Wavelet-domain multiway spectral separation of free drug, DNA, and drug-DNA complex profiles for quantitative binding...

Classical UV spectrophotometric analysis of drug-DNA interactions is fundamentally limited by severe spectral overlap and by linearization schemes relying on molar absorptivity assumptions. Under such conditions, conventional approaches, including the Benesi-Hildebrand method, often yield ambiguous or underestimated binding parameters. In this work, the UV spectroscopic binding analysis is reformulated within a wavelet-assisted multiway framework that enables resolution of highly overlapped UV spectra of the free drug, DNA, and the drug-DNA complex. Binding analysis is performed using the fractional occupancy ratio θ/(1 - θ), providing an assumption-free description of the drug-DNA equilibrium. UV absorption spectra recorded during the titration of hydroxychloroquine (HYD) with DNA were transformed into the wavelet domain using the continuous wavelet transform (CWT) and arranged as a three-way tensor. Parallel factor analysis (PARAFAC) decomposition enabled unique resolution of free HYD, free DNA, and the HYD-DNA complex without reference spectra or additional spectroscopic assumptions. Fractional occupancy values increased from 0.39 to 0.58 as DNA concentration increased from 4.0 × 10^-5 to 2.0 × 10^-4 M, indicating a consistent binding process. Binding constants evaluated using three definitions of free DNA ranged from (9.6 ± 3.8) × 10³ to (2.28 ± 1.42) × 10⁴ M^-1. In contrast, classical Benesi-Hildebrand analysis yielded a substantially lower apparent binding constant (∼1.77 × 10³ M^-1), underscoring the limitations of single-wavelength linearization in overlapped systems. The proposed wavelet-PARAFAC framework provides a robust, assumption-free strategy for resolving highly overlapped UV spectra and for quantitative drug-DNA binding analysis directly from UV spectroscopic data.