QSRR Study of Organic Dyes by Multiple Linear Regression Method Based on Genetic Algorithm (GA–MLR)

Quantitative structureretention relationships (QSRRs) are used to correlate paper chromatographic retention factors of disperse dyes with theoretical molecular descriptors. A data set of 23 compounds with known RF values was used. The genetic algorithmmultiple linear regression analysis (GAMLR) with three selected theoretical descriptors was obtained. The stability and predictability of the model was validated by use of leaveoneout (LOO), leavemanyout (LMO) crossvalidation, external validation, Yrandomization and applicability domain (AD) analysis. The GAMLR revealed a statistically meaningful model showing the dependence of the RF value on sum of topological distances between N and Br atoms (T(N..Br)), global topological charge index (JGT) and R autocorrelation of lag 5 / unweighted (R5u_A) of the compounds.