behaviour of extract_kraken_reads

[jorondo1]

Hi,

If I were to provide a genus ID, would using --exclude --include-children --include-parent manage to remove:

• the read from the phyla
• all its children,
• and (where I'm less sure) : all the reads from the parent taxid

We need the counts from, for example, the parent Kingdom to be readjusted by the same number of reads being removed (but not for the entire kingdom to be removed, obviously).

Also, does the script adjust every kraken output file, or does it only adjust the classified reads ? In other words, do I need to regenerate a kraken report afterwards using make_kreport.py?

By the way, thanks a lot for this tool, it's definitely an indispensable complement to Kraken, it has solved so many roadblocks we faced!