Journal of Cheminformatics

tracked for impact factor
Search for
Advanced search
Journal of Cheminformatics
Tools
Share this article

This article is part of the supplement: 5th German Conference on Cheminformatics: 23. CIC-Workshop

Open Access Oral presentation

Molecular bioactivity extrapolation to novel targets by support vector machines

Gerard JP Van Westen1*, JK Wegner2, AP IJzerman1, HWT Van Vlijmen2 and A Bender1

Author Affiliations

1 Amsterdam Center for Drug Research, Einsteinweg 55, 2333 CC, Leiden, The Netherlands

2 Tibotec, Gen De Wittelaan L 11B 3, 2800 Mechelen, Belgium

For all author emails, please log on.

Journal of Cheminformatics 2010, 2(Suppl 1):O3 doi:10.1186/1758-2946-2-S1-O3


The electronic version of this article is the complete one and can be found online at: http://www.jcheminf.com/content/2/S1/O3


Published:4 May 2010

© 2010 Gerard JP et al; licensee BioMed Central Ltd.

Oral presentation

The early phases of drug discovery use in silico models to rationalize structure activity relationships, and to predict the activity of novel compounds. However, the performance of these models is not always acceptable and the reliability of external predictions - both to novel compounds and to related protein targets - is often limited. Proteochemometric modeling [1] adds a target description, based on physicochemical properties of the binding site, to these models.

Our proteochemometric models [2] are based on Scitegic circular fingerprints on the compound side and on a customized protein fingerprint on the target side. This protein fingerprint is based on a selection of physicochemical descriptors obtained from the AAindex database. Through PCA we selected a number of physicochemical properties which are hashed in a fingerprint using the Scitegic hashing algorithm. We compared this fingerprint to a number of protein descriptors previously published, including the Z-scales, the FASGAI and the BLOSUM descriptors. Our fingerprint performs superior to all of these. In addition, we show that proteochemometric models improve external prediction capabilities. In the case of classification this leads to models with a higher specificity when compared to conventional QSAR. In the case of regression our models show an average lower RMSE of 0.12 log units when based on a pIC50 output variable compared to conventional QSAR modeling the same data-set. Furthermore, our models enable target extrapolation. As a result we can predict the activity of known and new compounds on new targets while retaining the same model quality as when performing external validation without target extrapolation.

References

  1. Freyhult E, Prusis P, Lapinsh M, Wikberg JE, Moulton V, Gustafsson MG: Unbiased descriptor and parameter selection confirms the potential of proteochemometric modelling.

    BMC Bioinformatics 2005, 6:50. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text OpenURL

  2. Doddareddy MKR, Van Westen GJP, Horst E, Peironcely JE, Corthals F, IJzerman AP, Emmerich M, Jenkins JL, Bender A: Chemogenomics: Looking at Biology through the Lens of Chemistry.

    Stat Anal Data Mining 2009, in press. OpenURL