This paper describes the derivation of a simple QSAR model for the prediction of log BB from a set of 55 diverse organic compounds. The model contains two variables: polar surface area (PSA) and calculated logP, both of which can be rapidly computed. It therefore permits the prediction of log BB for large compound sets, such as virtual combinatorial libraries. The performance of this QSAR on two test sets taken from the literature is illustrated and compared with results from other reported computational approaches to log BB prediction.

A method for the rapid computation of polar molecular surface area (PSA) is described. It is shown that consideration of only a single conformer when computing PSA gives an excellent correlation with intestinal absorption data-as good as previously reported methods employing multiple conformers. Circumventing a time-consuming conformational analysis opens the possibility of computationally screening large numbers of compounds for problems relating to absorption prior to synthesis. The robustness of the criterion for identifying poorly absorbed compounds (PSA >/= 140 A(2)) is illustrated through its application to a diverse test set of 74 drugs. The PSA-based method is also compared to an experimental method for absorption prediction recently described in the literature.

No abstract

No abstract

No abstract

No abstract

No abstract

No abstract