Background: Psoriasis is a recurrent chronic skin inflammatory disease. Its biomarkers are currently inaccurate and lack specificity. Hence, the immune infiltration mechanism of the key differential genes in psoriasis and related principles are discussed in this article. Methods: The GSE30999, GSE67853 and GSE78097 data sets were downloaded from the Gene Expression Omnibus(GEO) database, and the differentially expressed genes were screened with p＜0.05 & |Log2FC|＞1 by R language. Then, hub genes were picked out through LASSO and BORUTA analyses and verified in exterior dataset GSE83582. Further, we used ssGSEA to evaluate the effect of genes on immune infiltration. GSVA and GSEA algorithms were applied to assess associations between hub genes and different pathways for enrichment. Finally, the gene selection model of psoriasis-based immune infiltration will be examined by ROC. Results: The characteristic genes of psoriasis are determined to be TMPRSS11D, S100A12 and KYNU by the intersection of two algorithms. These three genes all have strong correlations with the content of immune cells and immune-related genes. Highly expressed TMPRSS11D, S100A12, and KYNU genes are involved in HEME_METABOLISM, XENOBIOTIC_METABOLISM, and MTORC1 pathways indicating that core genes affect the development and progression of psoriasis by regulating metabolism and T cell-related immunity. GESA showed that hub genes are linked to immune factors concerning enriched pathways. The AUC values of these three core genes are TMPRSS11D = AUC: 0.980 (0.959–1.000), S100A12= AUC: 0.982 (0.963–1.000), KYNU= AUC: 0.992 (0.982–1.000). Hub genes are actively associated with each other. Eventually, the differential gene expression of these key genes is validated by external datasets. Conclusion: Combining bioinformatics to analyze the immune infiltration mechanism of psoriasis provides a basis for discovering novel diagnostic biomarkers.