"FuncDiv" for contirbutional/functional diversity (alpha and beta diversities)?

[marwa38]

Hi team This is a powerful package and your up-to-date tutorial is very helpful. I was wondering if you could help with the downstream analysis for functional diversity? https://github.com/gavinmdouglas/FuncDiv FuncDiv is a new package developed by the picrsut2 developers for different workflows and not restricted to microbiota. I was wondering if you could help by giving just an example tutorial for it. Running the codes for functional alpha and beta diversity is straightforward but the downstream analysis needs some inputs from the microbiome world. In the meanwhile if you please could suggest some ideas for downstream analysis? that would be great. Cheers Marwa

[ChiLiubio]

Thanks. I will try to figure it out. Do you mean using the microeco package to do "downstream analysis" of the output of FuncDiv package?

Chi

[marwa38]

"downstream analysis" of the output of FuncDiv package?

yes, the package paper is straightforward and short and intentionally left without options for downstream analysis regarding the microbiome world so that everyone use it according to what they want to analyze.

Here is more info: https://twitter.com/gavin_m_douglas/status/1622951069293903874

"Essentially it is just like standard alpha diversity, but there are just many more dimensions (i.e., one for each function). So one could use ordination-based approaches (or other appropriate approaches for high-dimensional data) to compare the contributional alpha diversity across samples. You could also test whether there are specific functions that exhibit significant different contributional alpha diversity between sample groups of interest." Gavin

[marwa38]

For the functional analysis, it would great to help with the downstream analysis of POMSthis was developed by one of the main contributors to picrust2 but again needs help with downstream analysis. https://twitter.com/gavin_m_douglas/status/1595072133071347712 https://academic.oup.com/bioinformatics/article/38/22/5055/6731923?login=true https://github.com/gavinmdouglas/POMS

[ChiLiubio]

Thanks very much. I will learn more on it.

[marwa38]

ggpicrust2 a new package for visualisation of picrust2 especially that STAMP haven't been updated since 2015 that might be good

[ChiLiubio]

Yes. I also find it. Thanks.

[marwa38]

Hope you are doing well. @ChiLiubio Any updates on how to visualise the functional diversity? Writing my last manuscript/chapter and I was thinking adding this would be great. Thanks in advance

[ChiLiubio]

Hi. Sorry on that It was put on hold. Maybe the primary reason is I have no idea about how to extend functional diversity analysis based on the current framework. Please feel free to tell me if you have more detailed need on the analysis.

[marwa38]

Some suggestions are below that would be nice to consider: I think also what are the specific functions that need to have RDA? I think you are already have those functions active in microeco but will divert to the functional/contributional diversity: How you find that?

[marwa38]

@ChiLiubio

Just checked this paper referring to functional diversity with a simple figure https://www.nature.com/articles/s41398-023-02325-5/figures/3 Another manuscript: https://bmcoralhealth.biomedcentral.com/articles/10.1186/s12903-023-02911-5/figures/4

[ChiLiubio]

Hi These analysis methods can be easily achieved by the microeco package. The primary thing is to convert functional data to microtable object like what we show in the functional prediction part (https://chiliubio.github.io/microeco_tutorial/explainable-class.html#trans_func-class). Then all the operations are very similar with the steps in the tutorial. One notable thing is when you use cal_alphadiv function, please add parameter measures = "Shannon".