long reads

[davidvilanova]

Hi,
Working with nanopore reads i´m using mash screen to identify the genomes from an input reads file as follow:

mash screen -w -p 3 refseq.genomes.k21s1000.msh reads_nanopore.fq.gz
So far it is doing pretty well as i recover a specific genome (propionibacteria) i was expecting in the metagenome (just using a small subset of 4000 nanopore reads).

I was not expecting the plant genomes in there so i was wondering if there is some finetuning to do ?
For instance filtering on the number of shared-hashes ??

Is this the proper usage to get the mapped genomes of my nanopore reads ?

David

[tseemann]

@davidvilanova i don't think that is a plant genome, it looks like a plant pathogen; a microbe of some sort. it might have phage content similar to your bacterium

[davidvilanova]

Was not expecting this in a skin sample, but you´re right it´s probably close to the bacterium.