Smaller datasets after applying trans_alpha or trans_beta

[Dolores22]

Hi, I am new to microeco.
I have been using it first to perform alpha and beta diversity analyses, but after applying trans_alpha or trans_beta my dataset is reduced both in number of ASVs and samples.
Is there a filter applied that I´m not seeing?
Thank you so much !

[ChiLiubio]

Hi. Could you please provide an example that I can totally reproduce your issue?

[Dolores22]

This is what I did. When I looked at the saved file with bray distances all the samples are there, but in the generated PcoA only a subset of samples is displayed and the microtable is also reduced after applying trans_beta. Otu_table are ASVs counts.
So sorry if this is something basic, microeco is new to me. Thanks!

tax_ABCD.2 %<>% tidy_taxonomy
mt_ABCD.2 <- microtable$new(sample_table = ABCD.2_metadata, otu_table = counts_ABCD.2, tax_table = tax_ABCD.2)
mt_ABCD.2
microtable-class object:
sample_table have 132 rows and 63 columns
otu_table have 5760 rows and 132 columns
tax_table have 5760 rows and 6 columns
mt_ABCD.2$cal_betadiv(unifrac = FALSE)
The result is stored in object$beta_diversity ...
mt_ABCD.2$save_betadiv(dirpath = "beta_diversity")
mt_ABCD.2
microtable-class object:
sample_table have 132 rows and 63 columns
otu_table have 5760 rows and 132 columns
tax_table have 5760 rows and 6 columns
Beta diversity: calculated for bray,jaccard
t1_beta <- trans_beta$new(dataset = mt_ABCD.2, group = "sampleType", measure = "bray")
t1_beta$cal_ordination(method = "PCoA")
The result is stored in object$res_ordination ...

t1$res_ordination is the ordination result list

class(t1$res_ordination)
[1] "NULL"

plot the PCoA result with confidence ellipse

t1_beta$plot_ordination(plot_color = "Site", plot_shape = "sampleType")+

• scale_color_manual(values = color_site )+
• theme_gray()
  Scale for colour is already present.
  Adding another scale for colour, which will replace the existing scale.

mt_ABCD.2
microtable-class object:
sample_table have 47 rows and 63 columns
otu_table have 1942 rows and 47 columns
tax_table have 1942 rows and 6 columns
Beta diversity: calculated for bray,jaccard

[ChiLiubio]

Hi. Maybe there is some ASVs and samples that have 0 abundance. When you use trans_beta class, the useless data is automatically filtered.
To solve this, please remember to add tidy_dataset function in your pipeline to tidy all the data in the microtable object.

mt_ABCD.2 <- microtable$new(sample_table = ABCD.2_metadata, otu_table = counts_ABCD.2, tax_table = tax_ABCD.2)
mt_ABCD.2$tidy_dataset()
mt_ABCD.2

Another way is to set auto_tidy = TRUE

mt_ABCD.2 <- microtable$new(sample_table = ABCD.2_metadata, otu_table = counts_ABCD.2, tax_table = tax_ABCD.2, auto_tidy = TRUE)
mt_ABCD.2

Best,
Chi

[Dolores22]

All the samples and ASVs have abundances > 0, but appliying tidy_dataset solve the issue, thanks