Toward insights on determining factors for high activity in antimicrobial peptides via machine learning

• DOI: 10.7717/peerj.8265
• Publication Date: 2019-12-20T04:36:08Z
• Authors: Hao Li; Chanin Nantasenamat
• Keywords: Lytic cycle; Polarity (international relations)

The continued and general rise of antibiotic resistance in pathogenic microbes is a well-recognized global threat. Host defense peptides (HDPs), component the innate immune system have demonstrated promising potential to become next generation effective against plethora pathogens. While effectiveness antimicrobial HDPs has been extensively experimental studies, theoretical insights on mechanism by which these function comparably limited. In particular, studies AMP mechanisms are limited number different investigated type peptide parameters considered. This study makes use random forest algorithm for classifying activity as well identifying molecular descriptors underpinning peptides. Subsequent manual interpretation identified important revealed that polarity-solubility necessary membrane lytic HDPs.