Assessing the behavior of machine learning methods to predict the activity of antimicrobial peptides

This study demonstrates the importance of obtaining statistically stable results when using machine learning methods to predict the activity of antimicrobial peptides, due to the cost and complexity of the chemical processes involved in cases where datasets are particularly small (less than a few hu...

Full description