Bhushan A Baviskar, Sharada L Deore, Sagar Alone
Government College of Pharmacy, Amravati-444 604, Maharashtra, INDIA.
Published: January 2019
Type: Original Article
In the present communication, total seventeen saponin molecules were collected and subjected to quantitative structure-activity relationship (QSAR) analyses. 2D QSAR model is developed by using multiple linear regression method against 4 different cell lines. QSAR model was generated by using training set of 11 and test set of 6 molecules having correlation coefficient (r2), significant cross validated correlation coefficient (q2) and F-test (For statistical significance) is as given below (A-549: r2- 0.9281, q2- 0.8691, F-test- 51.6079), (SK-OV-3: r2- 0.9554, q2- 0.9184, F-test- 85.7357), (SK-MEL-2: r2- 0.9160, q2- 0.8285, F-test- 43.6084), (HCT15: r2- 0.9203, q2- 0.8357, F-test- 46.1887). In this QSAR study Alignment independent descriptors such as T_2_C_7, T_O_O_5 and physicochemical descriptors like Chain path count such as 6 chain count and Chi chain such as Chi 6 chain were most responsible descriptors for anticancer activity.
Key words: QSAR, Pulsatilla koreana, Cytotoxic activity, Multiple linear regressions.