omicverse.metabol.read_lcms

Contents

omicverse.metabol.read_lcms#

omicverse.metabol.read_lcms(path, *, feature_id_sep='/', sample_col=None, group_col=None, label_row=None, transpose=True)[source]#

Load an LC-MS peak table with m/z/RT feature IDs into AnnData.

Handles MetaboAnalyst’s LC-MS demo layout (peaks-in-rows with a dedicated “Label” row above the data), as well as plain wide tables. Feature IDs of the form <m/z><sep><RT> (e.g. "200.1/2926" or "85.065__24.64") are parsed into numeric var['m_z'] / var['rt'] columns so mummichog_basic() can consume the AnnData directly.

Parameters:

  • feature_id_sep (str (default: '/')) – Separator between m/z and RT inside each feature ID. MetaboAnalyst uses "/" for the small demo and "__" for the malaria table; XCMS exports sometimes use "_".

  • label_row (Optional[str] (default: None)) – When the raw file has features in rows, the first “feature” is often a dedicated group-label row (e.g. MetaboAnalyst’s "Label" row with values like "Naive" / "Semi_immue"). Pass the label of that row here and we’ll lift it into adata.obs['group'] before treating the rest as features.

  • transpose (bool (default: True)) – Default True (peaks-in-rows → samples-in-rows, AnnData orientation).

  • sample_col (Optional[str] (default: None)) – Only used when the table is already in samples-in-rows layout; both default to the first column / the first non-sample column.

  • group_col (Optional[str] (default: None)) – Only used when the table is already in samples-in-rows layout; both default to the first column / the first non-sample column.

Return type:

AnnData

Contents