search

thomasgurry / amplicon_sequencing_pipeline

Pipeline for computing OTU tables from 16S or ITS amplicon sequencing data.
MIT License
32 stars 16 forks source link

How can we calculate the beta diversity of unweighted_unifrac, weighted_unifrac? #2

Open xiaoyandong08 opened 5 years ago

xiaoyandong08 commented 5 years ago

Dear all, thanks for your fantastic work. How can we calculate the beta diversity of unweighted_unifrac, weighted_unifrac from your pipeline?

thomasgurry commented 5 years ago

Hello,

Thanks for your kind note. I'm afraid the pipeline stops at OTU tables, so it doesn't allow the calculation of beta diversity. If you want to do unifrac type calculations, you can use the *.otu_seqs.fasta file for the sequences of the 16S and build a phylogenetic tree using FastTree. You can then do the unifrac calculations using the skbio Python package (skbio.diversity), which I believe has documentation online.

Hope this helps!

Cheers, Thomas

On Thu, 11 Apr 2019 at 17:47, Yandong Xiao notifications@github.com wrote:

Dear all, thanks for your fantastic work. How can we calculate the beta diversity of unweighted_unifrac, weighted_unifrac from your pipeline?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/thomasgurry/amplicon_sequencing_pipeline/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/ASpvlznbsagAnzs4XmOvcxMNBtqmTqJkks5vf1kqgaJpZM4cp78a .

xiaoyandong08 commented 5 years ago

Thank you very much! How can I apply PICRUSt for the OTU tables generated by your pipeline to infer the function profile?

thomasgurry commented 5 years ago

I think for PiCrust you need the closed reference greengenes table, where OTUs have Greengenes IDs. Also you need the table to be in biom format (can use the biom-convert tool available online). Probably easier to use QIIME to generate the OtU table if you want to do PiCrust though.

On 12 Apr 2019, at 10:35, Yandong Xiao notifications@github.com wrote:

Thank you very much! How can I apply PICRUSt for the OTU tables generated by your pipeline to infer the function profile?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.