Combaning ASV tsble after decontam with qiime2 outputs into phyloseq object

General Discussion
1

Hello Everyone!
I have used package decontam to clean our table.qza from qiime2 output.
I want to create phyloseq object with qiime2 outputs like taxonomy,rooted-tree and AVS table decontaminated, that is already in R. I couldn't find anything about it.
I would appreciate any help.

2

Hello Alla,

Here are the docs on merging Phyloseq objects: Merge Data

When you merge two phyloseq objects, features not in both objects are dropped. So if you joined your decontam-filtered object with your full object, all features removed by decontam will also be dropped.

If decontam keeps features but changes their counts, you may want to merge a decontam table with an existing tree or something like that. In that case, you are basically making a new phyloseq object from the various parts you want to use, which is covered here: Importing phyloseq Data

3

Thank you)
I tried, but it didn't changed anything in data.
It's seems like I am working with the same data set from phyloseq I have.

mergedPilot = merge_phyloseq(phyloseq2, ASV_table_decontaminated_ps_noControl)
mergedPilot

mergedPilot
phyloseq-class experiment-level object
otu_table() OTU Table: [ 1698 taxa and 45 samples ]
sample_data() Sample Data: [ 45 samples by 8 sample variables ]
tax_table() Taxonomy Table: [ 1698 taxa by 7 taxonomic ranks ]
phy_tree() Phylogenetic Tree: [ 1698 tips and 1697 internal nodes ]

I though when R will merge ASV decontaminated non Controls with pyloseq object, which I did from qiime2 outputs, it will get rid of any data about controls, for example. But they are still there in plots and heatmaps.
Maybe I did something wrong.
I am just a beginner in R.

Here is phyloseq I did before merging

phyloseq-class experiment-level object
otu_table() OTU Table: [ 1698 taxa and 45 samples ]
sample_data() Sample Data: [ 45 samples by 6 sample variables ]
tax_table() Taxonomy Table: [ 1698 taxa by 7 taxonomic ranks ]
phy_tree() Phylogenetic Tree: [ 1698 tips and 1697 internal nodes ]

4

I though when R will merge ASV decontaminated non Controls with pyloseq object, which I did from qiime2 outputs, it will get rid of any data about controls, for example. But they are still there in plots and heatmaps.

Interesting! Have you tried building a new phyloseq object using the constructor classes? That would give you more control over exactly what goes into the object.