Application of\nMachine Learning in the Development\nof Fourth Degree Quantitative Structure–Activity Relationship\nModel for Triclosan Analogs Tested against Plasmodium\nfalciparum 3D7

Despite new therapies against malaria, this disease remains\none\nof the main causes of death affecting humanity. The phenomenon of\nresistance has caused concern as the drugs no longer have the same\nefficacy, forcing scientific research to develop new methods that\nprospect new molecules. With the advancement of artificial intelligence,\nit becomes possible to apply machine learning (ML) techniques in the\ndiscovery and evaluation of new molecules, by employing the quantitative\nstructure–activity relationship (QSAR), a classic method that\nuses regressions to create a model that allows identifying and evaluating\nnew drug candidates. This work combined QSAR with ML and developed\na supervised model that modeled a fourth degree polynomial equation\ncapable of identifying new drug candidates derived from the triclosan\ncompounda classic inhibitor of Plasmodium falciparum growth, the cause of severe malaria. The model produces an <i>R</i><sup>2</sup> greater than 80% for training and concurrent\ntesting, as well as a correlation index greater than 80% between the\ncalculated and experimental pEC<sub>50</sub> (negative logarithm of\nhalf maximal effective concentration) data. In addition, a web software\n(PlasmoQSAR) was created that allows researchers to calculate the\nEC<sub>50</sub> (half maximal effective concentration) of new molecules\nusing the developed analytical method.