“Respiratory viral infection is associated with increased Pseudomonas abundance in the cystic fibrosis airway” Yasmin Hilliam, Catherine R. Armbruster, Samar E. Atteih, Glenn J. Rapsinski, Monika Johnson, John Moore, Junu Koirala, Leah Krainz, Jordan Gaston, John Williams, Vaughn S. Cooper, Stella E. Lee and Jennifer M. Bomberger. ERJ Open Res 2026; 12: 01022-2025.

This article was originally published with discrepancies between the main text and figure 2. “There is a significant increase in distance to pre-virus state for samples collected during viral infection (median=0.75, p<0.0001) indicating substantial variation in the bacterial community composition during viral infection. The increase in distance to pre-virus centroid for samples collected post-viral infection is less than during viral infection (median=0.37), but is still significantly different from the pre-virus samples (p=0.01), indicating that the composition of airway microbiota is not entirely restored to baseline following viral infection.” has been corrected to “There is a significant increase in distance to pre-virus state for samples collected during viral infection (median=0.49, p<0.05) indicating substantial variation in the bacterial community composition during viral infection. The increase in distance to pre-virus centroid for samples collected post-viral infection is not significantly different to that during viral infection (median=0.59), but is still significantly different from the pre-virus samples (p=0.01), indicating that the composition of airway microbiota is not entirely restored to baseline following viral infection.” In figure 2, the significance of the relationship between “pre” and “active” has been corrected from p<0.001 to p<0.05. The corrected figure is shown below. These corrections do not affect the overall conclusions of the study.

FIGURE 2.

Airway community composition is disrupted by viral infection and does not fully return to baseline state. Centroids were calculated from Morisita–Horn distances from all samples per subject, per viral status (prior, active, post). Here, all sample distances were compared to the pre-virus centroid for each subject with distance to centroid representing shifts in community composition as compared to the pre-virus state. Data were compared by mixed-effect linear model with subject as a random effect. Differences in distance to centroid were tested by ANOVA and post hoc Tukey test. ns: nonsignificant. *: p<0.05.