search

microbiome / OMA

Orchestrating Microbiome Analysis
https://microbiome.github.io/OMA
75 stars 42 forks source link

Rarefaction as optional strategy for beta diversity #332

Open antagomir opened 11 months ago

antagomir commented 11 months ago

Ref: Pat Schloss' recent analyses on rarefaction in alpha & beta diversity. This addressess well some of the long debates and controversies surrounding rarefaction in microbial ecology.

Based on this I tend to think that rarefaction should be provided as an option for beta diversity calculations.

Here rarefaction refers to averaging alpha diversity estimates across many random rarifications (100x and 1000x are used in the preprint). This makes most difference in less diverse and low abundance environments but may influence also more rich ecosystems (see the preprint).

In that prepint: ""Beta-diversity values generated with rarefaction were calculated using the avgdist function in vegan. Briefly, vegan’s avgdist function calculates each pairwise dissimilarity index after obtaining a random draw of the same number of sequences from each sample. After obtaining 100 random draws it returns the mean value.""

We might be able to use this directly with scater::runMDS via the FUN argument from the latest PRs contributed by @BananaCancer - worth checking.

Then this could be provided as a recommended option in OMA.

Not sure if this can be addressed in reasonable new ways in mia, given that the beta diversity functionality largely comes from vegan and scater packages.

antagomir commented 8 months ago

See also https://github.com/microbiome/mia/issues/463

antagomir commented 2 months ago

@thpralas would you be interested looking into this once the situation relaxes? (assuming that @RiboRings is not already working on this - I guess he stays busy with other related topics). This is a good, more extensive task to look at and discuss, then providing a suitable solution in mia & OMA.

thpralas commented 2 months ago

@thpralas would you be interested looking into this once the situation relaxes? (assuming that @RiboRings is not already working on this - I guess he stays busy with other related topics). This is a good, more extensive task to look at and discuss, then providing a suitable solution in mia & OMA.

Thanks! I will look into this soon.