PARP1 inhibitors discovery: innovative screening strategies incorporating machine learning and fragment replacement
Jiahui Tu1, Jiaqi Chen2, Nan Zhou2
1The First Clinical College, Guangdong Medical University, Zhanjiang, 524023, China.
Abstract:
PARP1, the most prominent member of the PARP family, mediates DNA repair and cellular stress responses. PARP inhibitors (PARPi) show clinical promise in treating BRCA1/2-mutated or homologous recombination-deficient tumors, particularly in breast and ovarian cancers. However, acquired resistance remains a significant therapeutic challenge. This study developed a PARP1 inhibitor discovery pipeline integrating machine learning with conventional virtual screening methods. We introduced a novel strategy called fragment replacement to generate new compounds with optimized properties. Using the Maybridge compound library, we developed machine learning models to predict inhibitor activity. The Random Forest classifier demonstrated superior performance (AUC = 0.971, accuracy = 0.915) in tenfold cross-validation. This machine learning-driven approach outperformed conventional virtual screening in terms of efficiency. Subsequently, we conducted virtual screening using 2D fingerprints, shapes, and docking to retain the top-ranked ligands based on a standardized score (Z2-score). XP docking and ADMET prediction were used to select two molecules with strong drug-like properties. Fragment replacement was employed to reconstruct 101 new compounds with improved drug-like characteristics and increased activity. After validation, we identified three hits with docking scores between - 11.802kcal/mol and - 10.808kcal/mol, which were superior to the positive control Talazoparib (docking score: - 9.103kcal/mol). MD simulations assessed the binding stability of the compounds to proteins, with all three selected compounds exhibiting good binding stability, thus identifying them as potential candidates for development as PARP1 inhibitors.
Insights
Researchers developed a machine learning pipeline to discover novel PARP1 inhibitors, outperforming traditional methods. This approach identified promising drug candidates with enhanced activity and stability for cancer therapy.
Area of Science:
- Oncology
- Medicinal Chemistry
- Computational Biology
Background:
- Poly (ADP-ribose) polymerase 1 (PARP1) is crucial for DNA repair and cellular stress.
- PARP inhibitors (PARPi) are effective against BRCA1/2-mutated cancers but face resistance.
- Developing novel PARP1 inhibitors is critical for overcoming therapeutic challenges.
Purpose of the Study:
- To create an efficient PARP1 inhibitor discovery pipeline using machine learning and virtual screening.
- To identify novel compounds with optimized drug-like properties and high inhibitory activity.
- To validate potential PARP1 inhibitors through molecular docking and simulations.
Main Methods:
- Integrated machine learning models (Random Forest) with conventional virtual screening.
- Employed a fragment replacement strategy to generate novel compounds.
- Utilized 2D fingerprints, shape screening, docking (XP), ADMET prediction, and molecular dynamics (MD) simulations.
Main Results:
- The Random Forest model achieved high performance (AUC=0.971, accuracy=0.915).
- The machine learning approach proved more efficient than conventional virtual screening.
- Identified three lead compounds with superior docking scores and good binding stability compared to Talazoparib.
Conclusions:
- The developed pipeline effectively identifies potent PARP1 inhibitors.
- Novel compounds with improved activity and drug-like properties were discovered.
- The identified compounds represent promising candidates for further development as PARP1-targeted cancer therapeutics.
Related Concept Videos
02:46Genetic Screens
Forward genetic screens
Forward or “classical” genetic screens involve creating random mutations in an organism’s DNA using radiation, mutagens, or insertion of additional bases, which result in visible changes...
01:29Pharmacogenomics: Identification of New Drug Targets