BackgroundTraumatic brain injury (TBI) is more common than ever and is becoming a global public health issue. However, there are no sensitive diagnostic or prognostic biomarkers to identify TBI, which leads to long-term consequences. In this study, we aim to identify genes that contribute to brain injury and to identify potential mechanisms for its progression in the early stages.MethodFrom the Gene Expression Omnibus (GEO) database, we downloaded GSE2871’s gene expression profiles. Weighted gene coexpression network analyses (WGCNA) were conducted on differentially expressed genes (DEGs), and the DEGs were analyzed by Gene Set Enrichment Analysis(GSEA). An enrichment analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) was performed for understanding the biological functions of genes. The potential biomarkers were identified using 3 kinds of machine learning algorithms. Nomogram was constructed using the “rms” package. And the receiver operating characteristic curve (ROC) was plotted to detect and validate our prediction model sensitivity and specifificity.ResultsBetween samples with and without brain injury, 107 DEGs were identified, including 47 upregulated genes and 60 downregulated genes. On the basis of WGCNA and DEGs, 97 target genes were identified. In addition, biological function analysis indicated that target genes were primarily involved in the interaction of neuroactive ligands with receptors, taste transduction, cortisol synthesis and secretion, potassium ion transport. Based on machine learning algorithms, LOC103691092, Npw could be potentially useful biomarkers for TBI and showed good diagnostic values. Finally, a nomogram was constructed of the expression levels of these seven target genes to predict level of TBI, and the ROC showed that these genes can be used as hub genes after TBI.ConclusionLOC103691092, NPW, STK39, KCND3, APOC3, FOXE3, and CHRNB1 were identified as hub genes of TBI. These findings can provide a new direction for the diagnosis and treatment of TBI.