This study aims to analyze the regulatory non-coding RNAs in the pathological process of tuberculosis (TB), and identify novel diagnostic biomarkers. A longitudinal study was conducted in 5 newly diagnosed pulmonary tuberculosis patients, peripheral blood samples were collected before and after anti-TB treatment for 6 months, separately. After whole transcriptome sequencing, the differentially expressed RNAs (DE RNAs) were filtrated with |log2 (fold change) | > log2(1.5) and P value < .05 as screening criteria. Then functional annotation was actualized by gene ontology enrichment analysis, and enrichment pathway analysis was conducted by Kyoto Encyclopedia of Genes and Genomes database. And finally, the competitive endogenous RNA (ceRNA) regulatory network was established according to the interaction of ceRNA pairs and miRNA-mRNA pairs. Five young women were recruited and completed this study. Based on the differential expression analysis, a total of 1469 mRNAs, 996 long non-coding RNAs, 468 circular RNAs, and 86 miRNAs were filtrated as DE RNAs. Functional annotation demonstrated that those DE-mRNAs were strongly involved in the cellular process (n = 624), metabolic process (n = 513), single-organism process (n = 505), cell (n = 651), cell part (n = 650), organelle (n = 569), and binding (n = 629). Enrichment pathway analysis revealed that the differentially expressed genes were mainly enriched in HTLV-l infection, T cell receptor signaling pathway, glycosaminoglycan biosynthesis-heparan sulfate/heparin, and Hippo signaling pathway. CeRNA networks revealed that hsa-miR-17-5p, hsa-miR-106a-5p and hsa-miR-2355-5p might be regarded as potential diagnostic biomarkers for TB. Immunomodulation-related genes are differentially expressed in TB patients, and hsa-miR-106a-5p, hsa-miR-17-5p, hsa-miR-2355-5p might serve as potential diagnostic biomarkers.