LETHAL ERROR: Requested feature <PEPTIDOGLYCANSYN-PWY> was missing or not stratified

Microbial community profiling HUMAnN
1

Hello,

I have an issue with humann_barplot with --last-metadata option. When I use the option --last-metadata in humann_barplot, I have an error message as follows,

LETHAL ERROR: Requested feature was missing or not stratified

In the abundance table, I checked there is PEPTIDOGLYCANSYN-PWY. I am not sure what “not stratified” means in the error message.

Without the option --last-metadata, it is working without an error.

Would someone let me know what the problem is?

Thanks.

2

This script assumes you have potential metadata rows at the top of your table (e.g. sample location, sample time, etc.) followed by the HUMAnN rows. The --last-metadata flag tells the script where the metadata ends based on the name of the last metadata feature. I am not sure why using/not using this flag affects your error, but the error you’re seeing means that there are no species stratifications for the pathway you selected, i.e. it only have a community total abundance.

This script requires a feature that looks like…

Pathway
Pathway|Species1
Pathway|Species2
Pathway|unclassified

(for example.) If you don’t have stratified rows for your feature of interest, i.e. rows with headers containing |, then the script can’t plot that feature.

3

Hi Eric,

Thanks for your response.

I think that I have stratification in the joined table.

The following is an example of PEPTIDOGLYCANSYN-PWY.

PEPTIDOGLYCANSYN-PWY: peptidoglycan biosynthesis I (meso-diaminopimelate containing)
PEPTIDOGLYCANSYN-PWY: peptidoglycan biosynthesis I (meso-diaminopimelate containing)|g__Absiella.s__Absiella_dolichum
PEPTIDOGLYCANSYN-PWY: peptidoglycan biosynthesis I (meso-diaminopimelate containing)|g__Acidaminococcus.s__Acidaminococcus_sp_CAG_542

…

PEPTIDOGLYCANSYN-PWY: peptidoglycan biosynthesis I unclassified

I am not sure why using --last-metadata flag made an error.

I will try to figure out what the problem is with comparing a file of hmp_pathabund.pcl from humann tutorial.

Thanks,
Chang

4

You only need to invoke --last-metadata if you have metadata rows at the top of your table that you want to ignore. Is that the case here?