Group significance group means vs summarised taxa means

[Sam Noel]

HI,

I can see some differences in taxa abundances between my sample groups (High, Mid and Low) when I use summarise_taxa_through_plots and I wanted to test this with group_significance.py

I am wondering why the group means for the group_significance test are different from the summarised taxa?

What i did was run the summarised taxa script twice, once just on all the samples and once with the category -c sample group (High, Mid and Low).

I ran the group significance script on the L6 summarised biom file generated from all the samples using the same category as above. I am now comparing this to the L6 summarised taxa table from above.

For example

#OTU ID

High

Mid

Low

k__Archaea;p__Euryarchaeota;c__Thermoplasmata;o__E2;f__[Methanomassiliicoccaceae];g__vadinCA11

0.301903

0.268437

0.257573

vs

OTU

Test-Statistic

P

FDR_P

Bonferroni_P

High_mean

Mid_mean

Low_mean

k__Archaea;p__Euryarchaeota;c__Thermoplasmata;o__E2;f__[Methanomassiliicoccaceae];g__vadinCA11

1.115359

0.572536

0.915556

1

0.355990844

0.345151

0.3336768

Thanks

Sam