daa-guide

To choose a differential abundance analysis method for sparse microbiome data, apply evidence-based decision rules that map dataset characteristics like distribution and longitudinal design to LinDA, ZINB, Hurdle, NB, or LMM contexts.

Zero significant q-values in differential abundance analysis often indicate threshold mismatches or insufficient power. Troubleshoot by verifying method-specific thresholds and reviewing study design parameters against benchmarking references.

Interpret q-values in differential abundance analysis by applying method-specific thresholds and contextualizing results within your study design, as interpretation rules vary across LinDA, ZINB, Hurdle, NB, and LMM pipelines.

Select LinDA, ZINB, or Hurdle models for differential abundance analysis by mapping your dataset's sparsity and distribution patterns to benchmarking evidence, using concrete decision rules for unknown distributions.

Differential abundance analysis in longitudinal microbiome studies requires method-specific thresholds rather than uniform standards. Apply thresholds tailored to LMM or other chosen methods and interpret q-values within the study design context.

Unexpected differential abundance analysis results with zero-inflated data usually stem from method mismatch. Select ZINB or Hurdle models based on benchmarking decision rules and verify method-specific thresholds are correctly applied.