MLACP: machine-learning-based prediction of anticancer peptides.

Area of Science:

• Computational Biology
• Bioinformatics
• Oncology

Background:

• Cancer is a leading global cause of death, necessitating novel therapeutic strategies.
• Anticancer peptides (ACPs) show promise for targeted cancer cell destruction.
• Experimental identification of ACPs is resource-intensive, demanding computational solutions.

Purpose of the Study:

• To develop and evaluate machine-learning models for predicting anticancer peptides (ACPs).
• To identify potential ACP candidates efficiently for subsequent experimental validation.
• To provide a publicly accessible tool for the scientific community.

Main Methods:

• Development of support vector machine and random forest models.
• Feature extraction from amino acid sequences: composition, physicochemical properties.
• Model training and parameter optimization using the Tyagi-B dataset and 10-fold cross-validation.

Main Results:

• The random forest model demonstrated superior performance compared to existing methods.
• Achieved an average accuracy of 88.7% and a Matthews correlation coefficient of 0.78.
• Validated performance on independent benchmarking datasets.

Conclusions:

• Machine-learning approaches, particularly random forest, are effective for predicting ACPs.
• The developed computational method significantly accelerates the identification of potential ACPs.
• A publicly accessible web server (www.thegleelab.org/MLACP.html) is available to aid research.