Back

Integrated Bioinformatics and Pharmacogenomic Profiling of a Gene Panel in Diabetes Mellitus Treatment Response

de Oliveira Andrade, L. J.; Matos de Oliveira, G. C.; Leahy Neto, J. M.; de Mattos Salles, O. J.; Vinhaes Bittencourt, A. M.; Matos de Oliveira, L.

2025-10-07 endocrinology
10.1101/2025.10.04.25337324 medRxiv
Show abstract

BackgroundPharmacogenomic variability significantly influences diabetes mellitus (DM) treatment outcomes, yet systematic integration of multi-gene panels combining bioinformatics-driven discovery with cross-database validation remains limited across diverse populations. ObjectiveTo develop and validate a comprehensive 20-gene pharmacogenomic panel for predicting drug metabolism variability and treatment response in DM through integrated bioinformatics approaches. MethodsSystematic literature mining identified candidate genes through PubMed searches (2015-2025). Multi-criteria decision analysis prioritized genes across insulin secretion, insulin sensitivity, glucose metabolism, and drug metabolism pathways. Analyses included Gene Ontology enrichment, KEGG pathway mapping, STRING protein-protein interaction networks, variant annotation (dbSNP/ClinVar/PharmGKB), pathogenicity prediction (CADD/PolyPhen-2/SIFT), GTEx tissue-specific expression profiling, and DrugBank drug-gene interaction mapping. Cross-database validation assessed concordance across PharmGKB, DrugBank, GWAS Catalog, and PhKB. ResultsThe panel encompassed 20 genes distributed across 14 chromosomes. Network analysis revealed 87 edges with clustering coefficient 0.653, identifying 5 hub genes. Variant annotation catalogued 3,847 polymorphisms, including 247 pathogenic/likely pathogenic variants. Population analyses demonstrated 3.8-fold inter-ethnic allele frequency variations. PharmGKB integration identified 127 gene-drug pairs (23 Level 1A associations). Cross-database concordance achieved 87.3% (PharmGKB-DrugBank), 82.6% (GWAS Catalog), and 79.4% (PhKB). DrugBank identified 89 antidiabetic drug-gene interactions. Novel associations from recent publications demonstrated statistical significance in cohorts exceeding 2,000 patients. ConclusionsThis integrated framework provides validated foundations for precision diabetes therapeutics. Prospective clinical validation remains essential to translate computational discoveries into actionable decision-support tools optimizing therapeutic outcomes.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.