Computationally efficient approach for the identification of ice-binding surfaces and how they bind ice

Area of Science:

• Computational chemistry
• Biophysics
• Materials science

Background:

• Ice nucleation and growth are controlled by molecular recognition and binding to surfaces.
• Current computational methods like docking to identify ice-binding surfaces (IBS) have limitations, including requiring prior knowledge of ice planes and high computational cost.

Purpose of the Study:

• To develop and validate a robust, computationally efficient methodology for identifying IBS on molecules and crystals.
• To overcome the limitations of existing ice-docking approaches.

Main Methods:

• A novel methodology based on biased sampling using an order parameter that drives ice formation.
• Validation using all-atom and coarse-grained models of organic crystals and proteins.

Main Results:

• The new method identifies IBS and the corresponding ice plane simultaneously, without structure search algorithms.
• It is approximately 100 times more computationally efficient than current advanced ice-docking methods.
• The method can identify even small or weak surfaces that do not heterogeneously nucleate ice.

Conclusions:

• This biased simulation approach offers a significant advancement in identifying IBS for various materials, including antifreeze and ice nucleating proteins.
• It provides a faster and more comprehensive way to study ice-surface interactions and can be used to prepare systems for calculating ice nucleation rates.