Multi-scaled Modeling of Electrostatic and Polarization Effects in Biomolecules

University Of California-Irvine, Irvine CA

Investigators

Linked publications & trials

Abstract

PROJECT SUMMARY Atomistic molecular simulations provide a suite of testable observables that yield essential mechanistic insights into a variety of diseases, ranging from cancer to neurodegenerative disorders. These insights are vital for the development of therapeutic strategies. Yet, accurately modeling complex processes such as order-disorder transitions, ionic interactions, and events at the biomembrane interface remains a formidable challenge with current methods. This is largely due to the intricate modeling required for electrostatic and polarization effects across varying structural states and solvent environments – an endeavor that current approaches struggle to perform efficiently. Our hypothesis to address the accuracy requirement is that biomolecules situated in diverse chemical contexts are best modeled within a polarizable framework to ensure satisfactory transferability. Departing from traditional methods, our polarizable Gaussian Multipole (pGM) model represents charges and multipoles with Hermite-Gaussian functions, rather than the conventional delta functions. This advanced representation enhances the model's accuracy, self-consistency, and transferability. Nonetheless, a known drawback of polarization treatments is their tendency to compromise simulation efficiency. To address this, we are pioneering a multi-scaled framework that integrates all-atom polarizable, coarse-grained polarizable, and continuum polarizable models. These models are designed for consistent interaction within multi-scaled simulation methods, facilitating more efficient interfacing. Our plan can be summarized in the following four areas: (1) the development of pGM force fields; (2) the development of continuum pGM solvent models; (3) the development of coarse-grained pGM models; and (4) the validation and application of our computational models to biomedical challenges. We will apply our models to significant biomedical issues, with a particular focus on biomolecular recognition and its association with conformational changes and allostery. To ensure our models address the most pressing challenges, such as order-disorder transitions linked to biomolecular recognition, we will validate them against these complex phenomena. Our commitment extends to the annual release of new models and tools, aiming to provide a broad and enduring contribution to the biomedical research community.