Bio Applied Microbial Diversity

ASVs are now recommended (Callahan et al., 2017). DADA2 learns error rates per nucleotide transition, tests whether each unique sequence could be explained as an error from a more abundant sequence.

Taxonomy Assignment

Reference databases: SILVA (~2M seqs, most comprehensive), Greengenes2 (genome phylogeny), RDP, GTDB (genome-based gold standard).

Alpha Diversity

Measures diversity within a single sample:

• Richness: observed species count
• Evenness: how evenly distributed (Shannon, Simpson)
• Phylogenetic: Faith's PD incorporates evolutionary tree

Qualitative (presence/absence) vs quantitative (abundance-weighted).

Feature Table

Central data structure: rows = OTUs/ASVs, columns = samples, values = read counts. Simulate with Dirichlet-multinomial model per body site.

Pitfalls

• Rarefaction depth: Unequal library sizes bias richness; rarefy or use normalized metrics
• Batch effects: Always check for batch confounding before interpreting biological signal
• Multiple testing: Apply FDR correction (Benjamini-Hochberg) when testing thousands of features simultaneously