gravatar for Mensur Dlakic

1 hour ago by

USA

It would make no sense to do that because you are talking about two different datasets. The whole point of binning using coverage is to add yet another data column to the clustering procedure. That piece of information would be useless here because there is no way of knowing whether the two datasets are related in terms of coverage.

There should be no problem in binning without coverage, as tetranucleotide frequencies contain plenty of information. If you use software that absolutely requires coverage, you can make it up by giving all contigs an equal coverage value (1x, 100x, the exact number doesn't really matter). When all coverages are identical (which is actually true in the absence of that information), that data column will not contribute to the final binning result.



Source link