NOMAD: metagenomic characterisation of the viral pathogen composition in outbreaks of non-malaria acute febrile illness cases

Abstract

The clinical importance of non-malaria febrile acute illness (NM-AFI) in patients with a negative parasitological test has become apparent, with the progressive reduction in malaria transmission in endemic regions. Bacterial pathogens, for example Streptococcus pneumoniae and Haemophilus influenzae, which contribute disproportionally to febrile illness, are now preventable by vaccines. However, there are no vaccines, and little is known about viral NM-AFI prevalence, proliferation, virulence, and transmission chains between hosts. Although the predominant viral causes of NM-AFI are established, it is unclear if there are other NM-AFI associated emerging infectious viral pathogens that previously remained undetectable by conventional diagnostic strategies, for example severe acute respiratory syndrome coronavirus 2 (SARS-CoV-​2). Presumptive broad-spectrum antibiotic prescriptions to aparasitaemic patients not only drive drug resistance, but also lead to poor treatment outcomes. We hypothesized that insights on NM-AFI etiology, and consequently case management, could be improved by exploiting viral sequence diversity to identify viral pathogens present within metagenomics samples. We exploited simulated and existing infectious disease (Ebola, hepatitis C, chikungunya, and mosquito-borne arboviruses) metagenomic datasets to determine the composition of viral pathogens present, by implementing profile Hidden Markov Models derived from Swiss-Prot viral reference sequences for accurate pathogen detection and classification. Our analysis identified a combination of sequences from multiple viral etiological agents within the same disease sample. This approach provides a granular perspective of multiple viral etiological agents present within a single intra-host disease episode. It highlights prevalent viral strains that can subsequently be routinely detected using directed diagnostic tests to improve disease surveillance in endemic regions.