[Prediction of drug bioavailability by genetic algorithm and artificial neural network].

AIM

To set up an artificial neural network system and optimize by genetic algorithm (GA) to predict drug bioavailability.

METHODS

Genetic algorithm was used to optimize weights of the artificial neural network. The optimal solution of the artificial neural network model at a specific condition was obtained using the good search ability of genetic algorithm in order to predict drug bioavailability. Volume, refractivity, lgP(c), hydration, polarizability, E(HOMO) and E(LUMO) are inputs of the drug bioavailability prediction neural network, and its output is average drug bioavailability.

RESULTS

The prediction precision of average drug bioavailability of the GA- neural network model is 95.9%.

CONCLUSION

This model can be used in the forecasting of drug bioavailability.