| | | | The Mother of All QSAR (MOAQ) project will start in the last months of 2009 and will exploit the new Cloud Bus facilties that we have developed with the help of Microsoft Research and their Azure platform. It will take thousands of publically available experimental properties and chemical structure datasets from the StarLight database of the European Bioinformatics Institute and the WOMBAT database, kindly given to us by Prof Tudor Oprea of the University of New Mexico. The Discovery Bus is then expected to generate around 200,000 predictive QSAR models from this data for use in the meta analysis of modeliing performance as well as for use in multi-property predictions. We expect the MOAQ study to take around 5 years on our single worker server. Thanks to Microsoft and their Azure platform we expect this to be doen within several weeks of processing.
The models will be made freely available for inspection and prediction at the new Open QSAR web-site and we hope this will be used by members of the molecular design community and to support fundamental research, open source drug discovery and reduction of animal experimentation, particularly arising from teh new EU REACH legislation.
|
| |
|
