EvolStruct-Phogly: incorporating structural properties and evolutionary information from profile bigrams for the...

Area of Science:

• Biochemistry
• Computational Biology
• Proteomics

Background:

• Post-translational modifications (PTMs) alter proteome function, impacting cell biology and pathogenesis.
• Phosphoglycerylation is a PTM of growing interest, but its experimental identification is challenging.
• Existing computational predictors for phosphoglycerylated lysines lack sufficient accuracy.

Purpose of the Study:

• To develop an improved computational method for predicting phosphoglycerylated lysine residues.
• To leverage structural and evolutionary information for enhanced prediction accuracy.

Main Methods:

• Developed EvolStruct-Phogly, a novel predictor utilizing amino acid structural and evolutionary data.
• Extracted features including accessible surface area, torsion angles, local structure conformations, and position-specific scoring matrix profile bigrams.
• Validated the predictor using a benchmark dataset of experimentally identified phosphoglycerylated and non-phosphoglycerylated lysines.

Main Results:

• EvolStruct-Phogly demonstrated significant performance improvements over previous predictors.
• Achieved a sensitivity of 0.7744, specificity of 0.8533, precision of 0.7368, accuracy of 0.8275, and MCC of 0.6242.
• The predictor effectively utilizes structural and evolutionary features for phosphoglycerylation site identification.

Conclusions:

• EvolStruct-Phogly represents a noteworthy advancement in predicting phosphoglycerylated lysine residues.
• The developed tool offers a more efficient and accurate alternative to experimental methods.
• The software and data are publicly available for further research and application.