It calculates two descriptors provided by ALOGP=
S program:
- ALogPS_logP -- octanol/water partition coefficient. The algorithm was d=
eveloped with 12908 molecules from the PHYSPROP database using 75 E-state i=
ndicies. 64 neural networks were trained using 50% of molecules selected by=
chance from the whole set. The logP prediction accuracy is root mean squar=
ed error rms=3D0.35 and standard mean error s=3D0.26 [1,2].
- ALogPS_logS -- solubility in water. It was developed using 1291 molecul=
es and provided improved aqueous solubility prediction (rms=3D0.49, s=3D0.3=
8) compared to our previous analysis [3]. The molecules used in this study =
can be downloaded.
References<=
/h2>
- Tetko, I. V.; Tanchuk, V. Y. Application of associative neural networks=
for prediction of lipophilicity in ALOGPS 2.1 program, J. Chem. Inf. Compu=
t. Sci., 2002, 42, 1136-45.
- Tetko, I. V.; Tanchuk, V. Y.; Villa, A. E. Prediction of n-octanol/wate=
r partition coefficients from PHYSPROP database using artificial neural net=
works and E-state indices, J. Chem. Inf. Comput. Sci., 2001, 41, 1407-21.
- Tetko, I. V.; Tanchuk, V. Y.; Kasheva, T. N.; Villa, A. E. Estimation o=
f aqueous solubility of chemical compounds using E-state indices, J. Chem. =
Inf. Comput. Sci., 2001, 41, 1488-93.
------=_Part_108_480609816.1632427487195--