Decoding G-quadruplex stability: The role of loop architecture and sequence context in the human genome.

Abstract

Guanine-rich sequences are widely distributed throughout the human genome and are capable of forming intramolecular G-quadruplex (G4) structures through Hoogsteen hydrogen bonding. These structures have been implicated in diverse regulatory processes. While extensive studies have established that loop architecture-particularly loop length and composition-profoundly affects G4 structural stability, most investigations have relied on synthetic sequences with predefined loop configurations that do not accurately reflect genomic contexts. In the current study, we analyzed the chain composition and stability of G-quadruplexes within the human genome to clarify the relationship between them by high throughput sequencing data. We utilized G4-forming sequences identified by G4-seq and G4-miner-two sequencing-based methods that detect G4s through polymerase stalling-associated drops in sequencing quality scores, where more stable structures produce stronger signals and thus higher detection rates-as the primary dataset. Our analysis revealed a negative correlation between total loop length and G4 stability, whereas individual loop length distributions exhibited minimal influence. Interestingly, G4s with short loops frequently occur in the genome as microsatellites or tandem atypical G4 arrays, resulting in structural stability profiles that deviate from those observed in synthetic G4 motifs in vitro. Molecular dynamics simulations incorporating native flanking sequences further corroborated these findings, underscoring the importance of genomic context in determining G4 stability. We note that the research was restricted to canonical G4s, which may limit the generality of our conclusions.