A TXT-type anviāo artifact. This artifact is typically provided by the user for anviāo to import into its databases, process, and/or use.
š To the main page of anviāo programs and artifacts.
There are no anviāo tools that generate this artifact, which means it is most likely provided to the anviāo ecosystem by the user.
anvi-compute-functional-enrichment-across-genomes anvi-compute-metabolic-enrichment anvi-display-functions
This is a 2-column TAB-delimited text file to associate a given set of items with a set of groups. Depending on the context, items here may be individual samples or genomes. The first column can have a header name item
, sample
, genome
or anything else that is appropriate, and list the items that are relevant to your input data. The second column should have the header group
, and associate each item in your data with a group.
Each item should be associated with a single group, and it is always a good idea to define groups using single words without any fancy characters. For instance, HIGH_TEMPERATURE
or LOW_FITNESS
are good group names. In contrast, my group #1
or IS-THIS-OK?
, are not quite good names for groups and may cause issues downstream depending on who uses this file.
Here is an example groups-txt file:
item | group |
---|---|
item_01 | GROUP_A |
item_02 | GROUP_B |
item_03 | GROUP_A |
(ā¦) | (ā¦) |
If you are passing this file to the program anvi-compute-metabolic-enrichment, the names in the sample
column must match those in the āmodulesā mode output file that you provide to the program via the --modules-txt
parameter. If you know that the sample names match but you are still getting errors, you might need to specify which column in the āmodulesā mode output contains those sample names using the --sample-header
parameter.
Edit this file to update this information.