Structure-based prediction of post-translational modification cross-talk within proteins using complementary residue-...

Area of Science:

• Biochemistry
• Bioinformatics
• Computational Biology

Background:

• Post-translational modifications (PTMs) regulate protein function through single or multiple modifications.
• Predicting the interplay between different PTMs (PTM cross-talk) is crucial but remains challenging.
• Limited understanding of structural preferences and residue characteristics hinders accurate PTM cross-talk prediction.

Purpose of the Study:

• To develop an improved algorithm for predicting PTM cross-talk.
• To incorporate structural and residue-level features for enhanced prediction accuracy.
• To provide a reliable tool for analyzing PTM cross-talk.

Main Methods:

• Developed PCTpred, a structure-based algorithm for PTM cross-talk prediction.
• Designed comprehensive residue- and residue pair-based features at sequence and structural levels.
• Integrated sequence-based (PCTseq) and structure-based (PCTstr) predictors into a final model.

Main Results:

• PCTpred achieved high prediction performance with AUC values of approximately 0.9 (pair-based) and 0.8 (protein-based).
• Prediction accuracy remained stable even with simplified input or removal of distance preferences.
• Demonstrated significant improvements over existing state-of-the-art methods.

Conclusions:

• PCTpred offers a robust and reliable approach to PTM cross-talk prediction.
• The algorithm's performance highlights the importance of structural and residue-level features.
• Source code and data are publicly available for further research and application.