QSAR studies on some marine natural alkaloids as anticancer agents in cancer leukemia
QSAR studies on some marine natural alkaloids as anticancer agents in cancer leukemia
Mahmood Saeedi Kelishami1 Ghasem Ghasemi2 Reza gilanifar3 Mohammad Zakarianejad4
1) University professor,Rasht Branch, Islamic Azad University, Rasht, Iran E-mail:
2) University professor,Rasht Branch, Islamic Azad University, Rasht, Iran E-mail:
3) University professor,Payame Noor University, Sirjan Branch, Kerman, Iran E-mail:
4) University professor, Payame Noor University, Sirjan Branch, Kerman, Iran E-mail:
IOI Code
:
XYHA-BECCA
Publication :
The 4th International conference of Science and Engineering(4icesconf.com)
Abstract :
Alkaloids are naturally occurring nitrogen containing biologically active heterocyclic compounds. Over the last few years, a large number of biologically important alkaloids with antiviral, antibacterial, anti-inflammatory, antimalarial, antioxidant and anticancer activities have been isolated from marine source. Present article to use of method quantitative structure-activity relationship (QSAR) study has been done on some marine natural alkaloids as anticancer agents in cancer leukemia. Multiple linear regression (MLR), partial least squares (PLS) and principal component regression (PCR) were used to create QSAR models. For this purpose, geometry optimization performed at B3LYP level with a known basis set (6–31G (d)). Hyperchem, ChemOffice , Gaussian03W and Dragon softwares were used for geometry optimization of the molecules and calculation of the quantum chemical descriptors. Finally, Unscrambler program was used for analysis of data. For the compounds in the gas phase RMSEtrain ،RMSEtest and R2 with jack-knife method 0.1972, 0.7356 and 0.68 respectively. The values of R and R2 to GA-stepwise MLR model 0.984 and 0.969 respectively and also RMSEtrain and RMSEtest with Genetic algorithm-Artificial neural network (GA-ANN) respectively 0.1388 and 0.4208 are obtained. In end The GA-stepwise MLR method other than the method most appropriate for this cell is known.
Keywords : leukemia genetic algorithm multiple linear regression principal component regression partial least squares