Özet:
In the present study, the mutagenic activity of diverse structure of azo dyes and aromatic amine derivatives on TA98 Salmonella typhimurium bacterial strain with S9 activation was used to generate a quantitative structure – activity relationship (QSAR) model. The descriptors required for the model development were obtained by SPARTAN (v.10), DRAGON (v.7.0) software packages. The selection of descriptors was carried out by the tools implemented in QSARINS (v.2.2.3) software. Many division trials were performed on the dataset as training and test sets which comprise the 80% and 20% of the whole set, respectively. 6 descriptors (CIC2, Chi_D/Dt, L/Bw, TDB09p, Mor28s and piPC08) constitute the final model. The applicability domain (AD) of the generated QSAR model was defined by both the ranges of response and descriptors. The predictive ability of the final model was tested using an external dataset consisting of currently used 33 anionic water soluble textile dyes (eight anthraquinone and 25 azo dyes) with no experimental mutagenicity data. The proposed QSAR model had 70% structural coverage for the external set chemicals. The mutagenic activities of 22 current textile dyes belonging anthraquinone and azo dyes fell within the applicability domain of the proposed QSAR model which means that they were well predicted by the model. The order of 6 anthraquinone dyes which are mainly used for cotton, fiber dyeing and leather shading with the most mutagenic activity is: Acid Blue 62>Acid Blue 40>Acid Blue 45>Acid Blue 80>Acid Blue 230>Acid Blue 344. The least mutagenic azo dye is Direct Orange 34 which is mainly used for cotton, silk, wool and their blended fabric dyeing and printing, also can be used for leather and paper shading.
