Wonder if anyone can recommend in particular any o...
# taxprofiler
j
Wonder if anyone can recommend in particular any of the tools in Taxprofiler/specific databases for very broad-strokes classification (bacteria/eukarya - to phylum level maybe)? Want to assess how much contamination we have in some Illumina short read sequencing data
e
I'd try kraken2's standard 8G database first. If you have a lot of unclassified reads then I would recommend moving onto some more specific.
j
For broad strokes I would agree kraken for rough and ready
But if you want eukarya maybe one of the larger ones? https://benlangmead.github.io/aws-indexes/k2
j
I do have the core_nt one downloaded, but it really seems overkill. But then again, the standard databases don’t include much by way of eukaryotes 😢
j
Yeah that's why I suggested the larger ones
Oh no wait PFP is basically everything BUT eukarya...
The lack of standard databases are still really bothers me
j
Maybe the thing to do is actually just to quantify the samples with Sourmash/Sylph or similar + GTDB - really I just need to see if the proportion changes between two treatments…
j
But gtdb won't hold your eukaryotes right?
j
No, but if you see the proportion of unclassified change, that might do in a pinch
šŸ‘ 1