Integrating bacterial and viral genomic information enhances and extends the machine learning predictions of bacteriophage activity against pathogenic Escherichia coli
The use of bacteriophage (phage) to treat bacterial infections is undergoing a resurgence due to the rise of antibiotic resistance. Pathogenic Escherichia coli, including extraintestinal pathotypes such as uropathogenic E. coli (UPEC), pose a substantial clinical burden with the challenge of multi-drug resistance strains, making it important to progress alternative treatment options such as those…
Phage therapy is regaining attention as a solution to antibiotic-resistant bacterial infections, particularly pathogenic Escherichia coli strains like uropathogenic E. coli (UPEC). Predicting phage activity against unseen E. coli strains relies on machine learning models that incorporate bacterial genome data. However, these models lacked phage gene content, hindering their ability to generalize.
By integrating bacterial and phage pangenomes into a single predictive model, researchers have enhanced the accuracy of predicting phage-E. coli interactions. This unified approach, dubbed PanPhage, incorporates gene content from both phage and bacteria as machine learning features. The model's predictive prowess was demonstrated through leave-phage-out analyses, accurately predicting activity for phages with no training data.
Key bacterial determinants influencing prediction included surface recognition and anti-phage defense mechanisms, alongside phage-specific factors. PanPhage represents a significant advancement in capturing genetic information relevant to various stages of the phage-host interaction.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
