Introduction: In the present work, pharmacophore identification and biological activity prediction
for 86 pyrazole pyridine carboxylic acid derivatives were made using the electron conformational
genetic algorithm approach which was introduced as a 4D-QSAR analysis by us in recent years.
In the light of the data obtained from quantum chemical calculations at HF/6-311 G** level, the Electron
Conformational Matrices of Congruity (ECMC) were constructed by EMRE software. Comparing
the matrices, electron conformational submatrix of activity (ECSA, Pha) was revealed that are common
for these compounds within a minimum tolerance. A parameter pool was generated considering the obtained
Methods: To determine the theoretical biological activity of molecules and identify the best subset of
variables affecting bioactivities, we used the nonlinear least square regression method and genetic algorithm.
Results: The results obtained in this study are in good agreement with the experimental data presented
in the literature. The model for training and test sets attained by the optimum 12 parameters gave highly
satisfactory results with R2
training= 0.889, q2=0.839 and SEtraining=0.066, q2
ext1 = 0.770, q2
ext2 = 0.750,
ext3=0.824, ccctr = 0.941, ccctest = 0.869 and cccall = 0.927.