scanpy.api.pp.filter_genes_dispersion(data, flavor='seurat', min_disp=None, max_disp=None, min_mean=None, max_mean=None, n_bins=20, n_top_genes=None, log=True, subset=True, copy=False)

Extract highly variable genes [Satija15] [Zheng17].


Deprecated since version 1.3.6: Use highly_variable_genes() instead. The new function is equivalent to the present function, except that

  • the new function always expects logarithmized data

  • subset=False in the new function, it suffices to merely annotate the genes, tools like pp.pca will detect the annotation

  • you can now call:

  • copy is replaced by inplace

If trying out parameters, pass the data matrix instead of AnnData.

Depending on flavor, this reproduces the R-implementations of Seurat [Satija15] and Cell Ranger [Zheng17].

The normalized dispersion is obtained by scaling with the mean and standard deviation of the dispersions for genes falling into a given bin for mean expression of genes. This means that for each bin of mean expression, highly variable genes are selected.

Use flavor='cell_ranger' with care and in the same way as in recipe_zheng17().

data : AnnData, np.ndarray, sp.sparse

The (annotated) data matrix of shape n_obs × n_vars. Rows correspond to cells and columns to genes.

flavor : {'seurat', 'cell_ranger'}, optional (default: 'seurat')

Choose the flavor for computing normalized dispersion. If choosing ‘seurat’, this expects non-logarithmized data - the logarithm of mean and dispersion is taken internally when log is at its default value True. For ‘cell_ranger’, this is usually called for logarithmized data - in this case you should set log to False. In their default workflows, Seurat passes the cutoffs whereas Cell Ranger passes n_top_genes.

max_mean=3, min_disp=0.5, max_disp=`None` : min_mean=0.0125,

If n_top_genes unequals None, these cutoffs for the means and the normalized dispersions are ignored.

n_bins : int (default: 20)

Number of bins for binning the mean gene expression. Normalization is done with respect to each bin. If just a single gene falls into a bin, the normalized dispersion is artificially set to 1. You’ll be informed about this if you set settings.verbosity = 4.

n_top_genes : int or None (default: None)

Number of highly-variable genes to keep.

log : bool, optional (default: True)

Use the logarithm of the mean to variance ratio.

subset : bool, optional (default: True)

Keep highly-variable genes only (if True) else write a bool array for h ighly-variable genes while keeping all genes

copy : bool, optional (default: False)

If an AnnData is passed, determines whether a copy is returned.


If an AnnData adata is passed, returns or updates adata depending on copy. It filters the adata and adds the annotations


Means per gene. Logarithmized when log is True.


Dispersions per gene. Logarithmized when log is True.


Normalized dispersions per gene. Logarithmized when log is True.

If a data matrix X is passed, the annotation is returned as np.recarray with the same information stored in fields: gene_subset, means, dispersions, dispersion_norm.