As novel master regulators of initiation and progression, long non-coding RNAs (lncRNAs) can response to therapy in a wide variety of malignances. This study aimed to explore the regulatory mechanisms in non-small cell lung cancer (NSCLC) by constructing lncRNA-miRNA-mRNA networks. The whole transcriptome sequence was performed in five NSCLC tissues and their corresponding paired adjacent normal tissues. MicroRNAs (miRNAs) and mRNAs expression profiles were obtained from TCGA database. “EdgeR” R package was used for gene expression comparisons and the genes with | log₂FC |> 2 and false discovery rate (FDR) adjusted P<0.05 were considered significant. Totally, 559 differentially expressed lncRNAs (DELs) (235 up-regulated, 324 down-regulated) were identified via whole transcriptome sequence analysis. Subsequently, 14 up-regulated lncRNAs were obtained in the intersection of our RNA-seq data and TCGA dataset. The dysregulations of the selected lncRNAs were evaluated through GEPIA. Three candidate lncRNAs (LINC01426, TYMSOS, VPS9D1-AS1) were finally selected for further study. Through bioinformatics prediction, with the three lnRNAs and their associated miRNAs (n=14) and mRNAs (n=218), a lncRNA-miRNA-mRNA network was constructed. Furthermore, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway associated with the targeted genes were proceeded. In addition, the protein–protein interaction (PPI) network construction was performed by STRING and lncRNA-miRNA-mRNA regulatory network was visualized via Cytoscape. The top ten significant hub-genes in the PPI network are identified based on their node degrees. Moreover, the lncRNA-miRNA and miRNA-mRNA interactions were predicted using the StarBase and survival analyses were performed in Kaplan-Meier plotter website. We concluded that TYMSOS/hsa-miR-101-3p/CEP55 and TYMSOS/hsa-miR-195-5p/CHEK1 ceRNA regulatory aixs may play vital roles in the occurrence and development of NSCLC, and may provide novel therapeutic targets in the future.