AniAMPpred

AniAMPpred predicts antimicrobial peptides (AMPs) across animal genomes using machine learning and sequence analysis to identify probable antimicrobial proteins (PAPs).

Key Features:

• Machine Learning Integration: Combines support vector machine (SVM) classification with deep learning-derived sequence features to predict antimicrobial activity.
• Comprehensive Dataset Utilization: Trained on a curated dataset of known AMPs from animal species spanning 10–200 amino acids.
• High Predictive Accuracy: Demonstrated superior performance compared to existing classifiers in distinguishing AMPs from non-AMPs.
• Genome-Wide Application: Applicable genome-wide to identify PAPs across animal species, including identification of 436 PAPs in Helobdella robusta.
• Functional Validation Support: Supports functional validation via BLAST analysis against known AMP databases.

Scientific Applications:

• Antimicrobial Research: Identification of novel AMPs for investigation as antimicrobial agents against antibiotic-resistant pathogens.
• Protein Function Annotation: Annotation of protein functions within genomic databases to reveal roles in host defense mechanisms.
• Evolutionary Studies: Analysis of evolutionary conservation and divergence of AMPs across different animal species.

Methodology:

Model training used a curated dataset of known AMPs (10–200 amino acids); features were derived via deep learning and classified using a support vector machine (SVM), and predicted proteins can be validated by BLAST against AMP databases.