Prediction of fish bioconcentration factors of nonpolar organic pollutants based on molecular connectivity indices.

The relationship between bioconcentration factors and molecular connectivity indices was investigated. A regression model was developed using 80 measured BCFs of nonpolar organic pollutants. The five topological parameters used were 1 chi, 2 chi, 3 chi c, 0 chi v and 2 chi v. Modified jackknife tests were applied to examine the robustness of the model by repeatedly removing a set of or a class of compounds from the database. The model was compared with one using Kow as an independent parameter. The mean absolute errors for the 80 compounds studied were 0.288 and 0.302 log-unit for the two models, respectively.