Answers are grounded in this paper and cite their sources.
VEGFR-1 and VEGFR-2 have distinct but complementary roles in vascular endothelial growth factor signaling, providing a rationale for comparing ligand recognition by their kinase domains in neovascular age-related macular degeneration. We integrated dual-receptor docking, fragment tree Transformer-based variational autoencoder generation, and three-dimensional Monte Carlo tree search to identify candidate ligand scaffolds. Screening 37,445 natural product-like structures retained 989 compounds, and subsequent molecular generation and physicochemical filtering produced 1,315 structures for receptor-constrained optimization. Twenty candidates were prioritized, of which VDS-1, VDS-2, and VDS-3 were examined using three independent 100 ns molecular dynamics simulations per complex and MM/PBSA analysis. VDS-1 showed the most negative MM/PBSA estimates for VEGFR-2 (-49.79 ± 0.21 kcal/mol) and VEGFR-1 (-45.95 ± 1.84 kcal/mol), whereas VDS-2 exhibited substantial between-trajectory variability. VDS-3 showed limited between-trajectory variation in its MM/PBSA estimates for both receptors, although protein-aligned ligand RMSD revealed between-replicate heterogeneity, particularly in VEGFR-1. Van der Waals interactions predominated among favorable modeled contributions, with PHE1047 in VEGFR-2 and ILE881 in VEGFR-1 emerging as recurrent interaction regions. Although VDS-1 showed favorable modeled interactions, it had substantial physicochemical and predicted safety liabilities. VDS-3 therefore represented a comparatively balanced starting scaffold for further optimization, although experimental validation and toxicity assessment remain necessary.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: