Search

Additional file 7. Distinguishing taxa between dust and excreta stratified by company. Linear discriminant analysis effect size was performed on the top 50 most abundant bacterial taxa (genus level) across all ages.

Additional file 10. Distinguishing taxa between companies (A and B) stratified by sample type (dust and excreta). Linear discriminant analysis effect size was performed on the top 50 most abundant bacterial taxa (genus level) across all ages.

Additional file 13. Genera that were significantly different between high and low-performance farms of company A in pooled excreta samples. The results are based on differences of mean abundance tested with Wilcoxon ranksum test. P-values are corrected with false discovery rate (q-value).

Additional file 2. Permutational multivariate analysis of variance (PERMANOVA) results showing the influence of sample type, company, age, performance and their interaction on the bacterial community structure. Df = degrees of freedom, R² = proportion of the variance explained by each model, PERMDISP = permutational analyses of multivariate dispersion.

Additional file 1. Details of meat chicken farms used in this study. Farms A1 to A4 belong to company A and Farms B1 to B4 belong to company B.

Additional file 12. Genera that were significantly different between high and low-performance farms of company B in dust samples. The results are based on differences of mean abundance tested with Wilcoxon rank-sum test. P-values are corrected with false discovery rate (q-value).

Additional file 14. Genera that were significantly different between high and low-performance farms of company B in pooled excreta samples. The results are based on differences of mean abundance tested with Wilcoxon ranksum test. P-values are corrected with false discovery rate (q-value).

Additional file 4. Taxonomic assignment of the bacteria at phylum level in dust and excreta stratified by company.