Identification of gene markers in the development of smoking-induced lung cancer.

Lung cancer is a malignant tumor with high mortality in both women and men. To study the mechanisms of smoking-induced lung cancer, we analyzed microarray of GSE4115. GSE4115 was downloaded from Gene Expression Omnibus including 78 and 85 bronchial epithelium tissue samples separately from smokers with and without lung cancer. Limma package in R was used to screen differentially expressed genes (DEGs). Hierarchical cluster analysis for DEGs was conducted using orange software and visualized by distance map. Using DAVID software, functional and pathway enrichment analyses separately were conducted for the DEGs. And protein-protein interaction (PPI) network was constructed using Cytoscape software. Then, the pathscores of enriched pathways were calculated. Besides, functional features were screened and optimized using the recursive feature elimination (RFE) method. Additionally, the support vector machine (SVM) method was used to train model. Total 1923 DEGs were identified between the two groups. Hierarchical cluster analysis indicated that there were differences in gene level between the two groups. And SVM analysis indicated that the five features had potential diagnostic value. Importantly, MAPK1 (degree=30), SRC (degree=29), SMAD4 (degree=23), EEF1A1 (degree=21), TRAF2 (degree=21) and PLCG1 (degree=20) had higher degrees in the PPI network of the DEGs. They might be involved in smoking-induced lung cancer by interacting with each other (e.g. MAPK1-SMAD4, SMAD4-EEF1A1 and SRC-PLCG1). MAPK1, SRC, SMAD4, EEF1A1, TRAF2 and PLCG1 might be responsible for the development of smoking-induced lung cancer.

Yang Z ，Zhuan B ，Yan Y ，Jiang S ，Wang T ... -  《-》

Construction of a 26‑feature gene support vector machine classifier for smoking and non‑smoking lung adenocarcinoma sample classification.

Yang L ，Sun L ，Wang W ，Xu H ，Li Y ，Zhao JY ，Liu DZ ，Wang F ，Zhang LY ... -  《-》

Identification of feature risk pathways of smoking-induced lung cancer based on SVM.

Chen R ，Lin J 《PLoS One》

Bioinformatics analysis of gene expression profile data to screen key genes involved in pulmonary sarcoidosis.

Sarcoidosis is a multisystemic inflammatory and granulomatous disease that occurs in almost all populations and affects multiple organs. Meanwhile, its most common manifestation is pulmonary sarcoidosis. This study aimed to identify effective biomarkers for the diagnosis and therapy of pulmonary sarcoidosis. GSE16538 was downloaded from Gene Expression Omnibus, including 6 pulmonary sarcoidosis samples and 6 normal lung samples. Then, differentially expressed genes (DEGs) were identified by limma package in R. After the expression values of the DEGs were extracted, hierarchical clustering analysis was performed for the DEGs using the pheatmap package in R. Subsequently, protein-protein interaction (PPI) pairs among the DEGs were searched by STRING or REACTOME databases, and then PPI networks were visualized by Cytoscape software. Using DAVID and KOBAS, functional and pathway enrichment analyses separately were performed for the DEGs involved in the PPI network. Total 208 DEGs were identified in pulmonary sarcoidosis samples, including 179 up-regulated genes and 29 down-regulated genes. Hierarchical clustering showed that the DEGs could clearly distinguish the pulmonary sarcoidosis samples from the normal lung samples. In the PPI network constructed by STRING database, CXCL9, STAT1, CCL5, CXCL11 and GBP1 had higher degrees and betweenness values, and could interact with each other. Functional enrichment showed that CXCL9, CXCL11 and CCL5 were enriched in immune response. Moreover, STAT1 was enriched in pathways of chemokine signaling pathway and JAK-STAT signaling pathway. CXCL9, CXCL11, STAT1, CCL5 and GBP1 might be implicated in pulmonary sarcoidosis through interacting with each other.

Li H ，Zhao X ，Wang J ，Zong M ，Yang H ... -  《-》

Screening genes associated with myocardial infarction and transverse aortic constriction using a combined analysis of miRNA and mRNA microarray.

Myocardial infarction (MI) and transverse aortic constriction (TAC) are two models of cardiac hypertrophy. To study mechanisms of MI and TAC, GSE415 and GSE14267 were downloaded from Gene Expression Omnibus. GSE415 included left ventricle (LV) and intraventricular septum samples from mice that underwent MI, TAC or sham operation. GSE14267 included normal and MI samples from non-transgenic mice. Differentially expressed genes (DEGs) and microRNAs (miRNAs) were screened using limma package. Functional enrichment analysis was performed for DEGs using DAVID. Common DEGs of different groups were conducted for protein-protein interaction (PPI) analysis using STRING and visualized in PPI network by Cytoscape. Furthermore, targets were predicted for differentially expressed miRNAs using TarMir database. Totally, 277 DEGs, 31 common DEGs (e.g. SFRP2), 6 differentially expressed miRNAs (e.g. mmu-miR-448) and 1 miRNA-mRNA pair (mmu-miR-448→SIM2) were screened out. DEGs were significantly enriched in biological processes related to muscle development and ion transportation. In the PPI network for common DEGs, LOX (degree=7), POSTN (degree=5), SPARC (degree=4) and TIMP1 (degree=3) were with higher degrees. In addition, they might function by interacting with each other (e.g. LOX-TIMP1, LOX-POSTN, SPARC-TIMP1 and SPARC-POSTN). In conclusion, LOX, POSTN, SPARC, TIMP1 and SFRP2 might affect MI and TAC.3.

Wang M ，Luo J ，Wan L ，Hu T ，Li S ，Zhan C ... -  《-》