Removal rates for pharmaceuticals and dyes have been modelled using so-called quasi-SMILES, which are representations of the above processes. Quasi-SMILES is an extend of the simplified molecular input-line entry system (SMILES) where, in addition to information on the molecular structure, the codes of physicochemical conditions are included. In addition, these codes can be a representation for various eclectic circumstances, such as presence or absence of light, impact of x-Rays beems, as well seasons (e.g. summer-winter). Analysis of quasi-SMILES of pharmaceuticals by Monte Carlo technique, applied via the CORAL software, shows it is possible to build predictive models using a one-variable correlation between optimal (flexible) descriptors and the removal rates. Removal rates used to build the model were obtained from recent publications including seasonal differences. The statistical characteristics of the best models for removal rates of pharmaceuticals and dyes are quite good for external validation set. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.