Identification of hub genes with diagnostic values in pancreatic cancer by bioinformatics analyses and supervised learning methods

Systems biology approach: identification of hub genes, signaling pathways, and molecular docking of COL1A1 gene in cervical insufficiency

The collagen type I alpha 1 (COL1A1, OMIM #120,150) gene, encoding the alpha-1 chain of type I collagen (UniProt #P02452), plays a key role in life-homeostasis due to its remarkable involvement in collagen synthesis. It is a promising candidate gene implicated in the pathogenesis of cervical insufficiency (CI). This study aimed to identify genetic variations within the COL1A1 gene that contribute to the development of CI. Polymerase chain reaction (PCR) and amplicon sequencing were implemented for single nucleotide polymorphisms (SNPs) detection (+ 1245G/T, SP1 rs1800012), which revealed wild-type sequence for targeted SNPs in enrolled proband indicated negative results regarding COL1A1 gene involvement for current form of CI. It allows further investigation of other closely connected genes probed in this study. Computational approaches viz. Protein-protein interaction (PPI), gene ontology (GO), and pathway participation were used to identify the crucial hub genes and signaling pathways for COL1A1 and CI. Using the Yet Another Scientific Artificial Reality Application (YASARA) software, molecular docking, and molecular dynamic (MD) simulation with the oxytocin (CID 439,302), estradiol (CID 129,728,744), progesterone (CID 5994) and hydroxyprogesterone (CID 150,788) were done. Interactive bioinformatics analysis demonstrated that the COL1A1 and more than 10 collagen sister genes had a strong connection with CI. In sum, the findings of this study provide insights into a modus operandi that can be utilized to illuminate the path toward studying sister genes and smooth diagnosis of CI. These findings have implications for understanding the foundational process of the condition and potentially developing screening, diagnostic, and therapeutic interventions.

Graphical Abstract

Multiple-model machine learning identifies potential functional genes in dilated cardiomyopathy

Introduction

Machine learning (ML) has gained intensive popularity in various fields, such as disease diagnosis in healthcare. However, it has limitation for single algorithm to explore the diagnosing value of dilated cardiomyopathy (DCM). We aim to develop a novel overall normalized sum weight of multiple-model MLs to assess the diagnosing value in DCM.

Methods

Gene expression data were selected from previously published databases (six sets of eligible microarrays, 386 samples) with eligible criteria. Two sets of microarrays were used as training; the others were studied in the testing sets (ratio 5:1). Totally, we identified 20 differently expressed genes (DEGs) between DCM and control individuals (7 upregulated and 13 down-regulated).

Results

We developed six classification ML methods to identify potential candidate genes based on their overall weights. Three genes, serine proteinase inhibitor A3 (SERPINA3), frizzled-related proteins (FRPs) 3 (FRZB), and ficolin 3 (FCN3) were finally identified as the receiver operating characteristic (ROC). Interestingly, we found all three genes correlated considerably with plasma cells. Importantly, not only in training sets but also testing sets, the areas under the curve (AUCs) for SERPINA3, FRZB, and FCN3 were greater than 0.88. The ROC of SERPINA3 was significantly high (0.940 in training and 0.918 in testing sets), indicating it is a potentially functional gene in DCM. Especially, the plasma levels in DCM patients of SERPINA3, FCN, and FRZB were significant compared with healthy control.

Discussion

SERPINA3, FRZB, and FCN3 might be potential diagnosis targets for DCM, Further verification work could be implemented.

Early Detection of Pancreatic Cancers Using Liquid Biopsies and Hierarchical Decision Structure

OBJECTIVE

Pancreatic cancer (PC) is a silent killer, because its detection is difficult and to date no effective treatment has been developed. In the US, the current 5-year survival rate of 11%. Therefore, PC has to be detected as early as possible.

METHODS AND PROCEDURES

In this work, we have combined the use of ultrasensitive nanobiosensors for protease/arginase detection with information fusion based hierarchical decision structure to detect PC at the localized stage by means of a simple Liquid Biopsy. The problem of early-stage detection of pancreatic cancer is modelled as a multi-class classification problem. We propose a Hard Hierarchical Decision Structure (HDS) along with appropriate feature engineering steps to improve the performance of conventional multi-class classification approaches. Further, a Soft Hierarchical Decision Structure (SDS) is developed to additionally provide confidences of predicted labels in the form of class probability values. These frameworks overcome the limitations of existing research studies that employ simple biostatistical tools and do not effectively exploit the information provided by ultrasensitive protease/arginase analyses.

RESULTS

The experimental results demonstrate that an overall mean classification accuracy of around 92% is obtained using the proposed approach, as opposed to 75% with conventional multi-class classification approaches. This illustrates that the proposed HDS framework outperforms traditional classification techniques for early-stage PC detection.

CONCLUSION

Although this study is only based on 31 pancreatic cancer patients and a healthy control group of 48 human subjects, it has enabled combining Liquid Biopsies and Machine Learning methodologies to reach the goal of earliest PC detection. The provision of both decision labels (via HDS) as well as class probabilities (via SDS) helps clinicians identify instances where statistical model-based predictions lack confidence. This further aids in determining if more tests are required for better diagnosis. Such a strategy makes the output of our decision model more interpretable and can assist with the diagnostic procedure.

CLINICAL IMPACT

With further validation, the proposed framework can be employed as a decision support tool for the clinicians to help in detection of pancreatic cancer at early stages.

Identification and prognostic analysis of biomarkers to predict the progression of pancreatic cancer patients

BACKGROUND

Pancreatic cancer (PC) is a malignancy with a poor prognosis and high mortality. Surgical resection is the only "curative" treatment. However, only a minority of patients with PC can obtain surgery. Improving the overall survival (OS) rate of patients with PC is still a major challenge. Molecular biomarkers are a significant approach for diagnostic and predictive use in PCs. Several prediction models have been developed for patients newly diagnosed with PC that is operable or patients with advanced and metastatic PC; however, these models require further validation. Therefore, precise biomarkers are urgently required to increase the efficiency of predicting a disease-free survival (DFS), OS, and sensitivity to immunotherapy in PC patients and to improve the prognosis of PC.

METHODS

In the present study, we first evaluated the highly and selectively expressed targets in PC, using the GeoMxTM Digital Spatial Profiler (DSP) and then, we analyzed the roles of these targets in PCs using TCGA database.

RESULTS

LAMB3, FN1, KRT17, KRT19, and ANXA1 were defined as the top five upregulated targets in PC compared with paracancer. The TCGA database results confirmed the expression pattern of LAMB3, FN1, KRT17, KRT19, and ANXA1 in PCs. Significantly, LAMB3, FN1, KRT19, and ANXA1 but not KRT17 can be considered as biomarkers for survival analysis, univariate and multivariate Cox proportional hazards model, and risk model analysis. Furthermore, in combination, LAMB3, FN1, KRT19, and ANXA1 predict the DFS and, in combination, LAMB3, KRT19, and ANXA1 predict the OS. Immunotherapy is significant for PCs that are inoperable. The immune checkpoint blockade (ICB) analysis indicated that higher expressions of FN1 or ANXA1 are correlated with lower ICB response. In contrast, there are no significant differences in the ICB response between high and low expression of LAMB3 and KRT19.

CONCLUSIONS

In conclusion, LAMB3, FN1, KRT19, and ANXA1 are good predictors of PC prognosis. Furthermore, FN1 and ANXA1 can be predictors of immunotherapy in PCs.

Identification of Key mRNAs as Prediction Models for Early Metastasis of Pancreatic Cancer Based on LASSO

Pancreatic cancer is a highly malignant and metastatic tumor of the digestive system. Even after surgical removal of the tumor, most patients are still at risk of metastasis. Therefore, screening for metastatic biomarkers can identify precise therapeutic intervention targets. In this study, we analyzed 96 pancreatic cancer samples from The Cancer Genome Atlas (TCGA) without metastasis or with metastasis after R0 resection. We also retrieved data from metastatic pancreatic cancer cell lines from Gene Expression Omnibus (GEO), as well as collected sequencing data from our own cell lines, BxPC-3 and BxPC-3-M8. Finally, we analyzed the expression of metastasis-related genes in different datasets by the Limma and edgeR packages in R software, and enrichment analysis of differential gene expression was used to gain insight into the mechanism of pancreatic cancer metastasis. Our analysis identified six genes as risk factors for predicting metastatic status by LASSO regression, including zinc finger BED-Type Containing 2 (ZBED2), S100 calcium-binding protein A2 (S100A2), Jagged canonical Notch ligand 1 (JAG1), laminin subunit gamma 2 (LAMC2), transglutaminase 2 (TGM2), and the transcription factor hepatic leukemia factor (HLF). We used these six EMT-related genes to construct a risk-scoring model. The receiver operating characteristic (ROC) curve showed that the risk score could better predict the risk of metastasis. Univariate and multivariate Cox regression analyses revealed that the risk score was also an important predictor of pancreatic cancer. In conclusion, 6-mRNA expression is a potentially valuable method for predicting pancreatic cancer metastasis, assessing clinical outcomes, and facilitating future personalized treatment for patients with ductal adenocarcinoma of the pancreas (PDAC).

Quantitative evaluation of colon perfusion after high versus low ligation in rectal surgery by indocyanine green: a pilot study

BACKGROUND

In the field of rectal cancer surgery, there remains ongoing debate on the merits of high ligation (HL) and low ligation (LL) of the inferior mesenteric artery (IMA) in terms of perfusion and anastomosis leakage. Recently, infrared fluorescence of indocyanine green (ICG) imaging has been used to evaluate perfusion status during colorectal surgery.

OBJECTIVE

The purpose of this study is to compare the changes in perfusion status between HL and LL through quantitative evaluation of ICG.

METHODS

Patients with rectosigmoid or rectal cancer were randomized into a high or LL group. ICG was injected before and after IMA ligation, and region of interest (ROI) values were measured by an image analysis program (HSL video©).

RESULTS

From February to July 2020, 22 patients were enrolled, and 11 patients were assigned to each group. Basic demographics were similar between the two groups, except for albumin level and cardiac ejection fraction. There were no significant differences in F_max between the two groups, but T_max was significantly higher and Slope_max was significantly lower in the HL group than in the LL group. Anastomosis leakage was significantly associated with neoadjuvant chemoradiation and F_max.

CONCLUSION

After IMA ligation, T_max increased and Slope_max decreased significantly in the HL group. However, the intensity of perfusion status (F_max) did not change according to the level of IMA ligation.

NMFNA: A Non-negative Matrix Factorization Network Analysis Method for Identifying Modules and Characteristic Genes of Pancreatic Cancer

Pancreatic cancer (PC) is a highly fatal disease, yet its causes remain unclear. Comprehensive analysis of different types of PC genetic data plays a crucial role in understanding its pathogenic mechanisms. Currently, non-negative matrix factorization (NMF)-based methods are widely used for genetic data analysis. Nevertheless, it is a challenge for them to integrate and decompose different types of genetic data simultaneously. In this paper, a non-NMF network analysis method, NMFNA, is proposed, which introduces a graph-regularized constraint to the NMF, for identifying modules and characteristic genes from two-type PC data of methylation (ME) and copy number variation (CNV). Firstly, three PC networks, i.e., ME network, CNV network, and ME-CNV network, are constructed using the Pearson correlation coefficient (PCC). Then, modules are detected from these three PC networks effectively due to the introduced graph-regularized constraint, which is the highlight of the NMFNA. Finally, both gene ontology (GO) and pathway enrichment analyses are performed, and characteristic genes are detected by the multimeasure score, to deeply understand biological functions of PC core modules. Experimental results demonstrated that the NMFNA facilitates the integration and decomposition of two types of PC data simultaneously and can further serve as an alternative method for detecting modules and characteristic genes from multiple genetic data of complex diseases.

ITGA2, LAMB3, and LAMC2 may be the potential therapeutic targets in pancreatic ductal adenocarcinoma: an integrated bioinformatics analysis

Pancreatic ductal adenocarcinoma (PDAC) is the most common form of pancreatic cancer with an abysmal prognosis rate over the last few decades. Early diagnosis and prevention could effectively combat this malignancy. Therefore, it is crucial to discover potential biomarkers to identify asymptomatic premalignant or early malignant tumors of PDAC. Gene expression analysis is a powerful technique to identify candidate biomarkers involved in disease progression. In the present study, five independent gene expression datasets, including 321 PDAC tissues and 208 adjacent non-cancerous tissue samples, were subjected to statistical and bioinformatics analysis. A total of 20 differentially expressed genes (DEGs) were identified in PDAC tissues compared to non-cancerous tissue samples. Gene ontology and pathway enrichment analysis showed that DEGs were mainly enriched in extracellular matrix (ECM), cell adhesion, ECM-receptor interaction, and focal adhesion signaling. The protein-protein interaction network was constructed, and the hub genes were evaluated. Collagen type XII alpha 1 chain (COL12A1), fibronectin 1 (FN1), integrin subunit alpha 2 (ITGA2), laminin subunit beta 3 (LAMB3), laminin subunit gamma 2 (LAMC2), thrombospondin 2 (THBS2), and versican (VCAN) were identified as hub genes. The correlation analysis revealed that identified hub genes were significantly interconnected. Wherein COL12A1, FN1, ITGA2, LAMB3, LAMC2, and THBS2 were significantly associated with PDAC pathological stages. The Kaplan-Meier survival plots revealed that ITGA2, LAMB3, and LAMC2 expression were inversely correlated with a prolonged patient survival period. Furthermore, the Human Protein Atlas database was used to validate the expression and cellular origins of hub genes encoded proteins. The protein expression of hub genes was higher in pancreatic cancer tissue than in normal pancreatic tissue samples, wherein ITGA2, LAMB3, and LAMC2 were exclusively expressed in pancreatic cancer cells. Pancreatic cancer cell-specific expression of these three proteins may play pleiotropic roles in cancer progression. Our results collectively suggest that ITGA2, LAMB3, and LAMC2 could provide deep insights into pancreatic carcinogenesis molecular mechanisms and provide attractive therapeutic targets.

An Integrated Data Analysis of mRNA, miRNA and Signaling Pathways in Pancreatic Cancer

Although many genes and miRNAs have been reported for various cancers, pancreatic cancer's specific genes or miRNAs have not been studied precisely yet. Therefore, we have analyzed the gene and miRNA expression profile of pancreatic cancer data in the gene expression omnibus (GEO) database. The microarray-derived miRNAs and mRNAs were annotated by gene ontology (GO) and signaling pathway analysis. We also recognized mRNAs that were targeted by miRNA through the mirDIP database. An integrated analysis of the microarray revealed that only 6 out of 43 common miRNAs had significant differences in their expression profiles between the tumor and normal groups (P value < 0.05 and |log Fold Changes (logFC)|> 1). The hsa-miR-210 had upregulation, whereas hsa-miR-375, hsa-miR-216a, hsa-miR-217, hsa-miR-216b and hsa-miR-634 had downregulation in pancreatic cancer (PC). The analysis results also revealed 109 common mRNAs by microarray and mirDIP 4.1 databases. Pathway analysis showed that amoebiasis, axon guidance, PI3K-Akt signaling pathway, absorption and focal adhesion, adherens junction, platelet activation, protein digestion, human papillomavirus infection, extracellular matrix (ECM) receptor interaction, and riboflavin metabolism played important roles in pancreatic cancer. GO analysis revealed the significant enrichment in the three terms of biological process, cellular component, and molecular function, which were identified as the most important processes associated strongly with pancreatic cancer. In conclusion, DTL, CDH11, COL5A1, ITGA2, KIF14, SMC4, VCAN, hsa-mir-210, hsa-mir-217, hsa-mir-216a, hsa-mir-216b, hsa-mir-375 and hsa-mir-634 can be reported as the novel diagnostic or even therapeutic markers for the future studies. Also, the hsa-mir-107 and hsa-mir-125a-5p with COL5A1, CDH11 and TGFBR1 genes can be introduced as major miRNA and genes on the miRNA-drug-mRNA network. The new regulatory network created in our study could give a deeper knowledge of the pancreatic cancer.

Identification of the hub genes in gastric cancer through weighted gene co-expression network analysis

Background

Gastric cancer is one of the most lethal tumors and is characterized by poor prognosis and lack of effective diagnostic or therapeutic biomarkers. The aim of this study was to find hub genes serving as biomarkers in gastric cancer diagnosis and therapy.

Methods

GSE66229 from Gene Expression Omnibus (GEO) was used as training set. Genes bearing the top 25% standard deviations among all the samples in training set were performed to systematic weighted gene co-expression network analysis (WGCNA) to find candidate genes. Then, hub genes were further screened by using the “least absolute shrinkage and selection operator” (LASSO) logistic regression. Finally, hub genes were validated in the GSE54129 dataset from GEO by supervised learning method artificial neural network (ANN) algorithm.

Results

Twelve modules with strong preservation were identified by using WGCNA methods in training set. Of which, five modules significantly related to gastric cancer were selected as clinically significant modules, and 713 candidate genes were identified from these five modules. Then, ADIPOQ, ARHGAP39, ATAD3A, C1orf95, CWH43, GRIK3, INHBA, RDH12, SCNN1G, SIGLEC11 and LYVE1 were screened as the hub genes. These hub genes successfully differentiated the tumor samples from the healthy tissues in an independent testing set through artificial neural network algorithm with the area under the receiver operating characteristic curve at 0.946.

Conclusions

These hub genes bearing diagnostic and therapeutic values, and our results may provide a novel prospect for the diagnosis and treatment of gastric cancer in the future.

Integration of gene expression data identifies key genes and pathways in colorectal cancer

Colorectal cancer (CRC) is one of the most common malignant tumor and prevalent cause of cancer-related death worldwide. In this study, we analyzed the gene expression profiles of patients with CRC with the aim of better understanding the molecular mechanism and key genes in CRC. Four gene expression profiles including, GSE9348, GSE41328, GSE41657, and GSE113513 were downloaded from GEO database. The data were processed using R programming language, in which 319 common differentially expressed genes including 94 up-regulated and 225 down-regulated were identified. The gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG) enrichment analyses were conducted to find the most significant enriched pathways in CRC. Based on the GO and KEGG pathway analysis, the most important dysregulated pathways were regulation of cell proliferation, biocarbonate transport, Wnt, and IL-17 signaling pathways, and nitrogen metabolism. The protein-protein interaction (PPI) network of the DEGs was constructed using Cytoscape software and hub genes including MYC, CXCL1, CD44, MMP1, and CXCL12 were identified as the most critical hub genes. The present study enhances our understanding of the molecular mechanisms of the CRC, which might potentially be applied in the treatment strategies of CRC as molecular targets and diagnostic biomarkers.

Bioinformatic analysis of potential hub genes in gastric adenocarcinoma

Gastric adenocarcinoma is the most common histologic type of gastric cancer; however, the pathogenic mechanisms remain unclear. To improve mechanistic understanding and identify new treatment targets or diagnostic biomarkers, we used bioinformatic tools to predict the hub genes related to the process of gastric adenocarcinoma development from public datasets, and explored their prognostic significance. We screened differentially expressed genes between gastric adenocarcinoma and normal gastric tissues in Gene Expression Omnibus datasets (GSE79973, GSE118916, and GSE29998) using the GEO2R tool, and their functions were annotated with Gene Ontology and Kyoto Encyclopedia of Genes and Genomes signaling pathway enrichment analyses in the DAVID database. Hub genes were identified based on the protein-protein network constructed in the STRING database with Cytoscape software. A total of 10 hub genes were selected for further analysis, and their expression patterns in gastric adenocarcinoma patients were investigated using the Oncomine GEPIA database. The expression levels of ATP4A, CA9, FGA, ALDH1A1, and GHRL were reduced, whereas those of TIMP1, SPP1, CXCL8, THY1, and COL1A1 were increased in gastric adenocarcinoma. The Kaplan-Meier online plotter tool showed associations of all hub genes except for CA9 with prognosis in gastric adenocarcinoma patients; CXCL8 and ALDH1A1 were positively correlated with survival, and the other genes were negatively correlated with survival. These 10 hub genes may be involved in important processes in gastric adenocarcinoma development, providing new directions for research to clarify the role of these genes and offer insight for improved treatment.

Identification of potential crucial genes associated with the pathogenesis and prognosis of pancreatic adenocarcinoma

Pancreatic adenocarcinoma (PAAD) is a type of malignant tumor with the highest mortality rate among all neoplasms worldwide, and its exact pathogenesis is still poorly understood. Timely diagnosis and treatment are of great importance in order to decrease the mortality rate of PAAD. Therefore, identifying new biomarkers for diagnosis and prognosis is essential to enable early detection of PAAD and to improve the overall survival (OS) rate. In order to screen and integrate differentially expressed genes (DEGs) between PAAD and normal tissues, a total of seven datasets were downloaded from the Gene Expression Omnibus database and the ‘limma’ and ‘robustrankggreg’ packages in R software were used. The Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis of the DEGs was performed using the Database for Annotation, Visualization and Integrated Discovery website, and the protein-protein interaction network analysis was performed using the Search Tool for the Retrieval of Interacting Genes/Proteins database. A gene prognostic signature was constructed using the Cox regression model. A total of 10 genes (CDK1, CCNB1, CDC20, ASPM, UBE2C, TPX2, TOP2A, NUSAP1, KIF20A and DLGAP5) that may be associated with pancreatic adenocarcinoma were identified. According to the differentially expressed genes in The Cancer Genome Atlas, the present study set up four prognostic signatures (matrix metalloproteinase 12, sodium voltage-gated channel α subunit 11, tetraspanin 1 and SH3 domain and tetratricopeptide repeats-containing 2), which effectively predicted OS. The hub genes that were highly associated with the occurrence, development and prognosis of PAAD were identified, which may be helpful to further understand the molecular basis of pancreatic cancer and guide the synthesis of drugs for PPAD.

miR-216a-mediated upregulation of TSPAN1 contributes to pancreatic cancer progression via transcriptional regulation of ITGA2

Pancreatic cancer (PC) is recognized as the most aggressive and deadliest malignancy because it has the highest mortality of all cancers in humans. Mutations in multiple tumor suppressors and oncogenes have been documented to be involved in pancreatic cancer progression and metastasis. The upregulation of tetraspanin 1 (TSPAN1), a transmembrane protein, has been reportedly observed in many human cancers. However, the role of TSPAN1 and its underlying molecular mechanisms in PC progression have not been fully elucidated. In this study, we validated the oncogenic role of TSPAN1 in PC, showing that TSPAN1 reinforces cell proliferation, migration, invasion and tumorigenesis. To investigate the upregulation of TSPAN1 in PC, we showed that miR-216a is the upstream negative regulator of TSPAN1 via direct binding to the TSPAN1 3'-untranslated region. Through RNA-Seq analysis, we for the first time revealed that TSPAN1 expression transcriptionally regulates ITGA2, which is involved in the actin cytoskeleton pathway. The stimulated cell proliferation and invasion initiated by TSPAN1 overexpression could be abolished by knockdown of ITGA2 in PC cells. Furthermore, TSPAN1 epigenetically regulates the expression of ITGA2 by modulating the levels of TET2 DNMT3B and DNMT1, resulting in hypomethylation of the CpG island of the ITGA2 promoter. In conclusion, the newly identified miR-216a/TSPAN1/ITGA2 axis is involved in the modulation of PC progression and represents a novel therapeutic strategy for future pancreatic cancer treatment.

Development of a four-gene prognostic model for pancreatic cancer based on transcriptome dysregulation

We systematically developed a prognostic model for pancreatic cancer that was compatible across different transcriptomic platforms and patient cohorts. After performing quality control measures, we used seven microarray datasets and two RNA sequencing datasets to identify consistently dysregulated genes in pancreatic cancer patients. Weighted gene co-expression network analysis was performed to explore the associations between gene expression patterns and clinical features. The least absolute shrinkage and selection operator (LASSO) and Cox regression were used to construct a prognostic model. We tested the predictive power of the model by determining the area under the curve of the risk score for time-dependent survival. Most of the differentially expressed genes in pancreatic cancer were enriched in functions pertaining to the tumor immune microenvironment. The transcriptome profiles were found to be associated with overall survival, and four genes were identified as independent prognostic factors. A prognostic risk score was then proposed, which displayed moderate accuracy in the training and self-validation cohorts. Furthermore, patients in two independent microarray cohorts were successfully stratified into high- and low-risk prognostic groups. Thus, we constructed a reliable prognostic model for pancreatic cancer, which should be beneficial for clinical therapeutic decision-making.

Integrated transcriptomic analysis reveals hub genes involved in diagnosis and prognosis of pancreatic cancer

BACKGROUND

The hunt for the molecular markers with specificity and sensitivity has been a hot area for the tumor treatment. Due to the poor diagnosis and prognosis of pancreatic cancer (PC), the excision rate is often low, which makes it more urgent to find the ideal tumor markers.

METHODS

Robust Rank Aggreg (RRA) methods was firstly applied to identify the differentially expressed genes (DEGs) between PC tissues and normal tissues from GSE28735, GSE15471, GSE16515, and GSE101448. Among these DEGs, the highly correlated genes were clustered using WGCNA analysis. The co-expression networks and molecular complex detection (MCODE) Cytoscape app were then performed to find the sub-clusters and confirm 35 candidate genes. For these genes, least absolute shrinkage and selection operator (lasso) regression model was applied and validated to build a diagnostic risk score model. Cox proportional hazard regression analysis was used and validated to build a prognostic model.

RESULTS

Based on integrated transcriptomic analysis, we identified a 19 gene module (SYCN, PNLIPRP1, CAP2, GNMT, MAT1A, ABAT, GPT2, ADHFE1, PHGDH, PSAT1, ERP27, PDIA2, MT1H, COMP, COL5A2, FN1, COL1A2, FAP and POSTN) as a specific predictive signature for the diagnosis of PC. Based on the two consideration, accuracy and feasibility, we simplified the diagnostic risk model as a four-gene model: 0.3034*log2(MAT1A)-0.1526*log2(MT1H) + 0.4645*log2(FN1) -0.2244*log2(FAP), log2(gene count). Besides, a four-hub gene module was also identified as prognostic model = - 1.400*log2(CEL) + 1.321*log2(CPA1) + 0.454*log2(POSTN) + 1.011*log2(PM20D1), log2(gene count).

CONCLUSION

Integrated transcriptomic analysis identifies two four-hub gene modules as specific predictive signatures for the diagnosis and prognosis of PC, which may bring new sight for the clinical practice of PC.

Identification of key biomarkers and potential molecular mechanisms in lung cancer by bioinformatics analysis

Lung cancer is one of the most widespread neoplasms worldwide. To identify the key biomarkers in its carcinogenesis and development, the mRNA microarray datasets GSE102287, GSE89047, GSE67061 and GSE74706 were obtained from the Gene Expression Omnibus database. GEO2R was used to identify the differentially expressed genes (DEGs) in lung cancer. The Database for Annotation, Visualization and Integrated Discovery was used to analyze the functions and pathways of the DEGs, while the Search Tool for the Retrieval of Interacting Genes/Proteins and Cytoscape were used to obtain the protein-protein interaction (PPI) network. Kaplan Meier curves were used to analyze the effect of the hub genes on overall survival (OS). Module analysis was completed using Molecular Complex Detection in Cytoscape, and one co-expression network of these significant genes was obtained with cBioPortal. A total of 552 DEGs were identified among the four microarray datasets, which were mainly enriched in 'cell proliferation', 'cell growth', 'cell division', 'angiogenesis' and 'mitotic nuclear division'. A PPI network, composed of 44 nodes and 886 edges, was constructed, and its significant module had 16 hub genes in the whole network: Opa interacting protein 5, exonuclease 1, PCNA clamp-associated factor, checkpoint kinase 1, hyaluronan-mediated motility receptor, maternal embryonic leucine zipper kinase, non-SMC condensin I complex subunit G, centromere protein F, BUB1 mitotic checkpoint serine/threonine kinase, cyclin A2, thyroid hormone receptor interactor 13, TPX2 microtubule nucleation factor, nucleolar and spindle associated protein 1, kinesin family member 20A, aurora kinase A and centrosomal protein 55. Survival analysis of these hub genes revealed that they were markedly associated with poor OS in patients with lung cancer. In summary, the hub genes and DEGs delineated in the research may aid the identification of potential targets for diagnostic and therapeutic strategies in lung cancer.