This study applies the Lo-Shu Projection (LSP) framework to characterize the geometric pharmacological signatures of bioactive compounds across two pharmacologically distinct TCM formulas: Xuefu Zhuyu Tang (XFZYT, Huoxue/blood-activating) and Buyang Huanwu Tang (BYHWT, Buyang/Qi-tonifying). The Lo-Shu Local Transition Energy score (QLTE), derived from protein backbone phi/psi dihedral angle mapping onto the 3x3 Lo-Shu numeric grid, significantly discriminated mechanism-specific targets from controls in both formulas: XFZYT (p=7. 68e-6, d=-0. 790) and BYHWT (p<0. 0001, d=-1. 249). AutoDock Vina binding energies showed no discriminatory power in either formula (p=0. 082 and p=0. 470 respectively). Calycosin (Huang-Qi, Jun herb) demonstrated Buyang-selective binding (p=0. 047), providing the first compound-level geometric validation of Jun herb mechanism specificity. Total: 372 compound-target pairs across 22 bioactive compounds and 34 protein targets (with overlap). These findings establish QLTE as a universal, parameter-free geometric classifier for TCM pharmacological mechanism classification.
Yao-Kai Kao (Sun,) studied this question.