COMPUTATIONAL STUDY ON RNA DINUCLEOTIDES USING THE AM1 SEMI-EMPIRICAL METHOD

The thermodynamic stability, electronic structure, and global reactivity descriptors of all sixteen possible RNA dinucleotides (5′→3′: AA, AC, AG, AU, CA, CC, CG, CU, GA, GC, GG, GU, UA, UC, UG, and UU) were investigated using the semi-empirical Austin Model 1 (AM1) Hamiltonian implemented in MOPAC2016 (version 22.234w). Geometry optimisation was performed for each dinucleotide, followed by the calculation of the heat of formation, dipole moment, frontier molecular orbital (HOMO and LUMO) energies, COSMO surface area, COSMO volume, and molecular weight. Conceptual density functional theory (Conceptual DFT) global reactivity descriptors, including ionisation potential, electron affinity, electronegativity, chemical hardness, chemical softness, chemical potential, and electrophilicity index, were subsequently derived from the HOMO and LUMO energies using Koopmans' approximation. The calculated heats of formation ranged from −300.1 to −643.1 kcal mol⁻¹, indicating that all sixteen RNA dinucleotides are thermodynamically stable, with UU exhibiting the lowest heat of formation (highest thermodynamic stability) and GC the highest heat of formation (lowest thermodynamic stability). The HOMO–LUMO energy gap varied from 3.51 eV for CU to 7.73 eV for UG, suggesting that CU is the electronically softest and potentially the most chemically reactive dinucleotide, whereas UG and UU possess the greatest kinetic stability and the lowest predicted reactivity. The computed global reactivity descriptors consistently support these trends, with CU exhibiting the highest electrophilicity and lowest hardness, while UG and UU display the highest hardness and lowest electrophilicity. Overall, this study provides a systematic quantum chemical dataset describing the thermodynamic and electronic properties of all sixteen RNA dinucleotides. These results establish a comparative reference for understanding sequence-dependent electronic behaviour and intrinsic reactivity in RNA and provide useful baseline data for future computational investigations of RNA structure, function, and molecular interactions.