1.

Volcano plot of differentially expressed genes. Genes with a significant change of more than two-fold were selected. Green dot represented up regulated significant genes and red dot represented down regulated significant genes.

Fig. 2.

Heat map of differentially expressed genes. Legend on the top left indicate log fold change of genes. (A1 – A39 = COPD samples; B1 – B 43 = Normal control samples)

GO and pathway enrichment analyses of DEGs

We performed GO terms and REACTOME pathway enrichment analysis of DEGs in the COPD samples and normal control samples. In GO BP analysis, we found that DEGs were rich in cellular response to stimulus, signaling, response to stimulus and cell communication (Table 2). In GO CC analysis, DEGs were rich in extracellular region, cell periphery, endomembrane system and membrane (Table 2). In GO MF analysis, DEGs were mainly rich in signaling receptor binding, molecular function regulator activity, catalytic activity and small molecule binding (Table 2). In REACTOME pathway enrichment analysis, DEGs were primarily rich in immune system, signaling by interleukins, neutrophil degranulation and innate immune system (Table 3).

Table 2 The enriched GO terms of the up and down regulated differentially expressed genes

Table 3 The enriched pathway terms of the up and down regulated differentially expressed genes

Construction of the PPI network and module analysis

Using the pickle platform, PPI analysis of these DEGs identified 6497 nodes and 13223 interactions. MYC, LMNA, VCAM1, MAPK6, DDX3X, SHMT2, PHGDH, S100A9, FKBP5 and RPS6KA2 were hub nodes in PPI network (Table 4). And the most significant module 1 consisting of 88 nodes and 228 edges (Fig. 4A) and module 2 consisting of 101 nodes and 107 edges (Fig 4B) were found using the PEWCC plug-in of Cytoscape software. In GO terms and REACTOME pathway enrichment analysis, hub genes were rich in the cellular response to stimulus, signaling, immune system, signaling by interleukins, signal transduction, molecular function regulator activity, cell periphery, neutrophil degranulation, innate immune system, response to stimulus, cell communication, endomembrane system, catalytic activity, metabolism, small molecule binding and hemostasis.

Fig. 3.

PPI network of DEGs. Up regulated genes are marked in parrot green; down regulated genes are marked in red.

Fig. 4.

Modules selected from the PPI network. (A) The most significant module was obtained from PPI network with 88 nodes and 228 edges for up regulated genes (B) The most significant module was obtained from PPI network with 101 nodes and 107 edges for down regulated genes. Up regulated genes are marked in parrot green; down regulated genes are marked in red.

Table 4 Topology table for up and down regulated genes

Construction of the miRNA-hub gene regulatory network

To understand the potential regulation of hub genes, we performed a comprehensive network analysis of miRNA. The miRNA-hub gene regulatory network consists of 2630 (miRNA: 2279; Hub Gene: 351) nodes and 17158 edges (Fig. 5). We found 275 miRNAs (ex: hsa-mir-410-3p) regulating DDX3X, 194 miRNAs (ex: hsa-mir-186-5p) regulating MYC, 138 miRNAs (ex: hsa-mir-1277-5p) regulating MAPK6, 130 miRNAs (ex: hsa-mir-4723-5p) regulating ZFP36, 105 miRNAs (ex: hsa-mir-432-5p) regulating BHLHE40, 116 miRNAs (ex: hsa-mir-539-5p) regulating FKBP5, 96 miRNAs (ex: hsa-mir-1268a) regulating CDC25B, 89 miRNAs (ex: hsa-mir-2861) regulating SHMT2, 80 miRNAs (ex: hsa-mir-4533) regulating TRIB3 and 57 miRNAs (ex: hsa-mir-1229-3p) regulating PHGDH (Table 5).

Fig. 5.

Hub gene - miRNA regulatory network. The ash color diamond nodes represent the key miRNAs; up regulated genes are marked in green; down regulated genes are marked in red.

Table 5 MiRNA - hub gene and TF – hub gene topology table

Construction of the TF-hub gene regulatory network

To understand the potential regulation of hub genes, we performed a comprehensive network analysis of TF. The TF-hub gene regulatory network consists of 449 (TF: 96; Hub Gene: 353) nodes and 2990 edges (Fig. 6). We found 22 TFs (ex: BRCA1) regulating TP63, 14 TFs (ex: FOXA1) regulating FOS, 12 TFs (ex: CREB1) regulating DDX3X, 12 TFs (ex: SREBF2) regulating KRT17, 12 TFs (ex: MAX) regulating BHLHE40, 17 TFs (ex: ESR1) regulating FKBP5, 16 TFs (ex: PPARG) regulating FGR, 13 TFs (ex: NFIC) regulating CDC25B, 13 TFs (ex: STAT3) regulating SHMT2 and 13 TFs (ex: NOBOX) regulating S100A8 (Table 5).

Fig. 6.

Hub gene - TF regulatory network. The light green color triangle nodes represent the key TFs; up regulated genes are marked in dark green; down regulated genes are marked in dark red.

Receiver operating characteristic curve (ROC) analysis

ROC curve analysis was performed on the hub genes. The results indicated that all 10 hub gene had good diagnostic value. The AUC value (>□0.8) of hub genes are as follows: MYC (0.889), LMNA (0.925), VCAM1 (0.906), MAPK6 (0.913), DDX3X (0.905), SHMT2 (0.908), PHGDH (0.922), S100A9 (0.918), FKBP5 (0.901) and RPS6KA2 (0.894) (Fig.7).

Fig. 7.

ROC curve analyses of hub genes. A) MYC B) LMNA C) VCAM1 D) MAPK6 E) DDX3X F) SHMT2 G) PHGDH H) S100A9 I) FKBP5 J) RPS6KA2

Discussion

In recent years, due to the lack of reliable early diagnostic tools and methods, most COPD patients have suffered breathing difficulty, which would require bronchodilator and antiinflammatory treatments. This causes leading to severe physical, mental, and economic burdens to the patients. Previous investigations have shown that the immune microenvironment might play a key role in the occurrence and development of COPD [50–51]. However, the specific targets and therapeutic mechanisms of COPD remain unclear and require further investigation. With the advancement of NGS technology, it has started to be widely applied to find the key genes of diseases. The analysis of NGS data provides a convenient and comprehensive platform to reveal the molecular pathogenesis of COPD initiation and find effective therapeutic target for the treatment of COPD.

In this investigation, we analyzed the COPD expression profile GSE239897 screened from the GEO database. It includes 43 COPD and 39 normal control samples, we found 956 DEGs (including 478 up regulated genes and 478 down regulated genes). HBB (hemoglobin subunit beta) [52], HLA-DQA1 [53], EGR1 [54], ATF3 [55], XIST (X inactive specific transcript) [56], CD163 [57] and S100A8 [58] plays a crucial role in the pathogenesis of lung cancer and considered a promising target for pharmacological based therapies. HBB (hemoglobin subunit beta) [59], HBA1 [60], HLA-DQA1 [61], EGR1 [62], ATF3 [63], XIST (X inactive specific transcript) [64], S100A12 [65], CD163 [66] and S100A8 [67] are closely related to the progression of diabetes mellitus. HBA2 [68], HBA1 [68], EGR1 [69], XIST (X inactive specific transcript) [70], S100A12 [71] and CD163 [72] reported to play a central role in pulmonary hypertension. CHIT1 [73], EGR1 [74], S100A12 [75], CD163 [76] and S100A8 [77] have a critical role in initiating COPD. CHIT1 [78], EGR1 [79], ATF3 [80], S100A12 [81], CD163 [82] and S100A8 [81] were significantly associated with idiopathic pulmonary fibrosis. CHIT1 [83], EGR1 [84], ATF3 [85], XIST (X inactive specific transcript) [86], CD163 [66] and S100A8 [87] level were significantly altered in obesity. Increasing evidence demonstrated that EGR1 [88], RNASE2 [89], CD163 [90] and S100A8 [91], and its protein product has a function in respiratory viral infections. EGR1 [92], ATF3 [93], XIST (X inactive specific transcript) [94], S100A12 [95] and S100A8 [96] were identified as a candidate genes of heart problems. EGR1 [97], ATF3 [63] and XIST (X inactive specific transcript) [98] are involved in the development of osteoporosis. A recent research suggested that EGR1 [99] and S100A8 [100] contributed to the advancement of depression and anxiety. ATF3 [101], S100A12 [102], CD163 [103] and S100A8 [104] could act as diagnosis and prognosis biomarker for airway inflammation. ATF3 [105], S100A12 [106], CD163 [107] and S100A8 [108] played an important role in the progression of pneumonia. All these findings suggested that significant DEGs act as key factors in the pathological process of COPD and deserve more attention.

Next, the DEGs were annotated by performing GO term and pathway enrichment analysis, and we observed that these genes were closely related to signaling pathways. Signaling pathways include immune system [109], signal transduction [110], extracellular matrix organization [111], adaptive immune system [112], neutrophil degranulation [113], innate immune system [114], metabolism [115], diseases of metabolism [116] and Hemostasis [117] were responsible for advancement of COPD. PTGS2 [118], IL6 [119], CSF3 [120], CXCL8 [121], CXCL1 [122], SOCS3 [123], CCL3 [124], CCL5 [125], VCAM1 [126], CXCL13 [127], CXCL5 [128], CXCL12 [129], FGG (fibrinogen gamma chain) [130], IL1B [131], AREG (amphiregulin) [132], SERPINE1 [133], OSM (oncostatin M) [134], CCL11 [135], MFAP4 [136], APLN (apelin) [137], IL1RN [138], IFNG (interferon gamma) [139], CX3CL1 [140], CDKN1A [141], ICAM1 [142], SNAI1 [143], CD34 [144], IL33 [145], FASLG (Fas ligand) [146], FGF7 [147], TNF (tumor necrosis factor) [148], GDF15 [149], CCL22 [150], IL2 [151], CXCR6 [152], KLF5 [153], MCL1 [154], SRF (serum response factor) [155], FABP4 [156], CYP2C19 [157], CXCR4 [158], FOXC1 [159], EPHX1 [160], ANXA1 [161], DUSP6 [162], CPA3 [163], MYH10 [164], ICOS (inducible T cell costimulator) [165], CD1C [166], CD96 [167], S100A9 [168], FKBP5 [169], HMOX1 [170], PRTN3 [171], MARCO (macrophage receptor with collagenous structure) [172], STAB1 [173], SIGLEC9 [174], SLC11A1 [175], BPI (bactericidal permeability increasing protein) [176], PGF (placental growth factor) [177], FASN (fatty acid synthase) [178], S100A4 [179], CD33 [180], PTX3 [181], FLT1 [182], ABCA3 [183], DUOX1 [184], KL (klotho) [185], ALOX5AP [186], FPR2 [187], IL27 [188], FZD4 [189], MPO (myeloperoxidase) [190], CTSD (cathepsin D) [191], MUC1 [192], SIGLEC14 [193], VCAN (versican) [194] and KLK1 [195] have been identified as a target for COPD. PTGS2 [196], NR4A1 [197], IL6 [198], CXCL1 [199], SOCS3 [200], CCL3 [201], CCL5 [202], CCL2 [203], NR4A2 [204], CXCL12 [205], DUSP1 [206], GADD45B [207], APLNR (apelin receptor) [208], IL1B [209], SERPINE1 [210], CCL11 [211], APLN (apelin) [212], HCAR2 [213], IFNG (interferon gamma) [214], CX3CL1 [215], LDLR (low density lipoprotein receptor) [216], TNFAIP3 [217], BMP5 [218], ITK (IL2 inducible T cell kinase) [219], ICAM1 [220], SNCA (synuclein alpha) [221], CD34 [222], HOMER1 [223], IL33 [224], CPE (carboxypeptidase E) [225], KDM6B [226], TNF (tumor necrosis factor) [227], GDF15 [228], CCL22 [229], CCR4 [230], IL2 [231], IRF1 [196], FABP4 [232], CYP2C19 [233], CXCR4 [234], TFPI2 [235], ANXA1 [236], RGS5 [237], TFF3 [238], S100A9 [239], FKBP5 [240], ARG1 [241], RASD1 [242], ADRA1A [243], STAB1 [244], ABCB6 [245], IFNGR1 [246], GPER1 [247], SLC1A3 [248], GLUL (glutamate-ammonia ligase) [249], GATA2 [250], OLFM4 [251], FADS1 [252], KL (klotho) [253], FPR2 [254], FGF22 [255], MPO (myeloperoxidase) [256], MAOA (monoamine oxidase A) [257], BRINP1 [258], CTSD (cathepsin D) [259], BAX (BCL2 associated X, apoptosis regulator) [260], FADS2 [261], EHD3 [262] and PSAT1 [263] are a major mediators of depression and anxiety. PTGS2 [264], NR4A1 [265], EGR3 [266], KLRK1 [267], IL6 [268], CXCL8 [269], CXCL2 [269], SELE (selectin E) [270], MZB1 [271], CD69 [272], ZFP36 [273], CXCL1 [274], SOCS2 [275], NR4A3 [276], TRIB1 [277], SOCS3 [278], CCL3 [279], CCL5 [280], VCAM1 [281], CCL2 [282], BTG2 [283], IER3 [284], CXCL13 [285], ITGBL1 [286], CD24 [287], CXCL5 [288], CXCL12 [289], EYA4 [290], FGG (fibrinogen gamma chain) [291], RND1 [292], DUSP1 [293], CCL19 [294], GADD45B [295], KLK5 [296], CD3E [297], INHBA (inhibin subunit beta A) [298], CXCL6 [299], IL1B [300], AREG (amphiregulin) [301], PIM2 [302], SERPINE1 [303], OSM (oncostatin M) [304], CHRNA1 [305], CCL11 [306], MFAP4 [307], MLF1 [308], CH25H [309], CD27 [310], RHOH (ras homolog family member H) [311], CD5 [312], TP63 [313], LMNA (lamin A/C) [314], CD8A [315], WIF1 [316], KLF4 [317], IL1RN [318], IFNG (interferon gamma) [319], MS4A1 [320], FHL2 [321], CX3CL1 [322], BMP2 [323], EPHA2 [324], ANO1 [325], LDLR (low density lipoprotein receptor) [326], CDKN1A [327], CCR7 [328], GBP1 [329], BMP5 [330], BMP4 [331], CTNNAL1 [332], HTRA3 [333], ICAM1 [334], SNCA (synuclein alpha) [335], MSC (musculin) [336], CLDN1 [337], SNAI1 [338], CD34 [339], DUSP26 [340], IL33 [341], SULF1 [342], HAS2 [343], EYA2 [344], THBS1 [345], FASLG (Fas ligand) [346], CPE (carboxypeptidase E) [347], TRAT1 [348], KDM6B [349], CYGB (cytoglobin) [350], RND3 [351], DERL3 [352], TNF (tumor necrosis factor) [353], GDF15 [354], CCL22 [355], ARHGAP15 [356], CCR4 [357], NFKB2 [358], TESPA1 [359], TRIM52 [360], CD74 [361], CD6 [362], ALDH1A1 [363], IL2 [364], CXCR6 [365], DDX3X[366], KLF5 [367], IRF1 [368], PIM1 [369], PLK3 [370], IRF4 [371], EGLN3 [372], IL2RG [373], MCL1 [374], GSTM2 [375], SRF (serum response factor) [376], INPP4B [377], EDN1 [378], FOSL1 [379], FABP4 [380], CYP2C19 [381], CXCR4 [382], GJA1 [383], KDM6A [384], CTSV (cathepsin V) [385], FOXC1 [386], TFPI2 [387], KLRG1 [388], TIPARP (TCDD inducible poly(ADP-ribose) polymerase) [389], EPHX1 [390], GBP5 [391], DUSP10 [392], CD200R1 [393], DKK2 [394], EPHA3 [395], MDFI (MyoD family inhibitor) [396], ARID5A [397], XBP1 [398], ARRDC3 [399], FUT8 [400], ANXA1 [401], DLC1 [402], MAPK6 [403], DCN (decorin) [404], MYO1E [405], RGS5 [406], DUSP6 [407], CEACAM5 [408], TFF3 [409], AZGP1 [410], LUM (lumican) [411], MUC13 [412], MMP19 [413], PLA2G2D [414], COL10A1 [415], CORIN (corin, serine peptidase) [416], OLR1 [417], TSPAN8 [418], MMP10 [419], ICOS (inducible T cell costimulator) [420], CD1C [421], FBLN2 [422], CTSF (cathepsin F) [423], MASP1 [424], F8 [425], SERPINB2 [426], CD1E [427], CD48 [428], HCST (hematopoietic cell signal transducer) [429], CD96 [430], DDIT4 [431], TTTY15 [432], S100A9 [433], IL1R2 [434], HMOX1 [435], MT1M [436], TMEM100 [437], CLEC4E [438], F2RL3 [439], ARG1 [440], BTNL9 [441], HIF3A [442], AQP1 [443], TRIB3 [444], G6PD [445], FPR1 [446], PGC (progastricsin) [447], PADI4 [448], ADRA1A [449], MARCO (macrophage receptor with collagenous structure) [450], FCN3 [451], LILRB2 [452], AKR1C3 [453], VSIG4 [454], F13A1 [455], EIF4EBP1 [456], ABCB6 [457], SERPINA3 [458], SHMT2 [459], NLRC4 [460], NKD1 [461], ITGA2B [462], PGF (placental growth factor) [463], TYMS (thymidylate synthetase) [464], PTP4A3 [465], ZWINT (ZW10 interacting kinetochore protein) [466], ACVRL1 [467], SDCBP2 [468], ADAMTS9 [469], GPER1 [470], TOP2A [471], FASN (fatty acid synthase) [472], IGFBP6 [473], SPINK1 [474], DYSF (dysferlin) [475], HSPA6 [476], NUPR1 [477], S100A4 [478], CD33 [479], CTSL (cathepsin L) [480], RGS18 [481], MGST1 [482], PDK4 [483], KRT13 [484], AKR1C1 [485], BIRC5 [486], E2F1 [487], KLF15 [488], ENO1 [489], PTX3 [490], IDH1 [491], GATA2 [492], FLT1 [493], OLFM4 [494], ABCA3 [495], SEMA3F [496], KNSTRN (kinetochore localized astrin (SPAG5) binding protein) [497], FADS1 [498], DUOX1 [499], IRAK1 [500], DOK3 [501], KL (klotho) [502], SLC3A2 [503], SLC39A4 [504], TBX3 [505], RAC3 [506], PCK2 [507], PLK1 [508], RASSF4 [509], IL27 [510], STC2 [511], C1R [512], PON2 [513], MCF2L [514], PYCR1 [515], GAPDH (glyceraldehyde- 3-phosphate dehydrogenase) [516], FZD4 [517], SPOCK2 [518], ME1 [519], FGF22 [520], MPO (myeloperoxidase) [521], MAOA (monoamine oxidase A) [522], SCD (stearoyl-CoA desaturase) [523], LPL (lipoprotein lipase) [524], TSPAN14 [525], CYP2A6 [526], PLIN2 [527], S100A14 [528], DOK2 [529], S1PR1 [530], GKN2 [531], SRXN1 [532], AIF1 [533], NOVA2 [534], CTSD (cathepsin D) [535], ABCC3 [536], TLR8 [537], MUC1 [538], ZDHHC9 [539], NCF2 [540], FOXF1 [541], FN1 [542], BAX (BCL2 associated X, apoptosis regulator) [543], FLT4 [544], ARHGAP11A [545], VCAN (versican) [546], CA4 [547], METTL7B [548], TKT (transketolase) [549], EHD2 [550], HTATIP2 [551], TP53I3 [552], PSAT1 [553], UBE2C [554], MTHFD2 [555], CDC25B [556], TK1 [557], BDH1 [558], RRM2 [559], PHGDH (phosphoglycerate dehydrogenase) [560], CHAC2 [561], TDRD9 [562] and PSMC5 [563] expression was significantly altered in lung cancer. PTGS2 [564], IL6 [565], CXCL8 [566], NR4A3 [567], SOCS3 [568], CCL5 [569], CCL2 [570], CXCL13 [571], DUSP5 [572], CXCL12 [573], HBEGF (heparin binding EGF like growth factor) [574], AREG (amphiregulin) [575], APLN (apelin) [576], WIF1 [577], KLF4 [578], IFNG (interferon gamma) [579], CX3CL1 [580], BMP2 [581], ANO1 [582], LDLR (low density lipoprotein receptor) [583], TNFRSF13B [584], CCR7 [585], TNFAIP3 [586], HPGDS (hematopoietic prostaglandin D synthase) [587], ICAM1 [588], CD34 [589], IL33 [590], ASPN (asporin) [591], HAS2 [592], CAMK4 [593], THBS1 [594], FASLG (Fas ligand) [595], PDE4B [596], FGF7 [597], TNF (tumor necrosis factor) [598], GDF15 [149], CD74 [599], HTR2B [600], KLF5 [601], PIM1 [602], SRF (serum response factor) [603], EDN1 [604], CXCR4 [605], KDM6A [606], FOXC1 [607], SELP (selectin P) [608], FUT8 [609], ANXA1 [610], RGS5 [611], ECM2 [612], THBD (thrombomodulin) [613], MMP10 [614], ICOS (inducible T cell costimulator) [615], FBLN2 [616], F8 [617], HMOX1 [618], HIF3A [619], AQP1 [620], TRIB3 [621], G6PD [622], ABCC8 [623], EIF4EBP1 [624], SHMT2 [625], PGF (placental growth factor) [626], ACVRL1 [627], FASN (fatty acid synthase) [628], S100A4 [629], PDK4 [630], KLF15 [631], ENO1 [632], PTX3 [633], GATA2 [634], FLT1 [635], ABCA3 [636], KL (klotho) [637], PLK1 [638], GAPDH (glyceraldehyde-3- phosphate dehydrogenase) [639], MPO (myeloperoxidase) [640], LPL (lipoprotein lipase) [641], TLR8 [642], MUC1 [643], FOXF1 [644], VCAN (versican) [645] and FDPS (farnesyl diphosphate synthase) [646] have been proposed as the most promising biomarkers for pulmonary hypertension. PTGS2 [647], NR4A1 [648], IL6 [649], CXCL8 [650], CD69 [651], CXCL1 [652], SOCS2 [653], SOCS3 [654], CCL3 [655], CCL5 [656], VCAM1 [657], CCL2 [658], CXCL5 [659], DUSP5 [660], CXCL12 [661], KLF10 [662], KLK5 [663], CD3E [664], CD3D [665], IL1B [666], AREG (amphiregulin) [667], SERPINE1 [668], OSM (oncostatin M) [669], CD83 [670], MFAP4 [671], APLN (apelin) [672], RARRES2 [673], CH25H [674], CD27 [675], LMNA (lamin A/C) [676], KLF4 [677], IL1RN [678], IFNG(interferon gamma) [679], FHL2 [680], CX3CL1 [681], GZMA (granzyme A) [682], BMP2 [683], LDLR (low density lipoprotein receptor) [684], CDKN1A [685], RGS16 [686], TNFAIP3 [687], BMP4 [688], ICAM1 [689], SNCA (synuclein alpha) [690], CD34 [691], IL33 [692], FASLG (Fas ligand) [693], CPE (carboxypeptidase E) [694], PDE4B [695], FGF7 [696], ITGB7 [697], RND3 [698], TNF (tumor necrosis factor) [699], CASQ2 [700], GDF15 [701], CCR4 [702], CD74 [703], IL2 [704], PPP1R15B [705], CXCR6 [706], HTR2B [707], DDX3X [708], KLF5 [709], IRF1 [710], PIM1 [711], PLK3 [712], IRF4 [713], MCL1 [714], SRF (serum response factor) [715], SEMA3E [716], EDN1 [717], FOSL1 [718], FABP4 [719], RASGRP1 [720], CYP2C19 [721], CXCR4 [722], TFPI2 [723], EPHX1 [724], SELP (selectin P) [725], LTBP4 [726], ADCYAP1 [727], XBP1 [728], ANXA1 [729], DCN (decorin) [730], RGS5 [731], SIRPG (signal regulatory protein gamma) [732], PTPN22 [733], LUM (lumican) [734], STEAP4 [735], CORIN (corin, serine peptidase) [736], TSPAN8 [737], MMP10 [738], ICOS (inducible T cell costimulator) [739], MASP1 [740], F8 [741], CD48 [742], HLA-DOA [743], DDIT4 [744], S100A9 [745], P2RY1 [746], FKBP5 [747], HMOX1 [748], ITGA10 [749], ARG1 [750], HIF3A [751], AQP1 [752], TRIB3 [753], G6PD [754], ABCC8 [755], SLC11A1 [756], NPR3 [757], NLRC4 [758], HHEX (hematopoietically expressed homeobox) [759], ADAMTS9 [760], FASN (fatty acid synthase) [761], SPINK1 [762], S100A4 [763], CD33 [764], CTSL (cathepsin L) [765], NID1 [766], PDK4 [767], E2F1 [768], KLF15 [769], PTX3 [770], GLUL (glutamate-ammonia ligase) [249], FLT1 [771], OLFM4 [772], FADS1 [773], IRAK1 [774], KL (klotho) [775], ALOX5AP [776], IL15RA [777], EFNB1 [778], PCK2 [779], PLK1 [780], FPR2 [781], IL27 [782], PON2 [783], GAPDH (glyceraldehyde-3-phosphate dehydrogenase) [784], FZD4 [785], GPBAR1 [786], MPO (myeloperoxidase) [787], MAOA (monoamine oxidase A) [788], SCD (stearoyl-CoA desaturase) [789], LPL (lipoprotein lipase) [790], KCNJ11 [791], CYP2A6 [792], COL4A1 [793], AIF1 [794], GRB10 [795], DGKD (diacylglycerol kinase delta) [796], CTSD (cathepsin D) [797], TLR8 [798], MUC1 [799], BAX (BCL2 associated X, apoptosis regulator) [800], DOK1 [801], ECHDC3 [802], VCAN (versican) [803], PCSK2 [804], FADS2 [805], TKT (transketolase) [806], NCF4 [807], HTATIP2 [808], PNMT (phenylethanolamine N-methyltransferase) [809], KLK1 [810], HSD17B14 [811], ACOT1 [812], TK1 [813] and AFMID (arylformamidase) [814] have been identified as high-risk genes in diabetes mellitus. NR4A1 [815], IL6 [816], CXCL8 [817], CD69 [818], CXCL1 [819], SOCS3 [820], CCL3 [821], CCL5 [822], VCAM1 [823], CCL2 [824], CXCL13 [825], CD24 [826], CXCL5 [827], CXCL12 [828], OSM (oncostatin M) [829], CH25H [830], CD27 [831], KLF4 [832], IFNG (interferon gamma) [833], FHL2 [834], CX3CL1 [835], GZMA (granzyme A) [836], CCR7 [837], ICAM1 [838], CD34 [823], IL33 [839], FASLG (Fas ligand) [840], CPE (carboxypeptidase E) [841], KDM6B [842], TNF (tumor necrosis factor) [843], CCR4 [844], CD40LG [845], IL2 [846], IRF1 [847], IL2RG [848], FABP4 [849], CXCR4 [850], SELP (selectin P) [851], XBP1 [852], ANXA1 [853], MAPK6 [854], THBD (thrombomodulin) [855], ICOS (inducible T cell costimulator) [856], F8 [857], DDIT4 [858], S100A9 [108], ARG1 [859], ELANE (elastase, neutrophil expressed) [860], G6PD [861], MARCO (macrophage receptor with collagenous structure) [862], NLRC4 [863], BPI (bactericidal permeability increasing protein) [864], PGF (placental growth factor) [865], S100A4 [866], PTX3 [867], GATA2 [868], ABCA3 [869], FPR2 [870], IL27 [871], GAPDH (glyceraldehyde-3- phosphate dehydrogenase) [872], MPO (myeloperoxidase) [873], FOXF1 [874], VCAN (versican) [875] and FUOM (fucose mutarotase) [876] were revealed to serve an important role in pneumonia. NR4A1 [877], IL6 [878], CSF3 [879], CXCL8 [880], EGR2 [881], ZFP36 [882], CXCL1 [883], SOCS3 [884], CCL3 [885], CCL5 [886], CXCL13 [887], NR4A2 [888], GPR183 [889], CD24 [890], CXCL5 [891], DUSP5 [892], DUSP1 [892], CCL19 [893], CXCL6 [894], IL1B [895], SERPINE1 [896], OSM (oncostatin M) [897], CCL11 [898], CD83 [899], APLN (apelin) [900], RARRES2 [901], CH25H [902], IL1RN [903], IFNG (interferon gamma) [833], FHL2 [904], CX3CL1 [905], EPHA2 [906], ANO1 [907], LDLR (low density lipoprotein receptor) [908], CCR7 [909], SIT1 [910], TNFAIP3 [911], GBP1 [912], HPGDS (hematopoietic prostaglandin D synthase) [913], ICAM1 [914], CLDN1 [915], IL33 [916], THBS1 [917], FASLG (Fas ligand) [918], PDE4B [919], TNF (tumor necrosis factor) [843], GDF15 [920],D74 [921], IL2 [846], CXCR6 [922], DDX3X [923], IRF1 [924], PIM1 [925], ZBP1 [926], PLK3 [927], IRF4 [928], MCL1 [929], CYP2C19 [930], CXCR4 [931], GBP5 [932], SELP (selectin P) [933], DUSP10 [934], XBP1 [852], ANXA1 [935], DUSP6 [936], MMP19 [937], CFD (complement factor D) [938], TSPAN8 [939], MMP10 [940], F8 [857], HLA-DOB [941], CD177 [942], S100A9 [943], FKBP5 [944], HMOX1 [945], ARG1 [946], IL18R1 [947], TRIB3 [948], ELANE (elastase, neutrophil expressed) [949], G6PD [861], MARCO (macrophage receptor with collagenous structure) [950], IL17RB [951], F13A1 [952], NLRC4 [953], BPI (bactericidal permeability increasing protein) [954], GPER1 [955], FASN (fatty acid synthase) [956], HSPA6 [957], CD33 [958], CTSL (cathepsin L) [959], PTX3 [960], OLFM4 [961], ABCA3 [962], IRAK1 [963], DOK3 [964], PLK1 [965], FPR2 [966], IL27 [967], STC2 [968], GAPDH (glyceraldehyde-3-phosphate dehydrogenase) [969], ITGAM (integrin subunit alpha M) [970], MPO (myeloperoxidase) [971], TLR8 [972], MUC1 [973], FLT4 [974], VCAN (versican) [975] and CDC25B [976] have been reported to act as a potential biomarkers for respiratory viral infections treatment. NR4A1 [977], IL6 [978], CXCL8 [979], CD69 [980], ZFP36 [981], CXCL1 [982], SOCS2 [983], SOCS3 [984], CCL5 [985], VCAM1 [986], CCL2 [987], BTG2 [988], CD24 [989], CXCL5 [982], KLF10 [990], FGG (fibrinogen gamma chain) [991], DUSP2 [992], DUSP1 [993], CCL19 [994], APLNR (apelin receptor) [995], IL1B [996], SERPINE1 [997], OSM (oncostatin M) [998], APLN (apelin) [999], CH25H [674], CD27 [675], LMNA (lamin A/C) [676], WIF1 [1000], KLF4 [1001], IL1RN [996], IFNG (interferon gamma) [1002], FHL2 [1003], CX3CL1 [1004], BMP2 [1005], LDLR (low density lipoprotein receptor) [1006], CCR7 [1007], BMP4 [1008], CCDC3 [1009], ICAM1 [1010], CD34 [1011], IL33 [1012], THBS1 [1013], FASLG (Fas ligand) [1014], CPE (carboxypeptidase E) [694], PDE4B [1015], TNF (tumor necrosis factor) [1016], GDF15 [1017], CCL22 [1018], CD74 [1019], ALDH1A1 [1020], IL2 [1021], HTR2B [707], PIM1 [1022], IRF4 [1023], MCL1 [1024], SEMA3E [716], FABP4 [1025], CYP2C19 [1026], OMA1 [1027], CXCR4 [1028], KDM6A [1029], SELP (selectin P) [1030], EPHA3 [1031], XBP1 [1032], ANXA1 [1033], RGS5 [731], DUSP6 [1034], LUM (lumican) [734], STEAP4 [1035], MMP19 [1036], DPT (dermatopontin) [1037], CFD (complement factor D) [1038], CORIN (corin, serine peptidase) [1039], OLR1 [1040], MMP10 [1041], ENPP2 [1042], AIF1L [1043], DDIT4 [1044], S100A9 [1045], FKBP5 [1046], HMOX1 [1047], ARG1 [1048], HIF3A [1049], PRTN3 [1050], TRIB3 [1051],ELANE (elastase, neutrophil expressed) [1052], G6PD [1053], ABCC8 [1054], F13A1 [1055], NPR3 [1056], NLRC4 [1057], PGF (placental growth factor) [1058], TOP2A [1059], FASN (fatty acid synthase) [1060], SPINK1 [1061], NUPR1 [1062], S100A4 [1063], CTSL (cathepsin L) [1064], NID1 [1065], APCDD1 [1066], PDK4 [1067], BIRC5 [1059], E2F1 [1068], KLF15 [1069], PTX3 [1070], GATA2 [1071], OLFM4 [1072], FADS1 [1073], DUOX1 [1074], IRAK1 [1075], ABI3 [1076], ALOX5AP [1077], BCAT2 [1078], PCK2 [779], PLK1 [1079], IL27 [1080], PON2 [1081], ME1 [1082], GPBAR1 [1083], MPO (myeloperoxidase) [1084], MAOA (monoamine oxidase A) [1085], SCD (stearoyl- CoA desaturase) [789], LPL (lipoprotein lipase) [1086], MPST (mercaptopyruvate sulfurtransferase) [1087], CYP2A6 [1088], PLIN2 [1089], LAMC1 [1090], AIF1 [1091], CTSD (cathepsin D) [1092], TLR8 [798], DOK1 [1093], PICK1 [1094], FLT4 [1095], VCAN (versican) [1096], FADS2 [1097], TKT (transketolase) [1098], PNMT (phenylethanolamine N-methyltransferase) [1099], PHGDH (phosphoglycerate dehydrogenase) [1100] and CHDH (choline dehydrogenase) [1101] have been previously reported to be an effective diagnostic biomarkers for obesity. NR4A1 [648], IL6 [1102], CXCL2 [1103], MZB1 [1104], CD69 [92], CXCL1 [1105], SOCS2 [1106], TRIB1 [1107], SOCS3 [1108], CCL3 [655], CCL5 [1109], VCAM1 [1110], NFKBIZ (NFKB inhibitor zeta) [1111], CCL2 [1112], IER3 [1113], NR4A2 [1114], ITGBL1 [1115], DUSP5 [572], CXCL12 [1116], EYA4 [1117], JUND (JunD proto-oncogene, AP-1 transcription factor subunit) [1118], APLNR (apelin receptor) [208], IL1B [1119], AREG (amphiregulin) [1120], OSM (oncostatin M) [669], SMOC2 [1121], MFAP4 [1122], APLN (apelin) [212], RARRES2 [1123], CH25H [1124], TP63 [1125], AADAC (arylacetamide deacetylase) [1126], LMNA (lamin A/C) [1127], KLF4 [1128], IL1RN [1129], GPR174 [1130], IFNG (interferon gamma) [1131], FHL2 [1132], EPHA2 [1133], LDLR (low density lipoprotein receptor) [684], TNFAIP3 [1134], ICAM1 [1110], CD34 [1135], HOMER1 [1136], DUSP26 [1137], IL33 [1138], ASPN (asporin) [1139], FASLG (Fas ligand) [1140], CPE (carboxypeptidase E) [1141], RND3 [698], TNF (tumor necrosis factor) [1142], CASQ2 [700], GDF15 [1143], CCR4 [1144], IL2 [1145], KLF5 [709], PIM1 [711], MCL1 [714], SRF (serum response factor) [715], FABP4 [1146], CYP2C19 [1147], OMA1 [1148], CXCR4 [234], GJA1 [1149], FOXC1 [1150], TFPI2 [723], LTBP4 [1151], TNFSF18 [1152], XBP1 [1153], ACTN2 [1154], ANXA1 [1155], DCN (decorin) [730], RGS5 [1156], DUSP6 [1157], TFF3 [1158], CFD (complement factor D) [1159], THBD (thrombomodulin) [1160], MYH10 [1161], CORIN (corin, serine peptidase) [736], OLR1 [1162], F8 [1163], CPXM2 [1164], SYTL3 [1165], SLAMF7 [1166], DEFA3 [1167], TTTY15 [1168], S100A9 [239], IL1R2 [1169], P2RY1 [1170], HMOX1 [748], CLEC4E [1171], F2RL3 [1072], ARG1 [1173], HIF3A [751], AQP1 [752], ELANE (elastase, neutrophil expressed) [1174], G6PD [1175], VSIG4 [1176], F13A1 [1177], BPI (bactericidal permeability increasing protein) [1178], ADAMTS9 [1179], DYSF (dysferlin) [1180], S100A4 [629], CTSL (cathepsin L) [1181], BIRC5 [1182], KLF15 [769], PTX3 [1183], GATA2 [1071], FLT1 [1184], SEMA3F [1185], FADS1 [773], PRKCE (protein kinase C epsilon) [1186], KL (klotho) [1187], ALOX5AP [1188], TCAP (titin-cap) [1189], BCAT2 [1190], FPR2 [1191], IL27 [1192], PON2 [1193], ALPL (alkaline phosphatase, biomineralization associated) [1194], MPO (myeloperoxidase) [1195], MAOA (monoamine oxidase A) [1196], LPL (lipoprotein lipase) [1197], KCNJ11 [1198], CYP2A6 [1199], COL4A2 [1200], COL4A1 [1201], S1PR1 [1203], GRB10 [795], CTSD (cathepsin D) [1203], FOXF1 [1204], BAX (BCL2 associated X, apoptosis regulator) [1205], VCAN (versican) [1206], TKT (transketolase) [1207], NCF4 [807], LRRC25 [1208], KLK1 [1209], ACOT1 [812], ADAMTS2 [1210] and BDH1 [1211] have been reported to be associated with heart problems. Studies have shown that IL6 [1212], CSF3 [1213], CXCL8 [1214], CXCL2 [1215], CD69 [1216], CXCL1 [1217], SOCS2 [1218], SOCS3 [1219], CCL5 [1220], VCAM1 [1221], CCL2 [1222], CXCL12 [1223], KLF10 [1224], CCL19 [1225], CXCL6 [1226], IL1B [1227], AREG (amphiregulin) [1228], OSM (oncostatin M) [1229], CCL11 [1230], APLN (apelin) [1231], CH25H [1232], KLF4 [1233], IFNG (interferon gamma) [1234], FHL2 [1235], CX3CL1 [1236], EPHA2 [1237], LIF (LIF interleukin 6 family cytokine) [1238], RGS16 [1239], CCR7 [1240], TNFAIP3 [1241], BMP4 [1242], ITK (IL2 inducible T cell kinase) [1243], ICAM1 [1244], CD34 [1245], IL33 [1246], SULF1 [1247], HAS2 [1248], FASLG (Fas ligand) [1249], CPE (carboxypeptidase E) [1250], TNF (tumor necrosis factor) [1251], GDF15 [1252], CCL22 [1253], CCR4 [1254], CD74 [1255], CXCR6 [1256], HTR2B [1257], IRF1 [1258], PIM1 [1259], ZBP1 [926], MCL1 [1260], FABP4 [1261], CXCR4 [1262], KLRG1 [1263], XBP1 [1264], ANXA1 [1265], RGS5 [1266], MMP19 [1267], THBD (thrombomodulin) [1268], MYH10 [164], CD48 [1269], S100A9 [1270], ARG1 [1271], AQP1 [1272], MARCO (macrophage receptor with collagenous structure) [1273], FCN3 [1274], IL17RB [1275], SIGLEC9 [1276], IGFBP6 [473], S100A4 [1277], PTX3 [1278], ABCA3 [1279], DUOX1 [1074], PRKCE (protein kinase C epsilon) [1280], KL (klotho) [1281], SLC3A2 [1282], FPR2 [1283], IL27 [1284], MPO (myeloperoxidase) [1285], CCL26 [1286], CTSD (cathepsin D) [1287], MUC1 [1288], FOXF1 [1289], CTSE (cathepsin E) [1290] and VCAN (versican) [975] are associated with airway inflammation. Recent studies have indicated that IL6 [1291], CXCL8 [1292], CXCL2 [1293], MZB1 [1294], CD69 [1295], ZFP36 [1296], CCL3 [1297], CCL5 [1298], VCAM1 [1299], CCL2 [1300], CXCL13 [1301], ITGBL1 [1302], CXCL12 [1303], CXCL6 [1304], IL1B [1305], AREG (amphiregulin) [1306], SERPINE1 [1307], OSM (oncostatin M) [1308], CCL11 [1298], SMOC2 [1309], MFAP4 [1310], APLN (apelin) [1231], PPP1R15A [1311], TP63 [1312], LMNA (lamin A/C) [1313], WIF1 [1314], KLF4 [1315], IL1RN [1316], IFNG (interferon gamma) [1317], FHL2 [1318], CX3CL1 [1319], EPHA2 [1320], ANO1 [1321], LDLR (low density lipoprotein receptor) [1322], CCR7 [1323], BMP4 [1324], HTRA3 [1325], ICAM1 [1326], SNAI1 [1327], CD34 [1328], IL33 [1329], LTBP2 [1330], HAS2 [1331], FASLG (Fas ligand) [1332], PDE4B [1333], KDM6B [1334], RND3 [1335], TNF (tumor necrosis factor) [1336], GDF15 [1337], CCL22 [1338], CCR4 [1339], HTR2B [1340], IRF1 [1341], PIM1 [1342], IRF4 [1343], BMP3 [1344], SRF (serum response factor) [1345], EDN1 [1346], FABP4 [1347], CXCR4 [1348], WNT10A [1349], DUSP10 [1350], NREP (neuronal regeneration related protein) [1351], ANXA1 [1352], CPA3 [163], MMP19 [1353], MMP10 [1354], ICOS (inducible T cell costimulator) [1355], CD1C [1356], S100A9 [81], HMOX1 [1357], AQP1 [1358], IL18R1 [1359], TRIB3 [1360], FPR1 [1361], MARCO (macrophage receptor with collagenous structure) [1362], PGF (placental growth factor) [1363], FASN (fatty acid synthase) [1364], S100A4 [1365], CTSL (cathepsin L) [1366], E2F1 [1367], PTX3 [1368], GATA2 [634], FLT1 [1369], ABCA3 [636], DUOX1 [1370], IRAK1 [1371], KL (klotho) [1372], FPR2 [1373], IL27 [1374], MPO (myeloperoxidase) [1375], S1PR1 [1376], AIF1 [1377], MUC1 [1378], FOXF1 [1379], BAX (BCL2 associated X, apoptosis regulator) [1380], PSAT1 [1381] and PHGDH (phosphoglycerate dehydrogenase) [1382] are involved in the development of idiopathic pulmonary fibrosis. IL6 [1383], CSF3 [1384], SOCS3 [1385], CCL3 [1386], CCL5 [1387], VCAM1 [1388], CCL2 [1387], CXCL12 [1389], KLF10 [1390], FRZB (frizzled related protein) [1391], DUSP1 [1392], IL1B [1393], CCL11 [1394], LMNA (lamin A/C) [1395], WIF1 [1396], IL1RN [1397], IFNG (interferon gamma) [1398], CX3CL1 [1399], BMP2 [1400], CDKN1A [1401], BMP4 [1402], ICAM1 [1403], CD34 [1404], IL33 [1405], FASLG (Fas ligand) [1406], KDM6B [1407], TNF (tumor necrosis factor) [1408], GDF15 [1409], IL2 [1410], INPP4B [1411], FABP4 [1412], CYP2C19 [1413], CXCR4 [1414], DKK2 [1415], PLCB4 [1416], ANXA1 [1417], DUSP6 [1418], CORIN (corin, serine peptidase) [1419], F8 [1420], DDIT4 [1421], IL1R2 [1422], S100A4 [1417], CTSL (cathepsin L) [1423], PTX3 [1424], EFNB1 [1425], MCF2L [1426], FZD4 [1427], CCL26 [1428], PHGDH (phosphoglycerate dehydrogenase) [1429] and FDPS (farnesyl diphosphate synthase) [1430] have been identified as predictive biomarkers of osteoporosis. IL6 [1431], SOCS2 [1432], SOCS3 [1432], DUSP1 [1433], MLF1 [1434], EPHA2 [1435], CD34 [1436], IL33 [1437], GDF15 [1438], IL2 [1439], CD177 [1440], G6PD [1441], NFE2 [1442], PTX3 [1443], IDH1 [1444], MPO (myeloperoxidase) [1445] and KLK1 [1446] might be associated with the pathogenesis of polycythemia. The abnormal expression of IL33 [1447] and TNF (tumor necrosis factor) [1448] contributes to the pneumothorax. The above results suggest that enriched genes maight influence the progression of COPD.

Using the DEGs genes, a PPI network was constructed and significant modules were isolated on understanding the molecular pathogenesis of COPD, and predicting prognostic and diagnostic biomarkers as well as therapeutic targets. Here, we identified hub genes based on the topological properties (i.e., degree, betweenness, stress and closeness), which can be key biomarkers in COPD and linked with numerous molecular mechanisms. The top hub genes indicate most risk factors for the COPD. Research has shown that LMNA (lamin A/C) [314], VCAM1 [281], MAPK6 [403], DDX3X [366], SHMT2 [459], PHGDH (phosphoglycerate dehydrogenase) [560], S100A9 [433], SNCA (synuclein alpha) [335], TP63 [313], FOSL2 [1449], FHL2 [321], ATF3 [55], FOSL1 [379], DYSF (dysferlin) [475], CA4 [547], S100A8 [58] and MPO (myeloperoxidase) [521] plays an important role in the pathogenesis of lung cancer. LMNA (lamin A/C) [676], VCAM1 [657], DDX3X [708], S100A9 [745], FKBP5 [747], SNCA (synuclein alpha) [690], FOSL2 [1450], FHL2 [680], ATF3 [63], FOSL1 [718], S100A8 [67] and MPO (myeloperoxidase) [787] altered expression is associated with poor prognosis of diabetes mellitus. LMNA (lamin A/C) [676], VCAM1 [986], PHGDH (phosphoglycerate dehydrogenase) [1100], S100A9 [1045], FKBP5 [1046], FHL2 [1003], ATF3 [85] and MPO (myeloperoxidase) [1084] plays an important role in the development of obesity. LMNA (lamin A/C) [1127], VCAM1 [1110], S100A9 [239], TP63 [1125], JUND (JunD proto-oncogene, AP-1 transcription factor subunit) [1118], FHL2 [1132], ATF3 [93], BPI (bactericidal permeability increasing protein) [1178], DYSF (dysferlin) [1180], MPO (myeloperoxidase) [1195] and ALPL (alkaline phosphatase, biomineralization associated) [1194] were reported to be associated with the prognosis of heart problems. LMNA (lamin A/C) [1313], VCAM1 [1299], PHGDH (phosphoglycerate dehydrogenase) [1382], S100A9 [81], TP63 [1312], FHL2 [1318], ATF3 [80] and MPO (myeloperoxidase) [1375] have been identified as a key genes in idiopathic pulmonary fibrosis. LMNA (lamin A/C) [1395], VCAM1 [1388], PHGDH (phosphoglycerate dehydrogenase) [1429] and ATF3 [63] might be a potential targets for osteoporosis. VCAM1 [126], S100A9 [168], FKBP5 [169], BPI (bactericidal permeability increasing protein) [176] and MPO (myeloperoxidase) [190] are mainly associated with COPD. VCAM1 [823], MAPK6 [854], S100A9 [108], FHL2 [834], ATF3 [105], BPI (bactericidal permeability increasing protein) [864] and MPO (myeloperoxidase) [873] have a significant prognostic potential in pneumonia. VCAM1 [1221], S100A9 [1270], FHL2 [1235], ATF3 [101] and MPO (myeloperoxidase) [1285] are molecular markers for the diagnosis and prognosis of airway inflammation. DDX3X [923], S100A9 [943], FKBP5 [944], FHL2 [904], ITGAM (integrin subunit alpha M) [970], BPI (bactericidal permeability increasing protein) [954], MPO (myeloperoxidase) [971] and CD177 [942] have a significant prognostic potential in respiratory viral infections. Research reported that increased level of SHMT2 [625] and MPO (myeloperoxidase) [640] are related to pulmonary hypertension, as a novel target for pulmonary hypertension diagnosis and treatment. A recent study found that the S100A9 [239], FKBP5 [240], SNCA (synuclein alpha) [221] and MPO (myeloperoxidase) [256] are crucial for the progression of depression and anxiety. Regulation of MPO (myeloperoxidase) [1445] and CD177 [1440] levels might be a novel treatment option against polycythemia. The results of this investigation suggest that novel biomarkers include MYC (MYC proto-oncogene, bHLH transcription factor), RPS6KA2, SH2D1A, JUNB (JunB proto-oncogene, AP-1 transcription factor subunit), FOS (Fos proto-oncogene, AP-1 transcription factor subunit), RAB39A and RAB15 might play a key role in the pathogenesis of COPD. This investigation has been proved to be a useful approach to identify novel biomarkers in other diseases.

A miRNA-hub gene regulatory network and TF-hub gene regulatory network were constructed for the identified hub genes, miRNAs and TFs were defined by the degree rank. DDX3X [366], MAPK6 [403], ZFP36 [273], BHLHE40 [1451], TP63 [313], KRT17 [1452], CDC25B [556], SHMT2 [459], TRIB3 [444], PHGDH (phosphoglycerate dehydrogenase) [560], S100A8 [58], hsa-mir-410-3p [1453], hsa-mir-186-5p [1454], hsa-mir-432-5p [1455], BRCA1 [1456], FOXA1 [1457], CREB1 [1458], MAX (MYC associated factor X) [1459], ESR1 [1460], NFIC (nuclear factor I C) [1461] and STAT3 [1462] expression might be regarded as an indicator of susceptibility to lung cancer. DDX3X [708], BHLHE40 [1463], FKBP5 [747], TRIB3 [753], S100A8 [67], FOXA1 [1464], CREB1 [1465], SREBF2 [1466], ESR1 [1467] and STAT3 [1468] are a potential markers for the detection and prognosis of diabetes mellitus. A previous study reported that DDX3X [923], ZFP36 [882], FKBP5 [944], CDC25B [976], TRIB3 [948], hsa-mir-1277-5p [1469], MAX (MYC associated factor X) [1470] and STAT3 [1471] are altered expressed in respiratory viral infections. MAPK6 [854], BRCA1 [1472] and STAT3 [1473] are found to be associated with pneumonia. ZFP36 [981], FKBP5 [1046], TRIB3 [1051], PHGDH (phosphoglycerate dehydrogenase) [1100], BRCA1 [1474], SREBF2 [1475], ESR1 [1476] and STAT3 [1477] plays a vital role in the development of obesity. ZFP36 [1296], TP63 [1312], KRT17 [1478], TRIB3 [1360], PHGDH (phosphoglycerate dehydrogenase) [1382], FOXA1 [1479], CREB1 [1480], SREBF2 [1481] and STAT3 [1482] have been observed to be expressed in idiopathic pulmonary fibrosis patients. TP63 [1125], hsa-mir-410-3p [1483], hsa-mir-432-5p [1484], BRCA1 [1485], ESR1 [1486], NFIC (nuclear factor I C) [1487] and STAT3 [1488] might serve as therapeutic targets for heart problems. The expression levels of FKBP5 [169] and STAT3 [1489] have been proved to be altered in COPD patients. Recent studies have shown that altered expression of FKBP5 [240], BRCA1 [1490], CREB1 [1491], MAX (MYC associated factor X) [1492], ESR1 [1493] and STAT3 [1494] might be associated with the progression of depression and anxiety. Studies have found that SHMT2 [625], TRIB3 [621], BRCA1 [1495] and STAT3 [1496] expression was associated with pulmonary hypertension. PHGDH (phosphoglycerate dehydrogenase) [1429], BRCA1 [1497], FOXA1 [1498], CREB1 [1499], ESR1 [1500] and STAT3 [1501] might be a prognostic biomarker and potential therapeutic target for patients with osteoporosis. BRCA1 [1502], ESR1 [1503] and STAT3 [1504] have been linked to the start of airway inflammation. STAT3 [1505] is implicated in polycythemia. MYC (MYC proto-oncogene, bHLH transcription factor), FOS (Fos proto-oncogene, AP-1 transcription factor subunit), FGR (FGR proto-oncogene, Src family tyrosine kinase), hsa-mir-4723-5p, hsa-mir-539-5p, hsa-mir-1268a, hsa-mir-2861, hsa-mir-4533, hsa-mir-1229-3p, PPARG (peroxisome proliferator activated receptor gamma) and NOBOX (NOBOX oogenesis homeobox) migt be considered to be a novel biomarkers for COPD. These findings suggest that the hub genes, miRNAs and TFs can serve as candidate diagnostic biomarkers or drug targets for the clinical treatment of COPD.

In conclusion, we filtrated a total of 956 DEGs from GEO NGS dataset and further validated 10 hub genes (MYC, LMNA, VCAM1, MAPK6, DDX3X, SHMT2, PHGDH, S100A9, FKBP5 and RPS6KA2). The 10 hub genes were might associated with the prognosis of COPD. The GO functional terms and pathways identified in the investigation migt contribute to understand the molecular mechanisms of COPD. Our findings might provide novel therapeutic targets for COPD patients.

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Informed consent

No informed consent because this study does not contain human or animals participants.

Availability of data and materials

The datasets supporting the conclusions of this article are available in the GEO (Gene Expression Omnibus) (https://www.ncbi.nlm.nih.gov/geo/) repository. [(GSE239897) https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE239897]

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Author Contributions

B. V. - Writing original draft, and review and editing

C. V. - Software and investigation

Acknowledgement

I thanks very much to John McDonough, Yale School of Medicine, New Haven, CT, USA, the author who deposited their NGS dataset GSE239897, into the public GEO database.