Review user | www.CADASTER.eu

try another color:

try another fontsize: 60% 70% 80% 90%

www.CADASTER.eu

Name of method:

hybrid ensemble models

Descriptors used:
(e.g., sets provided on the challenge site)

E-state indices	Dragon descriptors
SimulationsPlus	QuantumChemistry
MOE	Only structures (e.g. in graph kernels)

Additional descriptors (indicate type, vendor or/and provide brief description and references):

we used our own molecular descriptors as implemented in the pipeline pilot software, in particular molecular properties, Electrical State Counts and Ghose-Crippen AlogP counts
(http://accelrys.com/products/scitegic/). Furthermore we used our own reactivit

Chemical classes:

Have you used information about chemical classes to build your model?
No
Yes
If yes, please, describe:

Mechanism and/or mode of action:

Have you used information about mechanisms and/or modes of action to build your model?
No
Yes
If yes, please, describe:

Descriptor selection:

Have you done selection of descriptors?
No
Yes
If yes, specify, number of descriptors selected for the final model:

If yes, describe the used procedure:

K-fold cross-validation

Method (select one):

If any other, please, specify:

We generated an hybrid ensemble model with 101 individual models that were trained on subsamples of the entire data set.
The final ensemble included 56 neural networks, 43 kridge-regression models and 4 svm with linear kernels.

Have you also used the Known set to develop the final model?

No
Yes

Software suite/vendor (select one):

If any other, please, specify:

ENTOOL, written in MATLAB (http://www.j-wichard.de/entool/)
Pipeline Pilot (http://accelrys.com/products/scitegic/)

Error Prediction

Have you provided the predicted errors?
No
Yes
If yes, provide the description:

We predicted the errors with the modeling errors from the known samples (residulas). Therefore we looked for the 3 nearest neighbors in the variable space and took the mean of the residuals. We used an euclidean metric with additional weights that were bas

Detailed description of the method:

The description should be sufficient to reproduce results, please, also provide references to the respective articles with more details, if available:

Hybrid ensemble model with selected molecular descriptors. For  the final model, training and validation set are merged together in order to have more training samples !
For model development, only the training samples were used. For the prediction of the blind data set, all available data was used in accordance with the instructions at this page:
http://www.cadaster.eu/node/71