Phagonaute: A web-based interface for phage synteny browsing and protein function prediction

Hadrien Delattre (1) , Oussema Souiai (1) , Khema Fagoonee (1) , Raphaël Guérois (2, 3) , Marie-Agnès Petit (1)

1 MICALIS - MICrobiologie de l'ALImentation au Service de la Santé
2 AMIG - Assemblage moléculaire et intégrité du génome
3 I2BC - Institut de Biologie Intégrative de la Cellule

Distant homology search tools are of great help to predict viral protein functions. However, due to the lack of profile databases dedicated to viruses, they can lack sensitivity. We constructed HMM profiles for more than 80,000 proteins from both phages and archaeal viruses, and performed all pairwise comparisons with HHsearch program. The whole resulting database can be explored through a user-friendly "Phagonaute" interface to help predict functions. Results are displayed together with their genetic context, to strengthen inferences based on remote homology. Beyond function prediction, this tool permits detections of co-occurrences, often indicative of proteins completing a task together, and observation of conserved patterns across large evolutionary distances. As a test, Herpes simplex virus I was added to Phagonaute, and 25% of its proteome matched to bacterial or archaeal viral protein counterparts. Phagonaute should therefore help virologists in their quest for protein functions and evolutionary relationships.

Mots clés: HSV1; Bacteriophages; Recombination functions; Archaeal viruses

Domaines: Sciences du Vivant [q-bio]

Hadrien Delattre, Oussema Souiai, Khema Fagoonee, Raphaël Guérois, Marie-Agnès Petit. Phagonaute: A web-based interface for phage synteny browsing and protein function prediction. Virology, 2016, 496, pp.42-50. ⟨10.1016/j.virol.2016.05.007⟩. ⟨hal-01357336⟩