gavinmdouglas
commented
1 year ago Hi Kinga,
Thanks for reporting this warning (or possible error?). It's hard to say whether there is a problem with the data, although I don't think so as this is a FutureWarning (i.e., a functionality will be changing). Would you mind providing me with the output of conda list in the environment you're working with? I think this is a new warning in newer versions of pandas.
Yes, you would only get the norm_taxon_function_contrib column with the pathway output if you use the --per_sequence_contrib option. And yes the interpretation would be the same as for the EC contributional output when this option is used. When the option is not used, pathway abundances correspond to the "community-wide" pathways. Essentially assuming that all genes / reactions can interact freely, regardless of which taxon encodes them. This is a very common assumption in microbiome data analysis, although I think it is seriously flawed.
Cheers,
Gavin
kingabo
Hi!
I run picrust2.5.1 using this code:
place_seqs.py -s seq.fa -o out.tre -p 30 --intermediate intermediate/place_seqs
hsp.py -i 16S -t out.tre -o marker_predicted_and_nsti.tsv.gz -p 30 -n
hsp.py -i EC -t out.tre -o EC_predicted.tsv.gz -p 30
metagenome_pipeline.py -i gr.biom -m marker_predicted_and_nsti.tsv.gz -f EC_predicted.tsv.gz -o EC_metagenome_out --strat_out
pathway_pipeline.py -i EC_metagenome_out/pred_metagenome_contrib.tsv.gz -o pathways_out -p 1
After the analysis I got this warning:
/soft/picrust2/picrust2-2.5.1/picrust2/pathway_pipeline.py:642: FutureWarning: The operation <function sum at 0x1489d7756160> failed on a column. If any error is raised, this will raise an exception in a future version of pandas. Drop these columns to avoid this warning.
Finally I got the unstratified and the stratified output, but I am bot sure if I can use it. Is it valid?
I also want to ask about the stratified table with predicted MetaCyc pathways. In the tutorial it is said that pathways table is similar to the EC table. However, the table with predicted pathways lacks the last column "norm_taxon_function_contrib" which gives proportional contribution of EC . Is this because such information is calculated only if I use --per_sequence_contrib option? If I use this option and I get the stratified table with pathways per sequence, is the last column also the proportional contribution of the given pathways? Is this last column analogous to the norm_taxon_function_contribution? Could you please clarify this?
Many thanks, Kinga