Atomistic simulations using molecular dynamics and current physical force fields have come of age to reproduce experimental structures with high accuracy. However, they require large computational resources that limit their use for studying slow biological processes such as folding and binding. We have developed a new method, MELD, for accelerating the sampling of those processes by using either sparse, ambiguous and/or experimental data or insights about proteins (e.g. globular proteins have hydrophobic cores). I will show how this method has been successful to predict binding poses of p53 related peptides to MDM2 and MDMX protein structures. Furthermore, we performed simulations of competitive binding between pairs of peptides to establish relative binding free energies between them in good agreement with experiment.

Relative binding free energies from flexible peptide binding simulations

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.