Refined Bonded Terms in Coarse-Grained Models for Intrinsically Disordered Proteins Improve Backbone Conformations

Area of Science:

• Computational Biology
• Biophysics
• Protein Dynamics

Background:

• Coarse-grained models for intrinsically disordered proteins and regions (IDP/Rs) conventionally omit bonded potentials to increase flexibility.
• This simplification often leads to inaccurate backbone conformations in simulations.
• Accurate modeling of IDP/Rs is crucial for understanding their biological functions.

Purpose of the Study:

• To address the inaccuracy of backbone conformations in coarse-grained IDP/R models.
• To develop an improved coarse-grained model that enhances backbone accuracy while maintaining flexibility.
• To ensure the new model accurately predicts secondary structure propensity and liquid-liquid phase separation (LLPS).

Main Methods:

• Introduced residue-specific angular, refined dihedral, and correction map (CMAP) potentials.
• Derived these potentials from statistical analysis of a customized coil database.
• Integrated the new potentials into the existing Mpipi model to create the Mpipi+ model.

Main Results:

• The Mpipi+ model demonstrates improved backbone conformations compared to previous models.
• Significantly enhanced secondary structure propensity (SSP) prediction, validated against experimental chemical shifts.
• Successfully captured transient secondary structures in IDP/Rs.
• Preserved the characteristic liquid-liquid phase separation (LLPS) propensities of IDPs.

Conclusions:

• The Mpipi+ model offers a more accurate representation of IDP/R backbone structures.
• It provides a valuable tool for studying the dynamics and functions of intrinsically disordered proteins.
• The model successfully balances flexibility with conformational accuracy for IDP/R simulations.