Background: Pulmonary tuberculosis (PTB) development is linked to dysregulated immune responses, with the innate immune cell barrier playing a crucial role in both disease initiation and progression. Methods: The acquisition of transcriptomic datasets pertaining to PTB was conducted using the GEO database. Modules associated with the innate immune cell barrier were identified using weighted gene co-expression network analysis. The intersection of the differentially expressed genes and the module genes yielded a set of overlapping genes. Biological pathways were characterized and candidate genes were screened through functional enrichment analyses and protein-protein interaction network modeling. Key genes were identified through the application of machine-learning algorithms and further confirmed via qPCR using clinical samples. Diagnostic models were developed and independently evaluated. Multiple analytical approaches were employed to explore the biological mechanisms and therapeutic potential. These included gene set enrichment analysis, immune cell infiltration profiling, regulatory network analysis, and molecular docking. Results: A total of 360 overlapping genes were identified, enriched in immune activation and inflammatory pathways. Nineteen candidates were screened, and five key genes-OAS1, STAT2, IFITM1, CCR7, and CD19-were prioritized through machine learning. qPCR validation confirmed consistent expression trends for all genes except CD19. These genes were enriched in lysosomal and ribosomal pathways, with aberrant neutrophil and monocyte infiltration strongly correlating with gene expression. Molecular docking identified peginterferon-alfa-2a and isotretinoin as potential therapeutic agents. Conclusion: This study identified five key genes and two potential drug candidates associated with innate immune dysregulation in PTB, providing new insights into disease mechanisms and supporting their potential value in PTB diagnosis and therapy.
