Mechanistic quantitative structure-activity relation ships (QSAR) variational principles relating data screening and data analysis are mainly introduced as: the longest simplified molecular-input line-entry system-SMILES' molecular chain (LoSMoC) to achieve for the maximum 1D-to-2D information of molecular input in chemical interaction with a receptor site; and, respectively, the shortest paths in between the endpoints of chemical-biological interaction, as an Euclidian metrics in modeling ligand-receptor space. Moreover, in prediction analysis, the max-min QSAR procedure employs molecular descriptors as electronegativity, chemical hardness, chemical power, electrophilicity, and lipophilicity which associate as well with variational principles of chemical reactivity viz.: mid of HOMO-LUMO annealing, equalization of HOMO-LUMO, minimize of charge flow, potential barrier tunneling and cell walls' transduction optimization, respectively, for the electrons or molecular ligand-receptor fragments on their highest occupied and lowest unoccupied molecular orbitals (HOMO and LUMO). As a working application the case of anti-HIV pyrimidinic 1,3-disubstituted uracil derivatives is employed towards revealing the chemical reactivity based QSAR optimum mechanism of interaction within sevenfold variational stages as provided by two different SMILES-based screening criteria.