skbio.stats.composition.ANCOMBCResult.dunnett_test#

ANCOMBCResult.dunnett_test(alpha='inherit', p_adjust='inherit', bootstraps=100, seed=None)[source]#

Perform Dunnett’s test (each group vs. reference) with mdFDR.

Parameters:
alphafloat or “inherit”, optional

Significance level. Default is “inherit”, which will use the value supplied upstream.

p_adjuststr, optional

Family wise error (FWER) controlling method. Default is “inherit”, which will use the p-value correction method supplied upstream.

bootstrapsint, optional

Number of bootstrap iterations. Default is 100.

seedint, Generator, or RandomState, optional

A user-provided random seed or generator for bootstrap samples. See details.

Returns:
pd.DataFrame

Dunnett test result with a (FeatureID, Comparison) multi-index. The index levels and columns are:

  • FeatureID: Feature identifier, i.e., dependent variable.

  • Comparison: Group-versus-reference contrast being tested.

  • Log(FC): Estimated group effect relative to the reference group on the natural-log abundance scale.

  • SE: Standard error of the estimated contrast.

  • W: W-statistic, calculated as the estimated contrast divided by its standard error.

  • pvalue: Uncorrected p-value of the W-statistic.

  • qvalue: p-value after mixed directional false discovery rate correction.

  • Signif: Whether the group differs significantly from the reference.