Abstract
The design of new medical drugs is a very complex process in which combinatorial chemistry techniques are used. The goal consists of discriminating between molecular compounds exhibiting or not certain pharmacological activities. Different machine learning approaches have been recently applied to different drug design problems leading to competitive results in pointing at particular compounds with high probability of exhibiting activity. The present work first deeps into the natural trade-off between accuracy in the much less populated active group and false alarm rate which could lead to too many expensive laboratory tests. Preliminary results show how different classification techniques are suited for this particular problem and throw light to keep improving the results by considering also the acceptance/rejection trade-off.
This work has been partially funded by spanish project TIC2003-08496.
Chapter PDF
Similar content being viewed by others
Keywords
- Receiver Operating Characteristic
- Receiver Operating Characteristic Curve
- False Positive Rate
- False Alarm Rate
- True Positive Rate
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Balaban, A.T., Motoc, I., Bonchev, D., Makenyan, O.: Topological indices for structure-activity correlations. Top. Curr. Chem. 114, 21–55 (1983)
Basak, S.C., Bertelsen, S., Grunwald, G.: Application of graph theoretical parameters in quantifying molecular similarity and structure-activty studies. J. Chem. Inf. Comput. Sci. 34, 270–276 (1994)
Castro, M.J., Díaz, W., Aibar, P., Domínguez, J.L.: Prediction and Discrimination of Pharmacological Activity by Using Artificial Neural Networks. In: Perales, F.J., Campilho, A.C., Pérez, N., Sanfeliu, A. (eds.) IbPRIA 2003. LNCS, vol. 2652, pp. 184–192. Springer, Heidelberg (2003) ISSN 0302-9743
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification, 2nd edn. John Wiley and Sons, Chichester (2001)
Fawcett, T.: Roc graphs: Notes and practical considerations for researchers. Machine Learning, submitted (2004), http://www.hpl.hp.com/personal/Tom_Fawcett/papers/ROC101.pdf
Gálvez, J., García-Domenech, R., de Julián-Ortiz, J.V., Soler, R.: Topological approach to drug design. J. Chem. Inf. Comput. Sci. 35, 272–284 (1995)
Jaén-Oltra, J., Salabert-Salvador, M.T., García-March, F.J., Péz-Giménez, F., Tomás-Vert, F.: Artificial neural network applied to prediction of fluorquinolone antibacterial activity by topological methods. J. Med. Chem. 43, 1143–1148 (2000)
Murcia-Soler, M., Pérez-Giménez, F., García-March, F.J., Salabert-Salvador, M.T., Díaz-Villanueva, W., Medina-Casamayor, P.: Discrimination and selection of new potential antibacterial compounds using simple topological descriptors. J. Mol. Graph. Model 21, 375–390 (2003)
Seybold, P.G., May, M., Bagal, U.A.: Molecular structure-propertiy relationships. J. Chem. Educ. 64, 575–581 (1987)
Tomás-Vert, F., Pérez-Giménez, F., Salabert-Salvador, M.T., García-March, F.J., Jaén-Oltra, J.: Artificial neural network applied to the discrimination of antibacterial activity by topological methods. J. Mol. Struct (THEOCHEM) 504, 272–276 (2000)
Witten, I.H., Frank, E.: Data Mining: Practical machine learning tools with Java implementations. Morgan Kaufmann, San Francisco (2000)
Zell, A., et al.: SNNS: Stuttgart Neural Network Simulator. User Manual, Version 4.2. Institute for Parallel and Distributed High Performance Systems, University of Stuttgart, Germany (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Díaz, W., Castro, M.J., Ferri, F.J., Pérez, F., Murcia, M. (2004). Improving Pattern Recognition Based Pharmacological Drug Selection Through ROC Analysis. In: Sanfeliu, A., Martínez Trinidad, J.F., Carrasco Ochoa, J.A. (eds) Progress in Pattern Recognition, Image Analysis and Applications. CIARP 2004. Lecture Notes in Computer Science, vol 3287. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30463-0_78
Download citation
DOI: https://doi.org/10.1007/978-3-540-30463-0_78
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23527-9
Online ISBN: 978-3-540-30463-0
eBook Packages: Springer Book Archive
