Integrative microRNA and gene profiling data analysis reveals novel biomarkers and mechanisms for lung cancer

// Ling Hu 1,2,* , Junmei Ai 2,* , Hui Long 3,* , Weijun Liu 4 , Xiaomei Wang 5 , Yi Zuo 6 , Yan Li 2 , Qingming Wu 7 and Youping Deng 7,2 1 Department of Anesthesiology, Tianyou Hospital, Wuhan University of Science and Technology, Wuhan, China 2 Department of Internal Medicine and Biochemistry, Rush University Medical Center, Chicago, IL, USA 3 Department of Gastroenterology, Tianyou Hospital, Wuhan University of Science and Technology, Wuhan, China 4 Department of Orthopedics, Pu Ai Hospital, Affiliated to Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China 5 Department of Biological Science and Technology, Wuhan Bioengineering Institute, Wuhan, China 6 Department of Orthopedic, Tianyou Hospital, Wuhan University of Science and Technology, Wuhan, China 7 Medical College, Wuhan University of Science and Technology, Wuhan, China * These authors have contributed equally to this work Correspondence to: Youping Deng, email: // Qingming Wu, email: // Keywords : microRNAs, lung cancer, meta-analysis, target gene, biomarker Received : July 14, 2015 Accepted : January 13, 2016 Published : February 08, 2016 Abstract Background: Studies on the accuracy of microRNAs (miRNAs) in diagnosing non-small cell lung cancer (NSCLC) have still controversial. Therefore, we conduct to systematically identify miRNAs related to NSCLC, and their target genes expression changes using microarray data sets. Methods: We screened out five miRNAs and six genes microarray data sets that contained miRNAs and genes expression in NSCLC from Gene Expression Omnibus. Results: Our analysis results indicated that fourteen miRNAs were significantly dysregulated in NSCLC. Five of them were up-regulated (miR-9, miR-708, miR-296-3p, miR-892b, miR-140-5P) while nine were down-regulated (miR-584, miR-218, miR-30b, miR-522, miR486-5P, miR-34c-3p, miR-34b, miR-516b, miR-592). The integrating diagnosis sensitivity (SE) and specificity (SP) were 82.6% and 89.9%, respectively. We also found that 4 target genes ( p < 0.05, fold change > 2.0) were significant correlation with the 14 discovered miRNAs, and the classifiers we built from one training set predicted the validation set with higher accuracy (SE = 0.987, SP = 0.824). Conclusions: Our results demonstrate that integrating miRNAs and target genes are valuable for identifying promising biomarkers, and provided a new insight on underlying mechanism of NSCLC. Further, our well-designed validation studies surely warrant the investigation of the role of target genes related to these 14 miRNAs in the prediction and development of NSCLC.