Re-evaluation of G-quadruplex propensity with G4Hunter

Area of Science:

• Genomics and Bioinformatics
• Molecular Biology
• Drug Discovery

Background:

• G-quadruplexes (G4) are non-canonical DNA structures with critical biological relevance.
• These structures are implicated in key biological processes and are promising drug targets.
• Existing G4 prediction tools suffer from false positives and false negatives.

Purpose of the Study:

• To develop and validate a novel algorithm, G4Hunter, for accurate prediction of G-quadruplex propensity.
• To assess the prevalence of G4-forming sequences in various genomes, including the human genome.

Main Methods:

• Development of the G4Hunter algorithm, considering G-richness and G-skewness.
• Validation using a dataset of 392 published sequences and experimental evaluation of 209 sequences via biophysical methods.
• Application of G4Hunter to the human mitochondrial genome and other species' genomes.

Main Results:

• G4Hunter provides a quadruplex propensity score, outperforming existing algorithms.
• Experimental validation confirmed the algorithm's accuracy.
• The human genome contains 2-10 times more stable G4-forming sequences than previously thought.

Conclusions:

• G4Hunter is a reliable tool for predicting G-quadruplex formation.
• The prevalence of G4 structures in the human genome is significantly underestimated.
• These findings expand the landscape of potential G4-based therapeutic targets.