GeneExpressionData

class alphagenome.data.gene_expression_data.GeneExpressionData(values, track_metadata, gene_metadata, interval)

Container for storing gene expression data.

Used to store gene-level predictions for denoised single-cell gene expression (SC_GENE_EXPRESSION_DENOISED; one channel per strand, in CPM) and protocol-specific single-cell gene expression (SC_GENE_EXPRESSION, in CPM).

values

A numpy array of floats representing the gene expression for each track.

track_metadata

A pandas DataFrame containing metadata for each track.

gene_metadata

A pandas DataFrame containing metadata for each gene.

interval

An Interval object representing the genomic region containing the genes.

names

Array of track names from track_metadata['name'].

strands

Array of track strands from track_metadata['strand'].

num_tracks

Total number of tracks along axis 1 of values.

Raises:

ValueError – If the number of tracks in values does not match the number of rows in track_metadata, or if the number of gene in values does not match the number of rows in gene_metadata or if track_metadata contains duplicate names.

Methods

copy()	Returns a deep copy of the GeneExpressionData object.
filter_to_negative_strand()	Filters tracks to the negative DNA strand.
filter_to_nonnegative_strand()	Filters tracks to the non-negative DNA strands (positive and unstranded).
filter_to_nonpositive_strand()	Filters tracks to the non-positive DNA strands (negative and unstranded).
filter_to_observed()	Filters tracks to observed tracks.
filter_to_positive_strand()	Filters tracks to the positive DNA strand.
filter_to_protein_coding()	Filters data to protein coding genes only.
filter_to_stranded()	Filters tracks to stranded tracks (excluding unstranded).
filter_to_unstranded()	Filters tracks to unstranded tracks.
filter_tracks(mask)	Filters tracks by a boolean mask.
get_genes_as_intervals()	Returns the gene intervals as a sequence of Interval objects.
resize(width)	Resizes the interval, removing genes falling outside the new width.
select_tracks_by_index(idx)	Selects tracks by numerical index.
slice_by_interval(interval)	Slices the GeneExpressionData to include only genes within an interval.
to_anndata()	Converts the GeneExpressionData to an AnnData object.
with_metadata_column(name, value)	Returns a new GeneExpressionData with updated track metadata.
with_name_suffix(suffix)	Returns a new GeneExpressionData object with suffixed track names.