How did you make the custom db and what is there in the console? Were there any error messages?

Hi Abhijeet,

No error messages. The db creation seemed to work and I have also try to run it with one of my samples and I got the kraken report (although I only got hits for only one taxa).

I downloaded the fasta files of these taxa from the NCBI website (for example: alnus_glutinosa_GCA_003254965.1.fna).

To build the db I followed what is written in the kraken2 manual:

kraken2-build --add-to-library $file --db PATH/kraken/plant

kraken2-build --build --db ~/kraken/plant

Valentina Vanghi

Does the fasta header in your sequence file has taxid as per the requirements of kraken2?

Try this

https://github.com/hcdenbakker/GTDB_Kraken

There you can find a huge database or instructions/script if you want to do manual db.

Hi,

I have run the fix_unmapped.py from KrakenTools and it produced the "still.unmapped.txt" file which is exactly the same as unmapped.txt (same accession number list).

Now the question is:

I have downloaded the genomes to build this custom db from the NCBI, why the software cannot find the taxa IDs if it does also use the NCBI genomes?

Thanks

Hi Nikolaos Psonis the file I have is pretty recent...I give up! But I have found my way using another program: Kaiju :) which seems to work.

Thanks for the help everyone

Hi thanks, do you know if Centrifuge is better than Kaiju?