Assessing Molecular Contacts Using Atom Environments Described by Ranked Lists

Area of Science:

• Computational chemistry
• Structural biology
• Bioinformatics

Background:

• Representing atomic and molecular environments is crucial for understanding protein structure and function.
• Existing methods may not fully capture the nuances of local atomic arrangements.

Purpose of the Study:

• To develop a novel method for representing atomic environments.
• To create a scoring function to assess the fitness of atoms within their local environment.
• To characterize atomic contact preferences.

Main Methods:

• Representing atomic environments as ranked lists of neighboring atoms by distance.
• Analyzing contact densities for origin-neighbor pairs stratified by rank.
• Developing a fitness score (FS) and contact preference (CP) metrics.
• Validating the FS score against the 3DRobot dataset.

Main Results:

• Contact densities exhibit regular and specific distributions when stratified by rank.
• The developed fitness score (FS) effectively discriminates between native protein structures and decoys in the 3DRobot dataset.
• Contact preferences (CP) provide insights into atomic interaction patterns.

Conclusions:

• The ranked neighbor list representation provides a robust way to characterize atomic environments.
• The fitness score (FS) is a powerful tool for structure assessment and quality control in structural biology.
• The developed methodology and code are valuable for the research community.