BACPHLIP
BACPHLIP predicts bacteriophage lifestyle (temperate versus virulent) from phage genome sequences to inform phage ecology and phage–host interactions.
Key Features:
- Lifestyle Classification: Classifies phages as temperate (capable of lysogeny and genomic integration) or virulent (lytic) based on genomic signals.
- Protein Domain Detection: Identifies conserved protein domains associated with temperate lifestyles and uses domain presence as predictive features.
- Machine Learning Approach: Implements a Random Forest classifier trained on 634 phage genomes to learn domain–lifestyle associations.
- High Accuracy: Achieves 98% accuracy on an independent test set of 423 phages, compared to 79% accuracy reported for previous tools.
Scientific Applications:
- Ecosystem Role Analysis: Enables assessment of phage contributions to microbial community dynamics and ecosystem functions by identifying lifestyle.
- Host Evolution Studies: Supports investigation of phage-driven selective pressures on bacterial hosts and co-evolutionary processes.
Methodology:
Analyzes phage genomes for conserved protein domains and inputs domain presence patterns into a Random Forest classifier trained on 634 genomes, with performance evaluated on an independent test set of 423 phages (98% accuracy).
Topics
Details
- License:
- MIT
- Programming Languages:
- Python
- Added:
- 1/18/2021
- Last Updated:
- 1/29/2021
Operations
Publications
Hockenberry AJ, Wilke CO. BACPHLIP: Predicting bacteriophage lifestyle from conserved protein domains. Unknown Journal. 2020. doi:10.1101/2020.05.13.094805.