Network-based genetic prioritization identifies targetable hub nodes and herbal compounds bridging hyper- and hypothyroidism

Objectives We aimed to develop prioritized integrative target optimization (PITO), an integrative computational phytomedicine framework that systematically bridges human genetics, pathway crosstalk analysis, and traditional Chinese medicine (TCM) pharmacology to prioritize therapeutic targets and recommend TCM compounds for autoimmune thyroid diseases (AITDs). The framework identified eight previously unreported hypothyroidism druggable targets and achieved over 80% literature validation for prioritize compounds. Methods PITO was developed as an extension of our original priority index algorithm, which integrates genome-wide association study summary statistics with multimodal regulatory genomics data (nGenes, cGenes, eGenes, and sGenes) and protein-protein interaction networks via random walk with restart. The framework was benchmarked against established methods, including Open Targets Genetics using area under the sensitivity curve and Leading-Gene-for-Target-Discovery metrics. We applied PITO to AITD overall, as well as to its two subtypes (hyperthyroidism and hypothyroidism), and mapped prioritized targets to TCM ingredients and formulas through the BATMAN-TCM 2.0 database, followed by Cytoscape-based network visualization and supraHex-based subtype-specific prioritization mapping. Results PITO demonstrated competitive performance against state-of-the-art methods and successfully recovered 25 phase-Ⅱ-or-above drug targets for AITD, 20 of which are already targeted by approved drugs for other indications. Pathway crosstalk analysis identified 30 crosstalk genes in AITD, of which >50% were seed genes. Integration with TCM pharmacology prioritized 50 compound components, of which 41 (>80%) were validated as thyroid disease-relevant through literature curation. Applying PITO to AITD subtypes revealed a 41-gene pathway crosstalk network for hyperthyroidism, and a 55-gene pathway crosstalk network for hypothyroidism, with 24 genes shared between the two subtypes, alongside 17 hyperthyroidism-specific and 31 hypothyroidism-specific genes. Shared crosstalk genes were enriched in the rat sarcoma virus (Ras) and phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) signaling pathways, hyperthyroidism-specific genes in mammalian target of rapamycin (mTOR) and Ras-related protein 1 signaling, and hypothyroidism-specific genes in nuclear factor kappa-B (NF-κB) and erythroblastic leukemia viral oncogene homolog signaling. Drug repurposing analysis identified five shared crosstalk genes [ epidermal growth factor receptor ( EGFR ), insulin-like growth factor 1 receptor , Kirsten rat sarcoma viral oncogene homolog , sarcoma , and tumor necrosis factor ( TNF )] already targeted by approved drugs, with convergence on wingless-related integration site (WNT)-β-catenin, TNF-α/NF-κB, and PI3K/AKT/mTOR pathways, alongside thirteen subtype-specific druggable genes. Notably, eight hypothyroidism-specific targets [ Bruton’s tyrosine kinase , Erb-B2 receptor tyrosine kinase 2 , Erb-B3 receptor tyrosine kinase 3 , Janus kinase 1 , KIT proto-oncogene receptor tyrosine kinase , MET proto-oncogene receptor tyrosine kinase , platelet-derived growth factor receptor α, and platelet-derived growth factor receptor β ]. have not been previously reported in the context of hypothyroidism. Edge-based network analysis and node attack analysis highlighted EGFR , AKT1 , MYC proto-oncogene , bHLH transcription factor and E1A binding protein p300 as critical regulatory nodes whose targeted disruption significantly destabilize the pathway crosstalk network. Conclusion PITO provides a scalable, genetics-led computational phytomedicine framework for therapeutic target prioritization and TCM compound recommendation in AITD, revealing both shared and subtype-specific molecular pathways and highlighting promising drug repurposing opportunities. By systematically linking genetic evidence with TCM pharmacology, this integrative strategy bridges modern genomics with traditional medicine, offering a rational foundation for precision phytomedicine approaches in complex diseases requiring multitarget interventions.