amPEPpy

amPEPpy predicts antimicrobial peptide (AMP) sequences from genome-scale data using a Python 3 implementation of a multi-threaded random forest classifier to identify candidate AMPs.

Key Features:

• Random Forest Classifier: Uses a random forest classifier implemented in Python 3 to classify sequence-derived descriptors for AMP prediction.
• Multi-threaded Processing: Supports multi-threaded computation to process genome-scale and large datasets.
• Global Protein Sequence Descriptors: Employs a distribution descriptor set derived from global protein sequence descriptors to represent peptide sequences.
• Training and Optimization Utilities: Includes utilities for training and optimizing random forest classifiers using novel training data.
• Genome-scale Data Handling: Designed to analyze genome-scale and metagenomic sequence data for AMP mining.

Scientific Applications:

• Antimicrobial Discovery: Identification of candidate antimicrobial peptides for downstream experimental validation.
• Antibiotic Resistance Research: Supports discovery of alternative antimicrobial agents relevant to addressing antibiotic resistance.
• Genomic and Metagenomic Screening: Mining of genome-scale and metagenomic datasets to detect putative AMP sequences.

Methodology:

Input genome-scale sequence data; calculate global protein sequence distribution descriptors; classify descriptors using a random forest classifier; output predicted AMP sequences.