The two scripts largely use the same MicrobiotaProcess functions; the main difference is *which `ps_` object you feed in for each task** (alpha, beta, composition plots).

What the “small” script currently does

• Creates one MPSE (often called mpse_abund) from a taxa-filtered plotting object (e.g. ps.ng.tax_abund / your ps_abund) and then runs alpha diversity + beta diversity + composition plotting all on that same MPSE.
• It also uses rarefied abundance (RareAbundance) as the input for some taxonomy abundance plots/heatmaps.

Implication: alpha/beta diversity are being computed on a taxa-filtered dataset, which can:

• artificially reduce richness (Observed/Chao1),
• shift Shannon/Simpson,
• and change distance structure (especially for presence/absence metrics, but also sometimes Bray).

What the updated MicrobiotaProcess workflow (recommended) does

It splits the workflow into two MPSE objects, each built from the correct upstream ps_*:

1) Diversity MPSE (mpse_div)

Input: ps_filt (QC-filtered samples, full taxa set; not filtered “for plotting”)

• Alpha diversity: do rarefaction inside MPSE (mp_rrarefy()) and compute alpha on RareAbundance.
• Beta diversity (your current Bray+Hellinger): compute Hellinger from non-rarefied Abundance and then Bray/PCoA/PERMANOVA.

2) Plotting MPSE (mpse_plot)

Input: ps_abund_rel (taxa filtered for readability; relative abundance)

• Use this MPSE only for composition plots (stacked bars, heatmaps).

Implication: you keep diversity analyses biologically faithful (no “plotting filter”) while still producing clean, readable composition plots.

Other (non-critical) differences

• The updated script uses prune_samples() + prune_taxa() instead of overwriting otu_table() manually (safer / less error-prone).
• Outputs are consistently written into a figures/ directory.

Bottom line

Same MicrobiotaProcess functions; the updated script mainly fixes the *recommended `ps_` input choice** for each analysis type (diversity vs. plotting).