Abstract (EN):
A series of alpha-N-heterocyclic carboxaldehyde thiosemicarbazones derivatives exhibit anticancer activity by inhibiting ribonucleotide reductase (RNR) enzyme was considered for the present computational study. The validated quantitative structure activity relationship (QSAR) models constructed with vsurf (HB7, WP7 and DW23) and molar refractivity (SMR_VSA5) descriptors yielded the cross-validated correlation coefficient of > 0.6, shows that the models have sufficient predictive ability. The SMR_VSA5 descriptor is the main contributor for the activity prediction in all models, which measure the steric factors and bulkiness of the given molecules. The negative contribution of the molar refractivity descriptor shows that the molecular volume should be low with its polar properties. The vsurf descriptors are dependent on the structure connectivity and conformation (dimensions are measured in ) which are useful in surface property prediction. The vsurf descriptors reveal that the surface polarity, hydrogen bond donor properties and hydrophilic contact surface of the molecules are important for the activity. The pharmacophore analysis results obtained from this study shows that the distance between the aromatic/hydrophobic and the PiN sites to the H-bond donor and acceptor groups should be connected with almost the same distance for significant RNR inhibitory activity.
Language:
English
Type (Professor's evaluation):
Scientific
Contact:
hari.moorthy@fc.up.pt; pafernan@fc.up.pt
No. of pages:
8