The tutorial aims at presenting the Generative Topographic Mapping (GTM) landscapes . For more details about the GTM algorithm, see [Bishop et al, Neural Computation 10, No. 1, 215–234 (1998)]. For details about GTM predictive models and landscapes, see [H.A. Gaspar et al. Mol. Informatics, 2015, 34 (6-7), 348-356]
The GTM is an unsupervised method to map high dimensional data to a two-dimensional representation. In the process, the GTM builds a probabilistic model of the data that can be exploited for data characterization, comparison or classification and regression model building. The GTM approach will be used to analyze a bioconcentration factor dataset and to explore structure-activity relationships. As a result, several property and activity landscapes will be generated and visualized.
Software
The tutorial is based on six pieces of software :
- xGTMapTool : a graphical user interface frontend for the preparation of a GTM.
- xGTMview : an application to link the GTM trained on chemical data and the chemical structures.
- xGTMReSample : an application to improve the resolution of a GTM model.
- xGTMReg : an application to build and validate a GTM-based regression model.
- xGTMClass : an application to build and validate a GTM-based classification model.
- xGTMLandscape : an application to visualize and navigate the activity and property landscapes.
The software are supplied online and can be downloaded for the OS of your choice. :
- Windows (Win10, 64bits)
- Mac (for Intel : Monterey 12.4. For M1 : BigSur or Monterey 12.4)
- Linux (Ubuntu 20 64bits or Ubuntu 20 64 bits)
Download link (Contains Software + License – valid until the 1st of August, 2022)
Software available until the 17th of July, 2022
Later, you can ask for updated versions here :
http://infochim.u-strasbg.fr/spip.php?rubrique41
The license of the software is distributed freely and a license file, called « licence.dat_CS32022 » (to be renamed to « licence.dat ») is distributed with the software for the OS of your choice (Windows, Mac or Linux).
The license file must be installed in a proper location to be found.
- On Windows : create the directory AppData\local\ISIDAGTM directory at the root of your home directory and copy the file license.dat in it. The absolute path of the file should be similar to this one :
C :\Users\username\AppData\local\ISIDAGTM\licence.dat
The file and the directory should have read and write permissions.
- On Mac : create the directory .config/ISIDAGTM directory at the root of your home directory and copy the file license.dat in it. The absolute path of the file should be similar to this one :
/Users/username/.config/ISIDAGTM/licence.dat
- On Linux : create the directory .config/ISIDAGTM directory at the root of your home directory and copy the file license.dat in it. The absolute path of the file should be similar to this one :
/home/username/.config/ISIDAGTM/licence.dat
Datasets and Step-by-step instructions
The tutorial uses a dataset of bioconcentration factor [Lunghini, F.et al. QSAR & Env. Protection, 30(7), 507-524].
The dataset and instructions can be downloaded here.