G4-mtQSAR – Multi-targeting G-quadruplex DNA
Title: Identifying new anti-cancer drugs by computational multi-target approaches targeting the G-quadruplex DNA
The proposed research aimed to develop computational methodologies to screen small ligands with the potential to selectively target cancer-associated G-quadruplex (G4) DNA. Since stabilization of oncogenic (c-myc, c-kit, k-ras, etc.) and telomeric G4s represented an effective strategy for targeted cancer therapy, the study focused on developing multi-target QSAR models to identify multi-target-directed ligands (MTDLs) capable of stabilizing multiple G4s simultaneously. This work represented the first computational effort to identify MTDLs against multiple G4s in cancer treatment. A novel approach was developed by integrating validated multi-target QSAR models and in-house ADMET models as a knowledgebase, which was subsequently used to screen potential MTDLs. The selectivity and binding characteristics of the screened ligands toward G4s over duplex DNA were analyzed using in silico and in vitro assays. This was the applied workflow:
Chemical and biological data were collected from literature and extensively curated, including structure correction, normalization, duplicate handling, and activity-cliff analysis.
Multi-target QSAR models were developed for different G4 types using regression and classification approaches, diverse molecular descriptors, and linear and non-linear chemometric techniques, and were validated following OECD guidelines.
A software tool with KNIME nodes and workflows was developed to screen, optimize, and design MTDLs against G4s.
Virtual screening of large chemical databases was performed using desirability-based multi-objective optimization, followed by evaluation of screened ligands through molecular docking, molecular dynamics simulations, and key biophysical assays.
Period: 2021-2023
Program: Marie Skłodowska-Curie Individual Fellowships
Financed by: H2020-MSCA-IF-2020