iDPPIV-SI: identifying dipeptidyl peptidase IV inhibitory peptides by using multiple sequence information

Area of Science:

• Biochemistry
• Computational Biology
• Pharmacology

Background:

• Diabetes mellitus poses a significant global health challenge.
• Dipeptidyl peptidase IV (DPP-IV) inhibitory peptides show promise as pharmaceutical agents for diabetes treatment.
• Accurate discrimination between DPP-IV inhibitory and non-inhibitory peptides is crucial for drug development.

Purpose of the Study:

• To develop a novel computational model, iDPPIV-SI, for identifying DPP-IV inhibitory peptides.
• To enhance the accuracy and efficiency of DPP-IV inhibitory peptide prediction.
• To provide a reliable tool for researchers in the field of diabetes therapeutics.

Main Methods:

• Utilized 50 physicochemical (PC) properties to represent peptide sequences.
• Applied 1-order, 2-order correlation methods, and discrete wavelet transform for feature extraction.
• Employed the least absolute shrinkage and selection operator (LASSO) for feature selection.
• Integrated selected features into a support vector machine (SVM) classifier.

Main Results:

• The iDPPIV-SI model achieved 91.26% accuracy on the training dataset.
• The model demonstrated 98.12% accuracy on an independent dataset.
• The proposed method significantly improved classification performance compared to existing state-of-the-art predictors.

Conclusions:

• The iDPPIV-SI model is a highly effective computational tool for identifying DPP-IV inhibitory peptides.
• This approach offers a significant advancement in the prediction of potential diabetes therapeutic agents.
• The developed model and associated code are publicly available to facilitate further research.