Metagenomic sequencing detects viruses and bacteria in a cross-sectional clinical cohort of undifferentiated febrile illness in Nigeria.

Molecular and microscopy-based diagnostic capacity is often insufficient or unavailable in places where infectious disease burdens are highest, such as in West Africa. Rapid diagnostic testing (RDT) can provide quick and affordable diagnoses of common infections but is an imperfect solution due to limitations around detecting and dealing with false-negative and false-positive results. An alternative to RDT is unbiased metagenomic sequencing for pathogen surveillance. Here, we present data from unbiased metagenomic sequencing used to identify causes of undiagnosed febrile illness in Jos, Plateau State, Nigeria. Proof of concept for this approach has been demonstrated by several groups who have identified epidemic and endemic viral diseases like Lassa fever, yellow fever, and chikungunya. We show that unbiased deep sequencing and metagenomic analysis can be used to identify RNA viruses in clinical samples. We sequenced RNA from sera of patients ( IMPORTANCE: In low-resource areas, fevers due to infectious pathogens are a major source of illness, but tools for detecting and identifying such pathogens are often limited. Unbiased approaches for identifying genetic material from all potentially infectious organisms in a sample represent an opportunity for discovering sources of fever. Metagenomic sequencing can improve insight into pathogen landscapes in low-resource settings, potentially providing early detection of disease outbreaks. However, unbiased metagenomic sequencing (mNGS) is no panacea; it is susceptible to contamination and false positives. We used mNGS to evaluate serum from >300 Nigerian clinic-goers in Jos, Nigeria, most of whom (>70%) had fevers of unknown origin. Our goal was to understand arbovirus prevalence in Jos, Nigeria, and identify the sources of infection not routinely monitored for at clinics. We detected hepatitis B virus, as well as nonpathogenic anelloviruses. Our study provides insight into the utility and limitations of mNGS for pathogen surveillance.