18th European Symposium on Quantitative Structure – Activity by n a

Read or Download 18th European Symposium on Quantitative Structure – Activity Relationships PDF

Best european books

The Intellectual Origins of the European Reformation, 2nd edition

The sixteenth-century Reformation continues to be a desirable and intriguing quarter of analysis. The revised variation of this exotic quantity explores the highbrow origins of the Reformation and examines the significance of rules within the shaping of historical past. offers an up to date and accelerated model of the unique, highly-acclaimed variation.

Greece in the 1980s

Papers initially offered at a convention, King's university, collage of London, 1981.

Additional resources for 18th European Symposium on Quantitative Structure – Activity Relationships

Example text

Dean, P. M. Compr Med Chem II 2006, 4, 283-305. 2. ; Guba, W. Recent developments in de novo design and scaffold hopping. Curr Opin Drug Discov Devel. 2008, 11, 365-374. 3. ; Stahl, M. Scaffold Hopping. Drug Discovery Today: Technologies 2004, 1, 217-224. 4. , Rarey M. Recore: A Fast and Versatile Method for Scaffold Hopping Based on Small Molecule Crystal Structure Conformations. J. Chem. Inf. Model. 2007, 47, 390-399. 5. , Stahl, M. Similarity Searching in Large Combinatorial Chemistry Spaces.

Bossa, C. (2008) Predictivity of QSAR, Journal of Chemical Information and Modeling, 48(5):971–980. Acknowledgements This research was partly funded by the EU FP7 project OpenTox - An Open Source Predictive Toxicology Framework (Health-F5-2008-200787) and the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety.

Training and test set compounds were prepared on the available experimental data by splitting using: Self Organizing Map (SOM) and random selection through activity sampling6. The training sets were used to derive statistically robust models based on both splitting criteria. The models were then externally validated for their predictivity on test sets. The common set of descriptors selected in both splitting was then used in full models to predict a total of 376 PFCs, including those in REACH preregistration list7.

Download PDF sample

Rated 4.01 of 5 – based on 38 votes