Random NAs inserted in the norm_taxon_function_contrib column

[Franek by]

Hi PICRUSt community,

 

I am just working through the revision of my manuscript and got into PICRUSt to spice up the data analysis and discussion. However, I came across some serious (?) issue which I cannot solve despite many attempts.

I run pathway pipeline including –per_sequence_contrib function to get the ASVs contribution to the pathway function:

pathway_pipeline.py -i EC_metagenome_out/pred_metagenome_unstrat.tsv.gz \

> -o pathways_out \

> --intermediate minpath_working \

> --per_sequence_abun EC_metagenome_out/seqtab_norm.tsv.gz \

> --per_sequence_function EC_predicted.tsv.gz \

> --per_sequence_contrib \

> -p 1

In the output file (path_abun_contrib.tsv.gz) the norm_taxon_function_contrib column has roughly 90% values denoted as NAs while some of the functions seems to be calcualted completely. Could you please let me know what could be the problem? I have tried also running the pathway pipeline with the input of EC stratified file but the result was the same.

 

Best regards

Franek

[Robyn Wright]

Hi Franek,

Could you maybe try running the full picrust2_pipeline.py with the --per_sequence_contrib option and see if you still run into the same issue? 

Thanks,

Robyn