This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Reads that used for re-constructing MAGs can not be classified by Kraken2

I have 2 paired shot gun sequencing data sets of human gut microbiome using Hiseq. I re-constructed MAGs using metaSpades, metabat2, checkM. For the 2 data sets, 55-65% paired reads concordantly aligned 1 time to the contigs with 75-88% overall alignment rate using bowtie2. Everything looks fairly Ok to me so far.

Then, I built a custom database using Kraken2 to classify the reads that were used for MAG re-constructing. <1% reads were classified by Kraken2. When I used a group of published genomes of human gut microbiome, ~50% reads could be classified.

My question: Can I expect that most reads are classified by Kraken2? Since MAGs are built with the same reads, I would guess most of them can be successfully classified. Thank you. Appreciate your ideas.

metagenomics mag kraken classification
• 404 views • link
updated by Mensur Dlakic Mensur Dlakic • written by Yifeng • Yifeng

Hello Yifeng!

We believe that this post does not fit the main topic of this site.

For this reason we have closed your question. This allows us to keep the site focused on the topics that the community can help with.

If you disagree please tell us why in a reply below, we'll be happy to talk about it.

Cheers!

• 0 views • link
written by Yifeng • Yifeng

Thank you for letting me know. I am sorry that I didn't see this message before I submit to close this post.

• 0 views • link
written by Yifeng • Yifeng

I figured it out that the fasta sequences have not be recognized by Kraken2. After reformatting the fasta files of my MAGs, ~88% reads aligned to them.

The fasta file format I used before:

>kraken:taxid|xxxxx sequence_description

Correct fasta file format for Kraken2 custom database:

>kraken:taxid|xxxxx|sequence_description

• 0 views • link
written by Yifeng • Yifeng

1 answer

I am curious: why do you need to classify the reads that are already assembled into MAGs, when you can classify the MAGs? MAGs are much longer and therefore contain stronger and more reliable signals.

To classify MAGs: https://github.com/Ecogenomics/GTDBTk

I would understand if you wanted to classify unmapped reads. While this is not the answer to your question, you may want to try sketching, which has been introduced on Biostars in this thread.

• 0 views • link
written by Mensur Dlakic Mensur Dlakic

Thank you for your reply. I have done GTDB-tk. I was simply curious, so I built a custom Kraken database using MAGs and contigs and then ran the scripts. I have used other bbtools but not bbsketch. I appreciate the post you provided and will read it soon.

• 0 views • link
written by Yifeng • Yifeng

Log in to answer this question.

More posts like this