Objective: The EC-GA method was employed in this study as a 4D-QSAR method, for the
identification of the pharmacophore (Pha) of ruthenium(II) arene complex derivatives and quantitative
prediction of activity.
Methods: The arrangement of the computed geometric and electronic parameters for atoms and bonds
of each compound occurring in a matrix is known as the electron-conformational matrix of congruity
(ECMC). It contains the data from HF/3-21G level calculations. Compounds were represented by a
group of conformers for each compound rather than a single conformation, known as fourth dimension
to generate the model. ECMCs were compared within a certain range of tolerance values by using the
EMRE program and the responsible pharmacophore group for ruthenium(II) arene complex derivatives
was found. For selecting the sub-parameter which had the most effect on activity in the series and the
calculation of theoretical activity values, the non-linear least square method and genetic algorithm
which are included in the EMRE program were used. In addition, compounds were classified as the
training and test set and the accuracy of the models was tested by cross-validation statistically.
Results: The model for training and test sets attained by the optimum 10 parameters gave highly satisfactory
results with R2
training= 0.817, q 2=0.718 and SEtraining=0.066, q2
ext1 = 0.867, q2
ext2 = 0.849, q2
=0.895, ccctr = 0.895, ccctest = 0.930 and cccall = 0.905.
Conclusion: Since there is no 4D-QSAR research on metal based organic complexes in the literature,
this study is original and gives a powerful tool to the design of novel and selective ruthenium(II) arene