Reply to Amarin et al., "Selection bias may compromise our understanding of the clinical significance of the co-detection of respiratory viruses".