LAUSR

Our study aimed to identify prognosis related epigenetic interactions of DNA methylation-miRNA-gene in lung adenocarcinoma. The RNA-seq, DNA methylation, and miRNA-seq data of squamous cell cancer samples were downloaded from TCGA. The DNA methylation-miRNA-gene interactions were collected via Illumina methylation platform and miRTarBase database. Linear regression model was utilized for the identification of epigenetic interactions. The epigenetic interactions related to prognosis were selected via Kaplan-Meier analysis. Genes in the interactions were used for pathway enrichment. Differentially expressed genes (DEGs) between high methylation level / high miRNA expression level (H/H) and low methylation level / low miRNA expression level (L/L) samples were screened. The correlations of epigenetic interactions with clinical features were also explored. Total of 454 lung adenocarcinoma patient samples were collected. The 1063 interactions were comprised of 1083 DNA methylation probes, 271 miRNAs and 528 genes, including cg14146378-hsa-mir-205-ARID1B, cg15375596-has-miR-1275-IGF1R, cg26691953-hsa-mir-195-CCNT1, etc. A total of 95 epigenetic interactions were significantly associated with prognosis. Among all the identified DEGs, low-expressed RASSF4, ZNF704, TFDP1, PLXNB2, TMC04, ZNF878, ARIDIB and high-expressed ZNF704, ZNF451, THOP1, IGF1R were related with poor prognosis; while low-expressed LDHB, ARID2, PRKCSH, HDAC4, NIPA1, RABAC1, TRIM28 and high-expressed FAM160B1, DNAAF3, CCNT1, ADAP1, ZFPM1, CCL11 were related with good prognosis. Fifteen epigenetic interactions were significantly related with clinical features. Gene expression and N-glycan trimming in the ER and Calnexin/Calreticulin cycle were two significant enriched pathways. Interactions of cg14146378-hsa-mir-205-ARID1B and cg15375596-has-miR-1275-IGF1R may be used as the prognosis indicators in lung adenocarcinoma.