Hit To Lead Optimization, And 3D QSAR Modeling of Imidazolopiperazines Derivatives against Malaria Based On PLSR Machine Learning Technique

Malaria is a mosquito-borne lifethreatening disease. Though it is treatable but due to emerging resistance against well-known drugs like 8aminoquinolines and artemisinin analogs, it is urgent need to develop novel drugs for the treatment of malaria. In order to develop a new drug, imidazolopiperazine derivatives have been shown antimalarial efficacy against wild-type as well as drug resistant malarial parasites. Hence, the experimental dataset (AID: 605733) of imidazolopiperazine series is a valuable resource to develop novel leads for antimalarial drug discovery. In this study, we conducted lead optimization using above dataset. For that, we carried out machine learning 3D QSAR modeling based on partial least square regression. For 3D QSAR modeling, firstly molecular descriptors were calculated using VLife Software followed by PLSR model evaluation by calculating statistical parameters. The value of correlation coefficient (r2 =0.9290) of 3D QSAR PLSR model indicated a strong positive correlation between biological activity and descriptors with high significance level. The high F test value (F= 69.83) suggested a low probability of the model’s failure. The regression model showed that steric (S_292) and electrostatic (E_284, and E_676) interactions play an important role in determining antimalarial activity. After that, we generated a computational library of imidazolopiperazine compounds using the template of core molecule i.e. imidazolopiperazine. The 3D QSAR PLSR model was applied on a computational library of compounds resulting top 10 novel antimalarial lead molecules. This work was extended by carrying out pharmacophore modeling of imidazolopiperazine scaffold which revealed that 5 pharmacophoric features: two aromatic, two aliphatic and one hydrogen bond donor are essential features for antimalarial activity. The Computational lead optimization may help in the design of novel antimalarial molecules. The information obtained from this could be applied for the future development of potent drugs against malaria.