The tutorial presents some data mining approaches suited to QSAR approach of profiling problem.
It is based on the MATLAB system and the MALSAR library.
The MathWorks company provided us with a trial version of their software. To get the trial MATLAB software use the login informations provided to you by eMail on Tuesday 17th of June and follow the instructions in this file. Note that you’ll need to create an account on the MathWorks web site.
The MALSAR is free, you can download it anytime.
Materials
The dataset, scripts and precomputed results that will be needed for the tutorial can be downloaded here:
After unpacking this archive you will find 4 folders:
– Exercises: the MATLAB scripts of each exercise.
– Precomputed: matlab data files (.mat) containing precomputed results to avoid lengthy calculations during the tutorial. Of cours, one can prevent the exercises to load these pre-computed results and reproduce them from scratch.
– Scripts: useful MATLAB scripts to factorise command lines for common tasks.
– Datasets: Molecule and descriptor files used in the tutorials. From this material, one can reproduce the tutorial using a different set of molecular descriptors.
Instructions
Detailed instructions for the tutorial are available here.
The slides of the Tutorial are available here.
Write to Gilles MARCOU for any comments on the material or if you noticed errors in it.