Appending Greengenes taxonomy to picrust .biom files instead of the map IDs

Hi
I have successfully used mothur to generate picrust formatted .biom files and used picrust thereafter. Works like a charm. However, during working i felt some tweaks may be very useful for mothur in terms of flexibility and have some questions.

  1. The picrust option in make.biom merges all the greengenes taxonomy appended to the OTUs in the study and outputs a .biom file here these merged taxonomies are presented as greengenes map IDs. Is there any way i can append the greengenes taxonomy ID to these ‘merged’ files? Like a reverse of the picrust option? I think currently mothur only allows greengenes taxonomy to be appended to the OTUs and not this merged OTU table. This will be an extremely helpful feature for the future. Can anyone suggest a workaround for now?
  2. Also, many a time we may work with different amplicons for the same gene to predict OTUs. These may result in similar taxonomic affiliations for OTUs for different studies. It will be very useful to have options to merge the “unduplicated, taxonomy table” mentioned above for the picrust .biom file with other.biom files. QIIME currently has a merge_otu_tables.py script for this.
  3. Is there any command to generate relative abundance taxonomy files from the OTU taxonomy files with absolute counts? Also, it will be very helpful to merge .lefse files as well.
    Please help.
  1. The picrust option in make.biom merges all the greengenes taxonomy appended to the OTUs in the study and outputs a .biom file here these merged taxonomies are presented as greengenes map IDs. Is there any way i can append the greengenes taxonomy ID to these ‘merged’ files? Like a reverse of the picrust option? I think currently mothur only allows greengenes taxonomy to be appended to the OTUs and not this merged OTU table. This will be an extremely helpful feature for the future. Can anyone suggest a workaround for now?

You would have to do this yourself. If I’m understanding your question correctly, why not just go back through make.biom to add the IDs?

  1. Also, many a time we may work with different amplicons for the same gene to predict OTUs. These may result in similar taxonomic affiliations for OTUs for different studies. It will be very useful to have options to merge the “unduplicated, taxonomy table” mentioned above for the picrust .biom file with other.biom files. QIIME currently has a merge_otu_tables.py script for this.

We won’t be doing this. I think mixing data from different regions is a really bad idea. We know that the primers used for each region have different biases. We also know that studies vary in how they do their sample prep and sequencing. Combining data across studies is a very dangerous approach. You would be best to analyze the datasets separately for the phenomenon you are looking for and then determine whether the results support each other across studies.

  1. Is there any command to generate relative abundance taxonomy files from the OTU taxonomy files with absolute counts? Also, it will be very helpful to merge .lefse files as well.

Why not use get.relabund on a shared file from the phylotype command? I’m not sure why you would merge .lefse commands.


Pat

Thank you Pat.