Discovering Dually\nActive Anti-cancer Compounds with\na Hybrid AI-structure-based Approach

Cancer’s persistent growth often relies on its\nability to\nmaintain telomere length and tolerate the accumulation of DNA damage.\nThis study explores a computational approach to identify compounds\nthat can simultaneously target both G-quadruplex (G4) structures and\npoly(ADP-ribose) polymerase (PARP)1 enzyme, offering a potential multipronged\nattack on cancer cells. We employed a hybrid virtual screening (VS)\nprotocol, combining the power of machine learning with traditional\nstructure-based methods. PyRMD, our AI-powered tool, was first used\nto analyze vast chemical libraries and to identify potential PARP1\ninhibitors based on known bioactivity data. Subsequently, a structure-based\nVS approach selected compounds from these identified inhibitors for\ntheir G4 stabilization potential. This two-step process yielded 50\npromising candidates, which were then experimentally validated for\ntheir ability to inhibit PARP1 and stabilize G4 structures. Ultimately,\nfour lead compounds emerged as promising candidates with the desired\ndual activity and demonstrated antiproliferative effects against specific\ncancer cell lines. This study highlights the potential of combining\nArtificial Intelligence and structure-based methods for the discovery\nof multitarget anticancer compounds, offering a valuable approach\nfor future drug development efforts.