Analysis of viral diversity in stool samples from infants and children with acute gastroenteritis in Kuwait using Metagenomics approach

Abstract Background Current molecular target-dependent methods are used to detect only known viruses. However, metagenomics based on next-generation sequencing (NGS) technique is a target-independent assay that enables simultaneous detection and genomic characterisation of all microorganisms present in a sample. In this study, we aimed to develop a metagenomics approach using NGS to identify and characterise viruses in stool samples from infants and children with Acute Gastroenteritis (AGE) in Kuwait. Methods We have investigated 84 stool samples from infants and children aged one month to ten years old with signs and symptoms of gastroenteritis who attended Mubarak Al-Kabeer and Al-Amiri hospitals in Kuwait from January to December 2017. A metagenomics approach using NGS to characterise viruses in clinical samples was used. Also, the commercial Real-Time PCR assay was used to detect viruses causing gastroenteritis. Results Metagenomics analysis revealed an average of 280,768 reads in which 5% of the reads were derived from viruses. The analysis of viral sequences verified that single infection of human adenovirus was the leading cause of gastroenteritis among infants and children, which was detected in 23.2% of the patients, followed by a mixed infection of human adenovirus and other viruses, which was detected in 20.9% of patients. Also, the newly discovered viruses known to cause gastroenteritis were detected, such as astrovirus MLB2, primate bocaparvovirus-1, Aichivirus A, cardiovirus, parechovirus A, astrovirus VA4, cosavirus-F, and bufavirus-3. Our results showed 71% agreement (k = 0.445, P = 0.000) between multiplex Real-Time PCR, which is used as a routine diagnostic test and metagenomics approach in the detection of viruses causing gastroenteritis in clinical samples. Conclusion Despite the difficulties in sample preparation and analysis process, we showed that metagenomics approach is a powerful and promising tool for the detection and characterisation of different viruses in clinical samples.