scanpy.pp.pca#
- scanpy.pp.pca(data, n_comps=None, *, layer=None, zero_center=True, svd_solver=None, random_state=0, return_info=False, mask=Empty.token, use_highly_variable=None, dtype='float32', copy=False, chunked=False, chunk_size=None)[source]#
Principal component analysis [Pedregosa11].
Computes PCA coordinates, loadings and variance decomposition. Uses the implementation of scikit-learn [Pedregosa11].
Changed in version 1.5.0: In previous versions, computing a PCA on a sparse matrix would make a dense copy of the array for mean centering. As of scanpy 1.5.0, mean centering is implicit. While results are extremely similar, they are not exactly the same. If you would like to reproduce the old results, pass a dense array.
- Parameters:
- data
AnnData
|ndarray
|spmatrix
The (annotated) data matrix of shape n_obs × n_vars. Rows correspond to cells and columns to genes.
- n_comps
Optional
[int
] (default:None
) Number of principal components to compute. Defaults to 50, or 1 - minimum dimension size of selected representation.
- layer
Optional
[str
] (default:None
) If provided, which element of layers to use for PCA.
- zero_center
Optional
[bool
] (default:True
) If True, compute standard PCA from covariance matrix. If False, omit zero-centering variables (uses scikit-learn
TruncatedSVD
or dask-mlTruncatedSVD
), which allows to handle sparse input efficiently. Passing None decides automatically based on sparseness of the data.- svd_solver
Optional
[str
] (default:None
) SVD solver to use:
- None
See chunked and zero_center descriptions to determine which class will be used. Depending on the class and the type of X different values for default will be set. If scikit-learn
PCA
is used, will give ‘arpack’, if scikit-learnTruncatedSVD
is used, will give ‘randomized’, if dask-mlPCA
orIncrementalPCA
is used, will give ‘auto’, if dask-mlTruncatedSVD
is used, will give ‘tsqr’- ’arpack’
for the ARPACK wrapper in SciPy (
svds()
) Not available with dask arrays.- ’randomized’
for the randomized algorithm due to Halko (2009). For dask arrays, this will use
svd_compressed()
.- ’auto’
chooses automatically depending on the size of the problem.
- ’lobpcg’
An alternative SciPy solver. Not available with dask arrays.
- ’tsqr’
Only available with dask arrays. “tsqr” algorithm from Benson et. al. (2013).
Changed in version 1.9.3: Default value changed from ‘arpack’ to None.
Changed in version 1.4.5: Default value changed from ‘auto’ to ‘arpack’.
Efficient computation of the principal components of a sparse matrix currently only works with the ‘arpack’ or ‘lobpcg’ solvers.
If X is a dask array, dask-ml classes
PCA
,IncrementalPCA
, orTruncatedSVD
will be used. Otherwise their scikit-learn counterpartsPCA
,IncrementalPCA
, orTruncatedSVD
will be used.- random_state
Union
[None
,int
,RandomState
] (default:0
) Change to use different initial states for the optimization.
- return_info
bool
(default:False
) Only relevant when not passing an
AnnData
: see “Returns”.- mask
UnionType
[ndarray
,str
,None
,Empty
] (default:_empty
) To run only on a certain set of genes given by a boolean array or a string referring to an array in
var
. By default, uses .var[‘highly_variable’] if available, else everything.- use_highly_variable
Optional
[bool
] (default:None
) Whether to use highly variable genes only, stored in .var[‘highly_variable’]. By default uses them if they have been determined beforehand.
Deprecated since version 1.10.0: Use mask instead
- layer
Layer of adata to use as expression values.
- dtype
str
(default:'float32'
) Numpy data type string to which to convert the result.
- copy
bool
(default:False
) If an
AnnData
is passed, determines whether a copy is returned. Is ignored otherwise.- chunked
bool
(default:False
) If True, perform an incremental PCA on segments of chunk_size. The incremental PCA automatically zero centers and ignores settings of random_seed and svd_solver. Uses sklearn
IncrementalPCA
or dask-mlIncrementalPCA
. If False, perform a full PCA and use sklearnPCA
or dask-mlPCA
- chunk_size
Optional
[int
] (default:None
) Number of observations to include in each chunk. Required if chunked=True was passed.
- data
- Return type:
- Returns:
If data is array-like and return_info=False was passed, this function returns the PCA representation of data as an array of the same type as the input array.
Otherwise, it returns None if copy=False, else an updated AnnData object. Sets the following fields:
- .obsm[‘X_pca’]
spmatrix
|ndarray
(shape (adata.n_obs, n_comps)) PCA representation of data.
- .varm[‘PCs’]
ndarray
(shape (adata.n_vars, n_comps)) The principal components containing the loadings.
- .uns[‘pca’][‘variance_ratio’]
ndarray
(shape (n_comps,)) Ratio of explained variance.
- .uns[‘pca’][‘variance’]
ndarray
(shape (n_comps,)) Explained variance, equivalent to the eigenvalues of the covariance matrix.
- .obsm[‘X_pca’]