Integrative Transcriptomic and Genetic Analysis Nominally Prioritizes SLC51A and TPMT as Candidate Diagnostic Genes for Thoracic Aortic Aneurysm

Abstract

Thoracic aortic aneurysm (TAA) is a life‑threatening disease with limited blood‑based biomarkers for early detection and risk stratification. Integrating transcriptomic and genetic data may help identify potentially relevant genes and develop suggestive diagnostic tools. Peripheral blood expression data from GSE9106 (comprising 59 TAA patients and 34 controls) were analyzed to identify differentially expressed genes (DEGs) between TAA patients and controls. Whole‑blood cis‑expression quantitative trait loci (cis‑eQTLs) from the eQTLGen consortium (31,684 individuals) were integrated with genome‑wide association study (GWAS) summary statistics (1,351 cases and 18,295 controls) using summary data–based Mendelian randomization (SMR) with Heterogeneity in Dependent Instruments (HEIDI) test to prioritize genes with suggestive causal effects. Overlapping DEGs and SMR genes were subjected to feature selection using least absolute shrinkage and selection operator (LASSO) logistic regression and random forest (RF). A multivariable logistic model based on core genes was visualized as a nomogram and evaluated by receiver operating characteristic analysis, calibration, and decision curve analysis. Single‑gene gene set enrichment analysis (GSEA) was performed to explore pathways associated with core‑gene expression. Transcriptomic analysis identified 1,108 DEGs, which, when integrated with SMR results, yielded 14 genes supported by both differential expression and genetic evidence at the nominal significance level. Forest plots demonstrated that all 14 genes had nominally significant SMR associations with TAA and passed the HEIDI test, providing limited evidence against heterogeneity, although this does not definitively rule out linkage or pleiotropy. Machine‑learning feature selection converged on two core genes, SLC51A and TPMT, which individually showed moderate diagnostic performance. A two‑gene nomogram constructed from these markers achieved good discrimination and calibration and provided net clinical benefit across a wide range of decision thresholds. Single‑gene GSEA indicated that variation in SLC51A and TPMT expression is associated with coordinated changes in RNA metabolism, macromolecular catabolism, energy utilization, and cell‑cycle–related processes. By integrating blood transcriptomics, eQTL data, and GWAS summary statistics, this study identifies SLC51A and TPMT as candidate diagnostic genes for TAA and proposes a simple two‑gene nomogram. Given the modest sample size, the lack of external validation, and the nominal nature of the genetic evidence, these findings should be considered exploratory and hypothesis‑generating rather than clinically actionable. Independent replication in larger cohorts and functional validation are required before any translational application.