Abstract
A neural network methodology based on the radial basis function (RBF) architecture is introduced in order to establish quantitative structure-toxicity relationship models for the prediction of toxicity. The dataset used consists of 221 phenols and their corresponding toxicity values to Tetrahymena pyriformis. Physicochemical parameters and molecular descriptors are used to provide input information to the models. The performance and predictive abilities of the RBF models are compared to standard multiple linear regression (MLR) models. The leave-one-out cross validation procedure and validation through an external test set produce statistically significant R 2 and RMS values for the RBF models, which prove considerably more accurate than the MLR models.
Experimental vs predicted toxicity using the RBF methodology for the test set (41 compounds)
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Lu FC, Kacew S (2002) Lu’s basic toxicology. Taylor & Francis, London
Karcher W, Devillers J (1990) SAR and QSAR in environmental chemistry and toxicology: scientific tool or wishful thinking?. In: Karcher W, Devillers J (eds) Practical applications of quantitative structure-activity relationships (QSAR) in environmental chemistry and toxicology. Kluwer, Dordrecht, pp 1–12
Nendza M (1998) Structure-activity relationships in environmental sciences, ecotoxicology series 6. Chapman & Hall, London
Schultz TW, Netzeva TI, Cronin MTD (2003) SAR QSAR Environ Res 14:59–81
Netzeva TI, Schultz TW, Aptula AO, Cronin MTD (2003) SAR QSAR Environ Res 14:265–283
Zahouily M, Rhihil A, Bazoui H, Sebti S, Zakarya D (2002) J Mol Model 8:168–172
Cronin MTD, Aptula AO, Duffy JC, Netzeva TI, Rowe PH, Valkova IV, Schultz TW (2002) Chemosphere 49:1201–1221
Ren S (2003) Chemosphere 53:1053–1065
Bukard U (2003) Methods for data analysis. In: Gasteiger J, Engel Th (eds) Chemoinformatics. Wiley VCH, Weinheim, pp 439–485
Devillers J (1996) Neural networks in QSAR and drug design. Academic Press, London
Afantitis Α, Melagraki G, Makridima K, Alexandridis A, Sarimveis H, Iglessi-Markopoulou O (2005) J Mol Struct: Theochem 716:193–198
Devillers J (2004) SAR QSAR Environ Res 15:237–249
Kaiser KLE (2003) Quant Struct-Act Relat 22:1–5
KaiserKLE (2003) J Mol Struct: Theochem 622:85–95
Gasteiger J (2003) Handbook of chemoinformatics: from data to knowledge, vol 3. Wiley VCH, Weinheim
Zupan J, Gasteiger J (1999) Neural networks in chemistry and drug design. Wiley VCH, Weinheim
Debnath AK (2001) Quantitative structure-activity relationship (QSAR): a versatile tool in drug design. In: Ghose AK, Viswanadhan VN (eds) Combinatorial library design and evaluation: principles, software tools, and applications in drug discovery. Marcel Dekker, New York, pp 73–129
Sarimveis H, Alexandridis A, Tsekouras G, Bafas G (2002) Ind Eng Chem Res 41:751–759
Lessigiarska I, Cronin MTD, Worth AP, Dearden JC, Netzeva TI (2004) SAR QSAR Environ Res 15:169–190
Aptula AO, Netzeva TI, Valkona IV, Cronin MTD, Schultz TW, Kuhne R, Schuurmann G (2002) Quant Struct-Act Relat 21:12–22
Darken C, Moody J (1990) Fast adaptive K-means clustering: some empirical results. IEEE INNS Int Joint Conf Neural Netw 2:233–238
Dunn JC (1974) J Cybernet 3:32–57
Leonard JA, Kramer MA (1991) Radial basis function networks for classifying process faults. IEEE Control Syst 11:31–38
Kennard RW, Stone LA (1969) Technometrics 11:137–148
Acknowledgements
G.M. wishes to thank the Greek State Scholarship Foundation for a doctoral assistantship.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Melagraki, G., Afantitis, A., Makridima, K. et al. Prediction of toxicity using a novel RBF neural network training methodology. J Mol Model 12, 297–305 (2006). https://doi.org/10.1007/s00894-005-0032-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00894-005-0032-8