Sorry for troubling you again...
qiime fragment-insertion filter-features --i-table merged_table.qza --i-tree insertion-tree.qza --o-filtered-table filtered_table.qza --o-removed-table removed_table.qza
The filtered_table.qza seemed the same as the original merged_table.qza. Is it a normal condition?
I also have tried other datasets in another project, the results turned out to be the same... No change has happened...
The description of the official document:
Usage: qiime fragment-insertion filter-features [OPTIONS]
Filters fragments not inserted into a phylogenetic tree from a feature-
table. Some fragments computed by e.g. Deblur or DADA2 are too remote to
get inserted by SEPP into a reference phylogeny. To be able to use the
feature-table for downstream analyses like computing Faith's PD or
UniFrac, the feature-table must be cleared of fragments that are not part
of the phylogenetic tree, because their path length can otherwise not be
determined. Typically, the number of rejected fragments is low (<= 10),
but it might be worth to inspect the ratio of reads assigned to those
rejected fragments.
Therefore, I believe that the this step is supposed to filter some features...
I have observed a similar question in the forum: no change after qiime fragment-insertion filter-features - #3 by Lennon_Lee
But I am still confused about it.