Nov 28 – 30, 2025
Centro Universitario Regional Litoral Norte, sede Salto (Gral. Rivera 1350)
America/Montevideo timezone

GeeMELD: In silico folding of DNA G-quadruplex guided by Circular Dichroism and pairing rules

Not scheduled
20m
Centro Universitario Regional Litoral Norte, sede Salto (Gral. Rivera 1350)

Centro Universitario Regional Litoral Norte, sede Salto (Gral. Rivera 1350)

Expositor (Póster) Diagnóstico y Terapia Sesión de posters 1

Speaker

Dr Pablo D. Dans (Universidad de la Republica)

Description

The G-quadruplex (G4) is a secondary structure adopted by both DNA and RNA, formed by the stacking of guanine tetrads stabilized by hydrogen bonds and monovalent cations such as K⁺ and Na⁺. These structures play key roles in gene regulation, DNA repair, stress signaling, and the formation of biomolecular condensates. G4s are also implicated in theories of cellular aging—by contributing to genome structural stabilization—as well as in neurodegenerative disorders such as Alzheimer’s disease. G4s also stand out as promising pharmacological targets (and scaffolds for new drugs) in oncologic, neurodegenerative, and viral contexts. However, structural knowledge of G4s remains limited: high-resolution techniques (NMR, X-ray crystallography, cryo-EM) have resolved atomic details for approximately 500 unique G4s deposited in the PDB, in stark contrast to more than 200,000 available protein structures. These methods are costly, difficult to access, and do not enable systematic structure generation.

To address these limitations, we developed GeeMELD, which plays on the English exclamation “Gee!” and the pronunciation of the letter G—evoking G-quadruplexes—while fusing the MELD (Modeling Employing Limited Data) algorithm with circular dichroism data to unravel their folding and dynamics. GeeMELD addresses a significant challenge in structural biology, with the potential to substantially enhance our understanding of biological processes where G4 formation is relevant and to inform the design of new therapeutic agents. For DNA G-quadruplex, GeeMELD outperforms sophisticated deep-learning approaches like Alphafold 3, providing, in addition, cation-specific insights on conformational sub-states and dynamics.

¿Desea que su trabajo sea considerado para presentación oral?
Documento de identidad 16951738
Fecha de nacimiento 23/05/1972
País de residencia Uruguay
Ciudad de residencia Salto
Título académico Doctor en Química (UdelaR)
Posición académica Investigador/a
Otra posición académica: Responsable del grupo de Biofísica Computacional, Prof. Agregado (DT)

Primary author

Dr Pablo D. Dans (Universidad de la Republica)

Co-authors

Mr Jokent Gaza (Universidad de Florida) Mr Santiago Pintos (Universidad de la Republica) Dr Ari Zeida (Universidad de la Republica) Dr Victoria Calzada (Universidad de la Republica) Ms Gabriela da Rosa (Universidad de la Republica) Dr Alberto Perez (Universidad de Florida)

Presentation materials

There are no materials yet.