Extracting AA sequences from each sample

[andressamv]

Hi, I have a question about the Kaiju output. I want to extract the AA sequences derived from fungi from each sample. My goal is to use the sequences as input in tools such as GhostKoala. I understand that using -v I will get those sequences, but how can I extract only the fungal ones? More importantly, how can I know the sample of origin of each read? Thank you

[pmenzel]

Hi,

• you could use kaiju-addTaxonNames -p to add the full taxonomic information to each line in the kaiju output file (see here) and then just keep files containing Fungi
• I don't get the second question

[pmenzel]

btw, please note that the sequences shown in the output file are the matched database sequences and not the translated amino acid sequences from the reads

[andressamv]

Hi @pmenzel. Thank you for this info! Can you explain what this means for mismatches? If I understood correctly, if I have mismatches, I will not be able to see them in the output, only the reference sequence. Is this correct?

[pmenzel]

yes, the output shows the matched database sequence.