Read length

[sankadinesh]

Is kraken suitable for taxonomic analysis of long reads (nanopore sequences) ?

[davidvilanova]

I have tested kraken2 with nanopore reads but due to high error rate kraken2 will not identify correctly the organisms.

[jenniferlu717]

It is more appropriate to change the kmer length in the database build for nanopore reads. With Kraken1, we used a kmer length of 28 I believe. I have yet to determine the best kmer length for Kraken 2 but this is something in the works.

[sylvain-ri]

Any update of the optimum kmer length for long reads such as nanopore ?

[zmunro]

@jenniferlu717 Any update on a good kmer length to use when building the kraken2 database for classifying nanopore reads?

[angelovangel]

Hi @zmunro, Alex Dilthey has some data about this in his metamaps paper. Check it out here

[ms-gx]

It is more appropriate to change the kmer length in the database build for nanopore reads. With Kraken1, we used a kmer length of 28 I believe. I have yet to determine the best kmer length for Kraken 2 but this is something in the works.

What would you suggest for minimizer length and minimizer spaces in addition to kmer size? An educated guess is also welcome :)

[ms-gx]

Hi @zmunro, Alex Dilthey has some data about this in his metamaps paper. Check it out here

Where do we find suggestions for parameters?