Detecting local residue environment similarity for recognizing near-native structure models.

Area of Science:

• Structural bioinformatics
• Computational biology
• Protein structure analysis

Background:

• Understanding local amino acid environments is crucial for protein structure prediction.
• Existing methods may not fully capture the nuances of residue-level structural context.
• Accurate scoring functions are needed to evaluate the quality of protein models.

Purpose of the Study:

• To introduce a novel representation for local amino acid environments in proteins: the Side-chain Depth Environment (SDE).
• To develop a computational procedure, PRESCO (Protein Residue Environment SCOre), utilizing SDEs for assessing protein model quality.
• To evaluate PRESCO's performance in identifying native or near-native protein models.

Main Methods:

• Defined the Side-chain Depth Environment (SDE) based on residue proximity and depth relative to the side-chain centroid.
• Developed the PRESCO scoring function to compare SDEs in a query model against a database of native protein structures.
• Benchmarked PRESCO against existing scoring functions using standard computational model datasets.

Main Results:

• Demonstrated that SDEs are general and can be found across proteins with diverse global folds.
• Showed that PRESCO effectively quantifies the 'native-likeness' of residue environments in computational models.
• PRESCO performance favorably compared to existing scoring functions in selecting native and near-native models.

Conclusions:

• The Side-chain Depth Environment (SDE) offers a robust representation of local protein structural context.
• PRESCO, powered by SDEs, provides an effective and accurate method for scoring protein models.
• This approach enhances the ability to identify high-quality protein structures from computational modeling efforts.