Hologram quantitative structure-activity relationship (HQSAR) study for anticancer activities of ۲H-chromen-۲-one derivatives

Cancer is one of the deadliest diseases and the second-leading cause of death in the world and remains asignificant health problem worldwide. The numerous drugs have been used for the cancer treatment but havesevere side effects. The current status in the fight against cancer suggests that there is a critical demand ofmore potent anticancer agents with the lowest toxicity and resistance profile. Consequently; increasinginterest has been devoted to the design and discovery of more effective anticancer agents in current medicinalchemistry. In this work, Hologram quantitative structure-activity relationship (HQSAR) models using ۲Hchromen-۲-one derivatives were generated to discover the relationship between the different chemicalstructures and the anti-cancer activity of agents. In the proposed HQSAR model, three fragment parameters,fragment distinction, fragment size and fragment length, were set to “A, C and H”, “۴-۷” and “۵۳”respectively. Conventional validation techniques, internal and external validations such as, non-crossvalidatedcorrelation coefficient (𝑟𝑛𝑐𝑣۲ ), cross-validated correlation coefficient (q۲) and predicted correlationcoefficient (𝑟𝑝𝑟𝑒𝑑۲ ), were utilized to evaluate the forecasting accuracy of proposed model. The HQSAR model(q۲, ۰.۸۵۱; 𝑟𝑛𝑐𝑣۲ , ۰.۹۷۴; 𝑟𝑝𝑟𝑒𝑑۲ , ۰.۸۴۱) for data set (training and test set) of anticancer agents yielded significantstatistical results. The HQSAR contribution maps generated from these models illustrated that the yellow,blue, green-blue and green fields played key roles for improve the anti-proliferative activity of anticanceragents. The final QSAR models could be useful for rational design and development of novel potentanticancer agents in cancer treatment.