A Report on Fourth Joint Sheffield Conference on
Chemoinformatics June 18-20, University of Sheffield, UK

The initial part of Joelle Gola’s talk consisted of generalities about local and global models and the features of BioFocusDPI’s product Admensa Interactive. Later on, Gola gave what approached a sales pitch about the proprietary algorithm Glowing Molecule which highlights problematic regions within potential drugs responsible for deficiencies in ADMET properties. Gola intended to present case studies of his company’s new automatic techniques for model building. These enable non-computational scientists to capture and share the knowledge contained in their experimental data by building local models for individual chemical series, iteratively improving models as more data are generated and using new models to predict properties for new chemical structures. Gola did describe the Gaussian Process method at the heart of this work^[10] but he ran out of time while rushing through his three examples.

Lastly, Damjan Krstajic, of the Research Centre for Cheminformatics in Serbia described yet another answer to the challenge of designing a system that will cope with constant influx of new information. Discovery Bus aims to automate QSAR modeling without sacrificing the quality of predictions. It is an implementation of Competitive Workflow, a novel software architecture, implemented using autonomous software agents^[11]. All possible combinations of components are explored leading to exhaustive evaluation of potential solutions. The idea is that we cannot know in advance which technique or approach to use in solving a QSAR problem, but if we apply most of the well known techniques and approaches, then we will have an explosion in the number of models, but we will also end up with multiple good solutions among them. A related poster, by David Leahy and co-workers at Newcastle University, described a multi-objective reverse QSAR search agent called Forager. This has been developed to search for non-dominated solutions to the research target profile definition of a new drug within a complex descriptor space, where the search heuristics are provided by multiple QSAR models. Forager uses a modified Particle Swarm Optimization algorithm.

I have reported on only 10 out of 24 presentations. It is hoped that another report will appear in the CSA Trust newsletter. Abstracts for all the papers and posters are online and readers are encouraged to seek out the many interesting papers for which I have not had space to comment. All in all, this was an excellent meeting. It is a shame that it could not accommodate more delegates, but if it did so, perhaps some of the more useful interactions and discussions would be impeded. I shall make a point of booking early for the fifth incarnation of this event.

References:

1. Schuffenhauer A.; Brown, N.; Ertl, P.; Jenkins, J. L.; Selzer, P.; Hamon, J. Clustering and Rule-based Classifications of Chemical Structures Evaluated in the Biological Activity Space. J. Chem. Inf. Model. 2007, 47,325-336.

2. Bemis, G. W.; Murcko, M. A. The Properties of Known Drugs. 1. Molecular Frameworks. J. Med. Chem. 1996, 39, 2887-2893.

3. Schuffenhauer, A.; Ertl, P.; Roggo, S.; Wetzel S.; Koch M. A.; Waldmann, H. The Scaffold Tree - Visualization of the Scaffold Universe by Hierarchical Scaffold Classification. J. Chem. Inf. Model. 2007, 47, 47-58.

4. Brewer, M. Development of a Spectral Clustering Method for the Analysis of Molecular Datasets. J. Chem. Inf. Model. 2007, in press.

5. Boda, K.; Seidel, T.; Gasteiger, J. Structure- and Reaction-based Evaluation of Synthetic Accessibility. J. Comput.-Aided Mol. Des. Published online February 9, 2007.

6. Terfloth, L.; Bienfait, B.; Gasteiger, J. Ligand-based Models for the Isoform Specificity of Cytochrome P450 3A4, 2D6, and 2C9 Susbstrates. J. Chem. Inf. Model. 2007, in press.

7. Zhang, Q.-Y.; Aires-de-Sousa, J. Structure-based Classification of Chemical Reactions without Assignment of Reaction Centers. J. Chem. Inf. Model. 2005, 45, 1775-1783.

8. Bender, A.; Mussa, H. Y.; Glen, R. C.; Reiling, S. Similarity Searching of Chemical Databases Using Atom Environment Descriptors (MOLPRINT 2D): Evaluation of Performance. J. Chem. Inf. Comput. Sci. 2004, 44, 1708-1718.

9. Agrafiotis, D. K.; Xu, H. A Self-organizing Principle for Learning Nonlinear Manifolds. Proc. Natl. Acad. Sci. U. S. A. 2002, 99(25) 15869-15872.

10. Obrezanova, O.; Csányi, G.; Gola, J.; Segall, M. Gaussian Processes: a Method for Automatic QSAR Modeling of ADME Properties. J. Chem. Inf. Model. 2007, in press.

11. Cartmell, J.; Enoch, S.; Krstajic, D.; Leahy, D. E. Automated QSPR through Competitive Workflow. J. Computer-aided Mol. Des. 2005, 19, 821-833.