Evaluating the Martini-Gō Coarse-Grained Model for Peptide Conformational Study

Author: Erik Koshnicharski

Field of Study: Chemistry; Mathematical Sciences

Faculty Mentors: Zack Jarin

Easel: 17

Timeslot: Afternoon

Abstract: Peptides are biologically important because of their ability to regulate physiological processes. Experimental study of peptides is challenging due to their equilibrium between disordered and ordered states. Coarse-grained computational methods supplement experiment by using reduced representations of molecular structures (beads vs. atoms), omitting details that are computationally costly while enabling microsecond-scale simulations to probe peptide behavior. One popular coarse-grained model is Martini3, despite its trouble capturing changes in protein secondary structure. The Martini-Gō model utilizes an applied Gō potential to describe non-bonded backbone bead interactions, promoting native structure retention while allowing flexibility to capture multi-state behavior. Here, fifteen peptides of varying structural properties are simulated using Martini-Gō with varying applied Gō potentials. Peptide secondary structures were quantified using root mean square deviation and native contact analyses and compared to experimental data. Results show that the Gō potentials over-stabilize the folded structures, contradicting the known disorder exhibited by these peptides.