ACP-DA: Improving the Prediction of Anticancer Peptides Using Data Augmentation.

Area of Science:

• Biochemistry
• Computational Biology
• Drug Discovery

Background:

• Anticancer peptides (ACPs) show promise for cancer therapy.
• Experimental identification of ACPs is costly and time-consuming.
• Computational methods, particularly machine learning, are crucial for efficient ACP prediction.

Purpose of the Study:

• To develop an improved computational model for predicting anticancer peptides (ACPs).
• To address performance limitations in existing models due to insufficient sample sizes.
• To enhance the accuracy and efficiency of identifying novel ACPs for drug development.

Main Methods:

• Proposed ACP-DA model integrating binary profile and AAindex features for peptide representation.
• Employed feature space data augmentation to increase training sample size.
• Trained a machine learning model on augmented data for ACP prediction.

Main Results:

• The ACP-DA model demonstrated superior performance compared to existing methods.
• Data augmentation in the feature space significantly improved prediction accuracy.
• ACP-DA showed enhanced performance in predicting anticancer peptides over models without data augmentation.

Conclusions:

• ACP-DA offers a more effective computational approach for anticancer peptide prediction.
• Data augmentation is a valuable strategy for improving machine learning model performance with limited biological data.
• The developed model and code are available to facilitate further research in ACP discovery.