AIM: To use neural networks, which simulate the functions of living ne
rvous systems, in QSAR studies; METHODS: Using the back-propagation ne
ural networks program devised by us, combining with partial least squa
res (PLS) method, we studied the relationships of quantum chemical ind
ices and analgesic activities of 25 3-methylfentanyl derivatives; RESU
LTS: Through learning process, a good QSAR model was established, and
the activities of these compounds were predicted; the correlation betw
een the activities and quantum chemical indices: the net charge of the
atom N-1, the net charge of the atom O-16, the torsional angle of ato
ms C-10-C-9-N-8-C-4, the interatomic distance between atom C-7 and the
center of phenyl plane C-9-14 (PhA), is quite well-matched. Based on
these results, an interactive pattern between 3-methylfentanyl derivat
ives and opioid receptors was suggested; CONCLUSION: Not only are the
results of neural networks superi- or to those of PLS method but they
also provide accurate predictions of the activity of the compounds and
also combine the PLS method with neural networks.,