AUTHOR=Dong Gai-Fang , Zheng Lei , Huang Sheng-Hui , Gao Jing , Zuo Yong-Chun 

TITLE=Amino Acid Reduction Can Help to Improve the Identification of Antimicrobial Peptides and Their Functional Activities

JOURNAL=Frontiers in Genetics

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.669328

DOI=10.3389/fgene.2021.669328

ISSN=1664-8021

ABSTRACT=Antimicrobial peptides (AMPs) are considered as potential substitutes for antibiotics in the field of new anti-infective drug design. There has been several machine learning algorithms and web servers have been developed in identifying antimicrobial peptides and their functional activities. However, there is still room for improvement in prediction algorithms and feature extraction methods. The reduced amino acid alphabet effectively solved the problems of simplifying protein complexity and recognizing the structure conservative region. This article went into details about evaluating the performances of more than 5000 amino acid reduced descriptors generated from 74 types of amino acid reduced alphabet in the first stage and the second stage to construct an excellent two-stage classifier, iAMP-RAAC, for identifying antimicrobial peptides and their functional activities. The results showed that the 1st stage AMP classifier was able to achieve the accuracy of 97.21% and 97.11% for the training data set and independence test data set. In the second stage, our classifier still showed good performance. At least three of the four metrics of SN, SP, ACC, and MCC exceeded the calculation results in the literature. Further, the ANOVA with incremental feature selection (IFS) was used for feature selection to further improve prediction performance. After the feature selection of each stage, the classifier obtained the best results with the fewest features and the highest accuracy. At last, a user-friendly web server for iAMP-RAAC was established at http://bioinfor.imu.edu.cn/iampraac  providing predictive services for interested inquiring sequences.