Consensus Antimicrobial Peptides Identified via Three Prediction Models from Peptide Libraries of Cephalopods' Posterior Salivary Glands
Description
The antimicrobial peptide (AMPs) subsets presented herein have been obtained through a consensus prediction approach involving three distinct machine learning models. These models were employed on peptide libraries associated with 13 digestion protocols (doi:10.17632/c3zhzgwsnw.1), which were applied to the reported omics data of Cephalopods' Posterior Salivary Glands (doi: 10.17632/gxmkytwdhx.1). The provided dataset encompasses 13 peptide FASTA files, each annotated with the enzyme(s) involved in the in silico digestion process, categorised as using a single enzyme (OE), or two enzymes (TE) applied either sequentially (S) or concurrently (C). The identified AMPs represent an intriguing and unexplored chemical landscape, with the potential to drive the discovery of novel peptide-based pharmaceutical agents.
Files
Steps to reproduce
Antimicrobial peptides screening was conducted on Peptide Libraries from Cephalopods' Posterior Salivary Glands, available at doi: 10.17632/6fjsdnvygb.1, utilizing the following machine learning models/tools: 1. Macrel 2. modlAMP_RF 3. modlAMP_SVM Subsequent analysis involved the examination of prediction outputs from each model for every peptide library. Peptides consistently detected by all three models/tools were identified across each library, representing the consensus AMPs for their respective peptide collections.