PhageAI - Bacteriophage Life Cycle Recognition with Machine Learning and Natural Language Processing

Abstract

As antibiotic resistance is becoming a major problem nowadays in a treatment of infections, bacteriophages (also known as phages) seem to be an alternative. However, to be used in a therapy, their life cycle should be strictly lytic. With the growing popularity of Next Generation Sequencing (NGS) technology, it is possible to gain such information from the genome sequence. A number of tools are available which help to define phage life cycle. However, there is still no unanimous way to deal with this problem, especially in the absence of well-defined open reading frames. To overcome this limitation, a new tool is definitely needed. We developed a novel tool, called PhageAI, that allows to access more than 10 000 publicly available bacteriophages and differentiate between their major types of life cycles: lytic and lysogenic. The tool included life cycle classifier which achieved 98.90% accuracy on a validation set and 97.18% average accuracy on a test set. We adopted nucleotide sequences embedding based on the Word2Vec with Ship-gram model and linear Support Vector Machine with 10-fold cross-validation for supervised classification. PhageAI is free of charge and it is available at https://phage.ai/. PhageAI is a REST web service and available as Python package. Machine learning and Natural Language Processing allows to extract information from bacteriophages nucleotide sequences for lifecycle prediction tasks. The PhageAI tool classifies phages into either virulent or temperate with a higher accuracy than any existing methods and shares interactive 3D visualization to help interpreting model classification results.

Genome Sequence and Characterization of Five Bacteriophages Infecting Streptomyces coelicolor and Streptomyces venezuelae: Alderaan, Coruscant, Dagobah, Endor1 and Endor2

“…These DTR were identified in the initial assembly by an approximately 2-fold increase in coverage clearly delimitated at single base positions. Phage lifecycle was predicted using PhageAI, which developed a lifecycle classifier based on machine learning and natural language processing [ 30 ].…”

“…The ends of the phage genomes were determined with PhageTerm [ 29 ] using default parameters. Phage lifecycle was predicted with PhageAI [ 30 ] using default parameters.…”

BACPHLIP: predicting bacteriophage lifestyle from conserved protein domains

“…Finally, we note that after completion of this work, another phage lifestyle classification tool came to our attention ( Tynecki et al, 2020 ). The PhageAI program uses machine learning and natural language processing to classify phage genome sequences, and is released as a web-based platform with an API that allows for programmatic access.…”

Isolation and Characterisation of Bacteriophages with Activity against Invasive Non-Typhoidal Salmonella Causing Bloodstream Infection in Malawi

“…Phage assemblies were annotated using Prokka v1.12 [ 43 ] with--proteins/blastdb/Viral_Genomes/all_viral.faa, visualised using SnapGene v5.2 (Insightful Science, snapgene.com) and manually curated ( Tables S2–S5 ). PhageAI [ 44 ] is a tool used to determine lifecycle of phages with high accuracy. Phage genome assemblies were analysed using PhageAI v0.1.0 LifeCycleClassifier using default parameters.…”