_Assay

missionbio.mosaic.assay._Assay

class _Assay(*args, **kwargs)

Abstract class for all assays

Each of Dna, Cnv, and Protein are inherited from this base class, hence all of them have the following functionality:

_Assay objects can be filtered using Python’s slice notation. It requires two keys (barcodes, ids). Both could be a list of positions, a Boolean list, or a list of the respective values.

Load the sample.

>>> import missionbio.mosaic.io as mio
>>> sample = mio.load('/path/to/h5')

Select the first 250 cells (these aren’t necessarily the cells with the highest reads, they’re arbitrary cells) and the dna variants obtained from filtering.

>>> select_bars = sample.dna.barcodes()[: 250]
>>> select_vars = sample.dna.filter_variants()

Slice the dna assay for the selected cells and variants.

>>> filtered_dna = sample.dna[select_bars, select_vars]

Note: This example is for DNA, but the same can be done for any _Assay object.

Variables

_Assay.selected_ids	The ids selected on the heatmap
_Assay.selected_bars	The barcodes selected on the scatterplot

Modifying data

rename_sample(old_name, new_name)	Rename sample to a new_name
deduplicate_barcodes()	Appends an integer to the barcode in case of duplication.
normalize_barcodes([allow_duplicates])	Remove all but [^ACGT] from barcodes
add_layer(name, array)	Add layer to the assay
add_row_attr(name, value)	Add row attribute to the assay
add_col_attr(name, value)	Add column attribute to the assay
add_metadata(name, value)	Add metadata to the assay
del_layer(key)	Delete a layer from the assay layers.
del_row_attr(key)	Delete an attribute from the row_attrs.
del_col_attr(key)	Delete an attribute from the col_attrs.
del_metadata(key)	Delete an attribute from the metadata.
has_info(name[, sample, any_sample, all_samples])	Check if the info exists in any of the samples
add_info(name, value[, sample, all_samples])	Add metadata to the assay
del_info(name[, sample, all_samples])	Delets info from the assay for a specific sample
sampleinfo	The info for the only sample in the assay

Selecting data

barcodes([label])	Get the list of barcodes.
clustered_barcodes([orderby, splitby, ...])	Hierarchical clustering of barcodes.
ids()	Get the list of ids.
clustered_ids(orderby[, features, override])	Hierarchical clustering of ids.
set_ids_from_cols(col_names)	Rename the ids based on columns
reset_ids()	Resets ids if they were modified using set_ids_from_cols()
drop(values[, value_type])	Drops the given values from the assay.
get_palette()	Get the label color palette.
set_palette([palette])	Set the label color palette.
get_labels([barcodes])	Get the labels corresponding to the barcodes.
set_labels(clusters[, default_label])	Set the labels for the barcodes.
set_selected_labels()	Custom clustering from scatter plots.
rename_labels(label_map)	Change the name of the labels.
get_attribute(attribute[, constraint, ...])	Retrieve any attribute in the assay
hierarchical_cluster(dat[, index])	Get the hierarchical clustering order of the index of the dataframe.

Transforming data

convolve(data, labels, convolve)	Convolve the data in-place based on the given labels.
scale_data(layer[, output_label])	Z-score normalization
run_pca(attribute, components[, ...])	Principal component analysis
run_umap(attribute[, output_label])	Perform UMAP on the given data.
run_lda(layer, attribute, output_label[, cycles])	Perform LDA on a row attribute using a layer to type clusters.
cluster(attribute[, method])	Identify clusters of cells.
group_clusters(attribute)	Relabel clusters based on expression.

Visualizations

scatterplot(attribute[, colorby, features, ...])	Scatter plot for all barcodes.
heatmap(attribute[, splitby, features, ...])	Heatmap of all barcodes and ids.
signaturemap(attribute[, kind, nan_value, ...])	Flattened heatmap showing the signature profile for each unique label
crosstabmap(columns[, index, normalize, ...])	Crosstab heatmap for all barcodes.
violinplot(attribute[, splitby, features, title])	Violin plot for all barcodes.
ridgeplot(attribute[, splitby, features, ...])	Ridge plot for all barcodes.
stripplot(attribute[, colorby, features, title])	Strip plot for all barcodes.
read_distribution([title])	Rank-ordered barcodes vs total reads.
feature_scatter(layer, ids, **kwargs)	Plot for given data across 2 ids.

Statistics

signature(attribute[, kind, splitby, ...])	The chosen signature for each cluster and feature.
test_signature(attribute[, kind])	Run a statistical test on the clusters
crosstab(columns[, index, normalize])	Compute a simple cross-tabulation of two attributes.

Parameters:

argslist

To be passed to missionbio.h5.data.Assay class.

kwargsdict

To be passed to missionbio.h5.data.Assay class.