feature request: send description from database headers to output table

[schultzm]

Hi @tseemann ,

Feature request: add a new column to the abricate output that contains 'DESCRIPTION' information. For example, the ncbi database has

>ncbi~~~mcr-1~~~A7J11_03461 phosphoethanolamine--lipid A transferase MCR-1
>ncbi~~~mcr-1.2~~~A7J11_03754 phosphoethanolamine--lipid A transferase MCR-1.2
>ncbi~~~mcr-1.3~~~A7J11_04944 phosphoethanolamine--lipid A transferase MCR-1.3

in the headers. I would like to be able to extract from the abricate stdout that mcr-1.3 is a phosphoethanolamine--lipid A transferase

Thanks

[tseemann]

Ok, there is a now a PRODUCT column.
Only some DBs and some genes have a product though.

[schultzm]

lightning-fast update. thanks!

[nalbright]

Where in the fasta header does the product information need to be. I have tried running a database with headers having a space and description (as per instructions for custom DB), but it does not display "product" in the output .csv. Is there a flag that I am missing? or do I have that information in the wrong spot in the fasta header?
Thanks!
Nicolette

[tseemann]

@nalbright can you paste the first line of your database sequences file?
and also what does abricate --version say?

[nalbright]

I am running abricate through biocontainer --version: abricate 0.5-dev

Header:

S_epidermidis_ISOLATEDB~~~NP_763556.1~~~OINGLKLB_00795 chromosome replication initiator DnaA; 1056816

Thanks!

[tseemann]

@nalbright your version of Abricate is too old. Support for PRODUCT came after 0.5.