SRS alternative tool to rarefaction

[marwa38]

Hi team

Would you agree to to add SRS as an alternative to rarefaction in ampliseq? It have lately recommended and prefered when compared to rarefaction.

I have Rcodes that I used for this here ChiLiubio/microeco#443

Also could try to implement SRS plugin from qiime2 if you prefer? as in the discussion here https://forum.qiime2.org/t/srs-as-a-normalisation-method/30531/4

But I will be a new contributer to nfcore pipelines so would need to figure it out.

Thanks for letting me know
Marwa

[erikrikarddaniel]

I've seen your discussion with @d4straub and agree: Why not if you have the time to implement it. Doing it via QIIME2 seems simplest as we already have a container that's used in several modules.

For your actual work, note that there are methods that circumvent subsampling altogether and uses the whole data set to calculate alpha diversity. I've been using the R package breakaway: https://adw96.github.io/breakaway/ (make sure you use the development version and not the CRAN version; for this reason I don't think we can find a container and it would hence be difficult to include in Ampliseq). DivNet (https://github.com/adw96/DivNet) from the same group provides more diversity measurements, but I have no experience with that package and I think it's rather heavy computationally.

[d4straub]

I tried DivNet some time ago and couldnt get it to run, see nf-core slack. Maybe it was updated though.