AUTHOR=Li Yanjuan , Ma Di , Chen Dong , Chen Yu 

TITLE=ACP-GBDT: An improved anticancer peptide identification method with gradient boosting decision tree

JOURNAL=Frontiers in Genetics

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2023.1165765

DOI=10.3389/fgene.2023.1165765

ISSN=1664-8021

ABSTRACT=Cancer is one of the most dangerous diseases, it annually kills millions of people globally. Anticancer peptides can treat cancer with small side effects. Therefore, identifying anticancer peptide has been a hot topic. In this study, we proposed an improved anti-cancer peptide predictor named ACP-GBDT to identify anticancer peptide based on sequence information. ACP-GBDT employed mixed-feature of AAIndex and SVMProt-188D to encode peptide sequences, and adopted GBDT to train the prediction model. The mixed-feature considers both the amino acid composition information of peptide sequence and the physicochemical properties of amino acid, it can effectively distinguish anticancer peptides from non-anticancer ones which is verified by experiments of independent test and 10-fold cross-validation. Finally, the comparison results on benchmark dataset proved that the developed ACP-GBDT is more simple and effective in the identification of anticancer peptides than existing methods.