dotools_py.bm.pcr_comparison

dotools_py.bm.pcr_comparison(adata_pre, adata_post, covariate, use_rep, n_comps=50, recompute_pca=False, linreg_method='numpy', scale=True)

Principal component regression score.

Compare the explained variance before and after integration. Return either the difference of variance contribution before and after integration or a score between 0 and 1 (if scaled is set to True) with 0 if the variance contribution hasn’t changed. The larger the score, the more different the variance contributions are before and after integration.

Parameters:

adata_pre AnnData

Annotated data matrix before the integration

adata_post AnnData

Annotated data matrix after the integration

covariate str

Column in adata.obs to regress against

use_rep str

Embedding to use for principal component analysis. If None, use the full expression matrix (adata.X), otherwise use the embedding provided in adata_post.obsm[use_rep].

n_comps int (default: 50)

Number of principal components to compute

recompute_pca bool (default: False)

Whether to recompute PCA with default settings

linreg_method Literal['numpy', 'sklearn'] (default: 'numpy')

Method for computing the linear regression

scale bool (default: True)

If set to True scale score between 0 and 1.

Return type:

float

Returns:

Returns the difference of variance contribution of PCR.

Examples

>>> import dotools_py as do
>>> adata = do.dt.example_10x_processed()
>>> adata_unintegrated = adata.copy()
>>> del adata_unintegrated.obsm
>>> do.bm.pcr_comparison(adata_pre=adata_unintegrated, adata_post=adata, covariate="batch", use_rep="X_CCA")
Out[47]: np.float64(0.832)