Iv. Tetko et al., HIV-1 REVERSE-TRANSCRIPTASE INHIBITOR DESIGN USING ARTIFICIAL NEURAL NETWORKS, Journal of medicinal chemistry, 37(16), 1994, pp. 2520-2526
Artificial neural networks were used to analyze and predict the human
immunodeficiency virus type 1 reverse transcriptase inhibitors. The tr
aining and control sets included 44 molecules (most of them are well-k
nown substances such as AZT, dde, etc.). The activities of the molecul
es were taken from literature. Topological indices were calculated and
used as molecular parameters. The four most informative parameters we
re chosen and applied to predict activities of both new and control mo
lecules. We used a network pruning algorithm and network ensembles to
obtain the final classifier. Increasing of neural network generalizati
on of the new data was observed, when using the aforementioned methods
. The prognosis of new molecules revealed one molecule as possibly ver
y active. It was confirmed by further biological tests.