doubt trans_beta

[jamorillo]

Hello, I apologize for this basic question regarding the calculation of beta diversity in microeco. In the following operation, on which otu table is the Bray-Curtis distance calculated? Is it a matrix already normalized by total sum scaling, or directly from the OTU matrix without any other transformations?

• create an trans_beta object
• measure parameter must be one of names(dataset$beta_diversity) t1 <- trans_beta$new(dataset = dataset, group = "Group", measure = "bray")

Given my general background, I understand that the steps should be: 1) rarefaction (optional...), 2) Total Sum Scaling (TSS) normalization, 3) optional transformation (sqrt, Hellinger, etc.,), 4) dissimilarity calculation, 5) finally,subsequent analyses (PERMANOVAs, NMDS, etc.). However, after reviewing the analyses with microeco, I've realized that I dont know how trans_beta works internally, and I would like to confirm this. Thank you!

[ChiLiubio]

Hi, Bray-Curtis distance calculation is performed in microtable class, not in trans_beta class. So the correct format should be dataset$cal_betadiv(). For the rarefaction, please see the microtable part of tutorial (https://chiliubio.github.io/microeco_tutorial/basic-class.html#microtable-class). The operation dataset$rarefy_samples() may be what you need and is very easy to use. I donnot suggest using TSS for bray distance. For the normalization methods, please also see the tutorial (https://chiliubio.github.io/microeco_tutorial/mecodev-package.html#trans_norm). The things that trans_beta does is the subsequent analysis.

[jamorillo]

Hello, I got it. I also did the most of the analysis directly in vegan, and my results are consistent with what I got from microeco. Thanks for your kind reply, and for developing this project. Best reagrds!