Regulatory network reconstruction of five essential microRNAs for survival analysis in breast cancer by integrating miRNA and mRNA expression datasets

• Published in: Functional & integrative genomics 2019-07, Vol.19 (4), p.645-658
• Main Authors: He, Kan, Li, Wen-Xing, Guan, Daogang, Gong, Mengting, Ye, Shoudong, Fang, Zekun, Huang, Jing-Fei, Lu, Aiping
• Format: Article
• Language: English
• Subjects: Animal Genetics and Genomics; Biochemistry; Bioinformatics; Biomedical and Life Sciences; Breast cancer; Cell Biology; Gene expression; Gene regulation; Genomics; Histocompatibility antigen HLA; Interferon; Life Sciences; Microbial Genetics and Genomics; MicroRNAs; miRNA; Original Article; Plant Genetics and Genomics; Precision medicine; Stat1 protein; Survival; Survival analysis; Transcription
• ISSN: 1438-793X; 1438-7948
• DOI: 10.1007/s10142-019-00670-7

Although many of the genetic loci associated with breast cancer risk have been reported, there is a lack of systematic analysis of regulatory networks composed of different miRNAs and mRNAs on survival analysis in breast cancer. To reconstruct the microRNAs-genes regulatory network in breast cancer, we employed the expression data from The Cancer Genome Atlas (TCGA) related to five essential miRNAs including miR-21, miR-22, miR-210, miR-221, and miR-222, and their associated functional genomics data from the GEO database. Then, we performed an integration analysis to identify the essential target factors and interactions for the next survival analysis in breast cancer. Based on the results of our integrated analysis, we have identified significant common regulatory signatures including differentially expressed genes, enriched pathways, and transcriptional regulation such as interferon regulatory factors (IRFs) and signal transducer and activator of transcription 1 (STAT1). Finally, a reconstructed regulatory network of five miRNAs and 34 target factors was established and then applied to survival analysis in breast cancer. When we used expression data for individual miRNAs, only miR-21 and miR-22 were significantly associated with a survival change. However, we identified 45 significant miRNA-gene pairs that predict overall survival in breast cancer out of 170 one-on-one interactions in our reconstructed network covering all of five miRNAs, and several essential factors such as PSMB9, HLA-C, RARRES3, UBE2L6, and NMI. In our study, we reconstructed regulatory network of five essential microRNAs for survival analysis in breast cancer by integrating miRNA and mRNA expression datasets. These results may provide new insights into regulatory network-based precision medicine for breast cancer.