Abstract / Summary
Abstract Diabetic nephropathy (DN) is a leading cause of end-stage renal disease, driven by metabolic dysregulation, chronic inflammation, and fibrotic remodeling. Elucidating the molecular networks underlying DN progression is essential for prioritizing therapeutic strategies. In this study, we employed an integrated in silico approach combining transcriptomic analysis, molecular network reconstruction, and structure-based virtual screening to characterize candidate disease-associated gene modules and prioritize multi-target natural compounds. RNA sequencing data from human kidney biopsies were analyzed. To prioritize hyperactive pathological cascades for pharmacological inhibition, downstream network clustering and virtual screening focused exclusively on upregulated differentially expressed genes (DEGs). Intersection with curated metabolic drug targets identified disease-associated genes. Protein–protein interaction and regulatory network analyses revealed subnetworks enriched for immune activation, inflammatory signaling, and extracellular matrix organization, highlighting highly connected nodes including CD4, CCR7, CSF1R, FBN1, ITGA4, LCK, SYK , and TLR6. Candidate regulatory elements, including transcription factors (NFKB1, RELA, SP1, STAT3 ) and miRNAs, were identified as potential modulators of these transcriptional programs. To bridge these network findings to therapeutic discovery, natural compounds from the LOTUS database were screened against 29 prioritized DN-associated protein targets using molecular docking. Network analysis prioritized ergostane-type steroids as candidate multi-target antagonists exhibiting favorable theoretical binding affinities. Collectively, these findings delineate key molecular networks in DN and prioritize mechanistically grounded therapeutic candidates, generating robust, data-driven hypotheses for future experimental validation.