Output with absolute counts instead of relative abundances

[davidecrs]

Hi,

is it possible to have absolute counts in the output file of the tool (instead of relative abundances)?

Thanks!

Best Regards
Davide

[dkoslicki]

Sorry for missing this @davidecrs . The way things are set up currently, relative abundances are all we can output due to using number of k-mers vs. total number of k-mers in the file. @mahmudhera , correct me if I am wrong, but sourmash prefetch at most gives an estimate of the counts, but this is still relative to what is essentially the number of reads/k-mers in the input data. I.e. absolute counts (in terms of reads or k-mers) are still relative to your sequencing depth

[mahmudhera]

@dkoslicki you are right. We can only estimate the content (in terms of distinct kmers) within a KO present in a sample. If a larger fraction of a KO's content is present, we estimate the KO to be "more" present than others. We do not directly have access to the number of reads mapped to a KO, and so I don't see a straightforward way to get "absolute counts". Does that make sense?