Identification of S-nitrosylation sites based on multiple features combination

Area of Science:

• Biochemistry
• Proteomics
• Bioinformatics

Background:

• Protein S-nitrosylation (SNO) is a vital post-translational modification regulating cell function and pathophysiology.
• Accurate identification of S-nitrosylation sites is essential for analyzing large biological datasets.
• Computational tools are needed to assist in the rapid and efficient identification of S-nitrosylation sites.

Purpose of the Study:

• To develop and evaluate a computational method for predicting protein S-nitrosylation sites.
• To utilize various sequence-derived features for enhanced prediction accuracy.
• To assess the method's performance using cross-validation and independent tests.

Main Methods:

• Employed multiple sequence features: PC-PseAAC, kmer1, kmer2, PC-PseAAC_G, ANBPB, DBPB, BPB, IAAPair, and PSTAAP.
• Utilized Information Gain (IG) to reduce feature redundancy and assess amino acid importance.
• Validated prediction performance using cross-validation and independent datasets.

Main Results:

• The developed computational method achieved high prediction accuracy.
• For the training dataset, Accuracy (Acc) was 83.11% and Matthew's Correlation Coefficient (MCC) was 0.6617.
• For the independent test dataset, Acc was 73.17% and MCC was 0.3788.

Conclusions:

• The proposed method demonstrates significant potential for identifying protein S-nitrosylation sites.
• This computational tool can complement existing prediction methods.
• The approach offers a valuable resource for effective S-nitrosylation site identification in biological research.