Combine OTU tables of different samples into one and add metadata of mapping file

[Qing Zhang]

Hi, everyone!

I have reads from soil samples and I want to analyze the fungi community.

I had 83 samples in total and I processed the reads of each sample separately. And I have already finished quality filtering, picking otus, taxonomic assignment and make otu table. Then I want to use the script summarize_taxa_through_plots.py to do the downstream analysis. But in this step a combined OTU table is needed. How can I combined the individual OTU tables into a big OTU table (i.e. rows are OTUs and columns are samples)?

Could you please provide me some instructions to combine the OTU table? Everything you say will be welcome.

Thank you so much for your help!

Best wishes,

Qing

[Colin Brislawn]

Good morning Qing,

Great question! Yes, you will need a combined OTU table for these kinds of analysis. 

My recommendation is to combine your samples into a single seqs.fna file, directly after demultiplexing (so they have qiime labels) and directly before OTU picking (so that all the samples will end in a single table).

So the process goes like this with a [new step] in the middle:

quality filtering,  [ cat sample1.fna sample2.fna sample3.fna > all_seqs.fna ],  picking otus,  downstream analysis

Let me know if this helps!

Colin

P.S. Which OTU picking script did you use? I ask because if you used closed-ref OTU picking, you can merge all your OTU tables using this script. 

http://qiime.org/scripts/merge_otu_tables.html 

(Warning: That script only worked with tables make with close-ref OTU picking. If you used denovo or open-ref OTU picking, you cannot merge your OTU tables, because the new OTUs will be different in different tables.)