omicverse.metabol.asca_variance_bar

omicverse.metabol.asca_variance_bar#

omicverse.metabol.asca_variance_bar(asca_result, *, ax=None, figsize=(5.5, 3.0))[source]#

Horizontal bars of per-effect variance-explained fractions.

Visualises the partition of total sum-of-squares from omicverse.metabol.asca() across the named effects (factors and their interactions) plus the residual. The grey “residual” bar at the bottom shows what the model didn’t explain — a useful sanity check: if residual variance dominates, the planned factors don’t capture the dominant signal and the model needs revision.

Parameters:

  • asca_result (ASCAResult) – Output of asca(). Reads .effects[name].variance_explained for each named effect plus .residual_ss / .total_ss.

  • ax (Optional[Axes] (default: None)) – Standard matplotlib hooks.

  • figsize (tuple[float, float] (default: (5.5, 3.0))) – Standard matplotlib hooks.

Returns:

  • (fig, ax) tuple. Bar labels show variance percentage with one

  • decimal place.

Contents