grassp.tl.resolve_diffusion

resolve_diffusion(data, gene_sets, key_added='ann_diffusion', mode='likelihood', min_term_size=0, min_probability=0.5, eta=1.0, tau=0.4, maxk=3, cap=12, gene_key='gene_symbol', species='hsap', out_key=None)

(Re)resolve stored diffusion probabilities into a per-protein label, in place.

Reads obsm[{key_added}_probabilities] / uns[{key_added}_categories] (written by independent_diffusion()) and writes obs[out_key] (default f"{key_added}_resolved" for likelihood/argmax, f"{key_added}_resolved_specific" for specific). mode and min_term_size are as in independent_diffusion(). Lets you obtain several resolutions (e.g. likelihood and specific) or sweep min_term_size without re-diffusing.

Return type:

None

Parameters:

data AnnData

key_added str

mode Literal['likelihood', 'specific', 'argmax']

min_term_size int

min_probability float

eta float

tau float

maxk int

cap int

gene_key str

species str

out_key str | None