HUMAnN: Unable to convert gene table value to float

Microbial community profiling HUMAnN
1

Using humann v3.9 I try to run it on a dummy gene table which looks as follows:

# Gene Family   E100002696_L01_1_kneaddata_Abundance-RPKs
UNMAPPED    4887640.0000000000
UniRef50_Q9RCA1 2135.6903965600
UniRef50_D7UV20 2135.6122163071
UniRef50_A0A7X0X377 2135.6122163071

I verified that the file is tab-separated and yet I am getting a following error in the gene_table_humann_temp/gene_table.log :

12/11/2024 12:56:48 PM - humann.humann - INFO: Process the gene table ...
12/11/2024 12:56:48 PM - humann.store - DEBUG: Unable to convert gene table value to float: UNMAPPED    4887640.0000000000
12/11/2024 12:56:48 PM - humann.store - DEBUG: Unable to convert gene table value to float: UniRef50_Q9RCA1 2135.6903965600
12/11/2024 12:56:48 PM - humann.store - DEBUG: Unable to convert gene table value to float: UniRef50_D7UV20 2135.6122163071
12/11/2024 12:56:48 PM - humann.store - DEBUG: Unable to convert gene table value to float: UniRef50_A0A7X0X377 2135.6122163071
12/11/2024 12:56:48 PM - humann.humann - INFO: TIMESTAMP: Completed 	processing gene table 	:	 0	 seconds

The resulting gene_table_pathabundance.tsv file looks as follows:

# Pathway	gene_table_Abundance
UNMAPPED	0.0000000000
UNINTEGRATED	0.0000000000

I also tried to use the example data posted in this issue: Example for `File Type 4 (gene table) tsv` as input to HUMAnN? - #2 by w_ceasea with the same result.

What is the correct format of gene table if that’s what I want to use as an input for HUMAnN?

2

It looks like something has gone wrong with the tab (\t) characters that separate the gene ID column (1) from the abundance column (2), such that it is seeing both as a single column. Did you open and/or edit the files in any way? It’s possible that a text editor replaced the \ts with spaces (for example) which could then no longer be correctly parsed.