omicverse.single.pyMOFA

class omicverse.single.pyMOFA(omics, omics_name)

Train MOFA models for latent factor discovery across multiple omics layers.

Parameters:

• omics (list) – List of omics data matrices (for example RNA/ATAC/proteomics).

• omics_name (list) – Names corresponding to each entry in omics.

Returns:

Initializes MOFA training inputs and metadata.

Return type:

None

Examples

>>> test_mofa = ov.single.pyMOFA(omics=[rna, atac], omics_name=["rna","atac"])

__init__(omics, omics_name)

Initialize a multi-view MOFA training container.

Parameters:

• omics (list) – List of input omics objects, typically AnnData matrices for each modality (for example RNA, ATAC, protein).

• omics_name (list) – View names aligned with omics order. These names are used in the exported MOFA model and downstream interpretation.

Methods

__init__(omics, omics_name)	Initialize a multi-view MOFA training container.
mofa_preprocess()	Convert input omics objects into MOFA-ready dense matrices.
mofa_run([outfile, factors, iter, ...])	Train and save a MOFA model using current multi-omics inputs.