# the more permutations you request, the longer it takes # but also the more stable and precise your p-values become aitchison_perm % dist_permanova ( seed = 1234, # for set.seed to ensure reproducibility of random process n_processes = 1, n_perms = 99, # you should use at least 999! variables = "bmi_group" ) #> 10:04:14 - Starting PERMANOVA with 99 perms with 1 processes #> 10:04:14 - Finished PERMANOVA # view the permanova results perm_get ( aitchison_perm ) %>% as.ame ( ) #> Df SumOfSqs R2 F Pr(>F) #> bmi_group 2 104.0678 0.04177157 4.773379 0.01 #> Residual 219 2387.2862 0.95822843 NA NA #> Total 221 2491.3540 1.00000000 NA NA # view the info stored about the distance calculation info_get ( aitchison_perm ) #> psExtra info: #> tax_agg = "Family" tax_trans = "identity" dist_method = "aitchison"
# calculate distances aitchison_dists % tax_filter (min_prevalence = 0.1 ) %>% tax_transform ( "identity", rank = "Family" ) %>% dist_calc ( "aitchison" ) #> Proportional min_prevalence given: 0.1 -> min 23/222 samples.
0 kommentar(er)
