dForml(KNN)-PseAAC: Detecting formylation sites from protein sequences using K-nearest neighbor algorithm via Chou's...

Area of Science:

• Biochemistry
• Computational Biology
• Proteomics

Background:

• Post-translational modifications (PTMs) like protein formylation on lysine residues are vital for cellular functions.
• Accurate identification of formylation sites is essential for understanding these biological mechanisms.
• Existing experimental methods for site identification are time-consuming and less efficient than computational approaches.

Purpose of the Study:

• To develop the first computational predictor, LFPred, for identifying lysine formylation sites in proteins.
• To provide a faster and more convenient alternative to experimental methods for formylation site prediction.

Main Methods:

• Utilized sequence features including amino acid composition (AAC), binary profile features (BPF), and amino acid index (AAI).
• Employed the K-nearest neighbor (KNN) algorithm as the classification model.
• Implemented a discrete window approach based on information entropy and addressed data imbalance with carefully selected negative samples.

Main Results:

• The LFPred predictor achieved a specificity of 79.9% and a sensitivity of 81.4% in jackknife validation.
• Demonstrated the effectiveness of combining sequence features and the KNN algorithm for lysine formylation prediction.

Conclusions:

• LFPred serves as a valuable computational tool for accurate prediction of lysine formylation sites.
• The developed method offers a significant advancement in the study of protein formylation and its biological implications.