ABSTRACT
Abstract ................................................................................................. 208 10.1 Introduction ................................................................................ 208 10.2 General Framework.................................................................... 209 10.3 Application Examples ................................................................ 212 10.4 Conclusions ................................................................................ 221 Acknowledgments ................................................................................. 221 Keywords .............................................................................................. 221 References ............................................................................................. 221
EMILI BESALÚ1,*, LIONELLO POGLIANI2, and J. VICENTE JULIAN-ORTIZ2
1Departament de Química, Institut de Química Computacional i Catàlisi (IQCC), Universitat de Girona, 17003 Girona, Catalonia, Spain
2Departamento de Química Física, Unidad de Investigación de Diseño de Fármacos y Conectividad Molecular, Facultad de Farmacia, Universitat de València, Burjassot, València, Spain and MOLware SL, Valencia, Spain, E-mail: liopo@uv.es, jejuor@uv.es
*Corresponding author. E-mail: emili.besalu@udg.edu
ABSTRACT
After the presentation of the Superposing Significant Interaction Rules (SSIR) method in a previous volume of this AAP collection, this chapter shows how the procedure can be applied to inspect the data attached to an experimental design. This is possible due to two main reasons: first, the symbolic treatment of the data, that confers to it potential use in many fields; and second, because the rules being considered by the procedure are attached to events that are probabilistically evaluated. This chapter presents two example applications performed with data originally prepared for design of experiments. It is shown how the SSIR method is fast and helps to point to the correct direction of response optimization. The advantages of SSIR in terms of simplicity and the availability to deal with unbalanced experimental designs are discussed.
