Skip to main content
. 2012 Sep 26;7(9):e45001. doi: 10.1371/journal.pone.0045001

Table 2. P-values derived from PCoA for comparisons between individuals and patient groups defined by culture status.

P-values
All sequences Pseudomonas and Burkholderia sequences excluded
Null hypothesis Coordinate 1 Coordinate 2 Coordinate 1 Coordinate 2
1Patients with Burkholderia (n = 8) and Pseudomonas (n = 14) cluster together. 0.0002 0.008 0.10 0.71
2Samples from individual patients with Pseudomonas (n = 14) cluster together (40 timepoints). 0.06 0.003 0.02 0.08
2Samples from individual patients with Burkholderia (n = 8) cluster together (18 timepoints). 0.04 0.13 0.11 0.64
2Samples from individual patients (n = 23) cluster together across time. 0.0003 0.0004 0.001 0.15
1

The average value for each coordinate of the PCoA was taken for each patient (collapsing across all timepoints) and the Wilcoxon test was used to evaluate the hypothesis that the distribution of the resultant average coordinates was the same for Pseudomonas- and Burkholderia-positive patients.

2

To evaluate whether patients from the same culture status group clustered together across time, we used the Krusall-Wallis test with a factor of patient ID in which samples from different timepoints were treated as repeated samples for each patient. Because patients who were culture positive for Pseudomonas and Burkholderia clustered separately, we performed the analysis on each patient group as well as a combined analysis on all patients (Pseudomonas-positive, n = 14; Burkholderia-positive, n = 8; culture negative, n = 1).

HHS Vulnerability Disclosure