search

dib-lab / 2020-paper-sourmash-gather

Here we describe an extension of MinHash that permits accurate compositional analysis of metagenomes with low memory and disk requirements.
https://dib-lab.github.io/2020-paper-sourmash-gather
Other
8 stars 1 forks source link

do we want to discuss gather & abundance? #4

Closed ctb closed 3 years ago

ctb commented 3 years ago

related to #3, we have an opportunity to benchmark abundance calculations vs read mappings. should we? or does it confuse the message?

this could part of a different paper, too - perhaps something to do with strain-level detection and quantification. see below excerpt from e-mail to david koslicki -

I’m pretty confident that this approach can be used to do accurate strain-level resolution, based on the combinatorial analysis of the content of each strain’s accessory elements. That was a mouthful, but, basically, as long as each strain’s genome has a unique combination of k-mers (which each must, because otherwise it would be the same as another genome…) our approach can match metagenomes to it based on that unique combination.

ctb commented 3 years ago

preliminary discussion with @luizirber => abundance just makes things more confusing, so we plan to omit it from this paper.