%matplotlib inline

Calculate distances to a user-defined anchor point

This example shows how to use squidpy.tl.var_by_distance() to calculate the minimum distances of all observations to a user-defined anchor point, store the results in anndata.AnnData.obsm and plot the expression by distance. using squidpy.pl.var_by_distance().

import squidpy as sq

First, let’s download the MIBI-TOF dataset.

adata = sq.datasets.mibitof()

This data set contains a cell type annotation in anndata.AnnData.obs["Cluster"] and a slide annotation in anndata.AnnData.obs["library_id"]

adata.obs

	row_num	point	cell_id	X1	center_rowcoord	center_colcoord	cell_size	category	donor	Cluster	batch	library_id
3034-0	3086	23	2	60316.0	269.0	7.0	408.0	carcinoma	21d7	Epithelial	0	point23
3035-0	3087	23	3	60317.0	294.0	6.0	408.0	carcinoma	21d7	Epithelial	0	point23
3036-0	3088	23	4	60318.0	338.0	4.0	304.0	carcinoma	21d7	Imm_other	0	point23
3037-0	3089	23	6	60320.0	372.0	6.0	219.0	carcinoma	21d7	Myeloid_CD11c	0	point23
3038-0	3090	23	8	60322.0	417.0	5.0	303.0	carcinoma	21d7	Myeloid_CD11c	0	point23
...	...	...	...	...	...	...	...	...	...	...	...	...
47342-2	48953	16	1103	2779.0	143.0	1016.0	283.0	carcinoma	90de	Fibroblast	2	point16
47343-2	48954	16	1104	2780.0	814.0	1017.0	147.0	carcinoma	90de	Fibroblast	2	point16
47344-2	48955	16	1105	2781.0	874.0	1018.0	142.0	carcinoma	90de	Imm_other	2	point16
47345-2	48956	16	1106	2782.0	257.0	1019.0	108.0	carcinoma	90de	Fibroblast	2	point16
47346-2	48957	16	1107	2783.0	533.0	1019.0	111.0	carcinoma	90de	Fibroblast	2	point16

3309 rows × 12 columns

For each slide we now want to calculate the distance of all observations to the closest Epithelial cell. In addition we want to include the condition of the donors and the donor id in the resulting design matrix As we don’t create a copy, the result will be stored in anndata.AnnData.obsm.

sq.tl.var_by_distance(
    adata=adata,
    groups="Epithelial",
    cluster_key="Cluster",
    library_key="library_id",
    covariates=["category", "donor"],
)

Since we didn’t specify a name, the resulting data frame is called “design_matrix”. NaN values indicate, that the observation belongs to an anchor point or that the coordinates for this observation weren’t available from the beginning on.

adata.obsm["design_matrix"]

	Cluster	library_id	Epithelial	Epithelial_raw	category	donor
3034-0	Epithelial	point23	NaN	0.000000	carcinoma	21d7
3035-0	Epithelial	point23	NaN	0.000000	carcinoma	21d7
3036-0	Imm_other	point23	0.043157	33.105891	carcinoma	21d7
3037-0	Myeloid_CD11c	point23	0.066190	50.774009	carcinoma	21d7
3038-0	Myeloid_CD11c	point23	0.109999	84.380092	carcinoma	21d7
...	...	...	...	...	...	...
47342-2	Fibroblast	point16	0.849905	651.958588	carcinoma	90de
47343-2	Fibroblast	point16	0.418362	320.923667	carcinoma	90de
47344-2	Imm_other	point16	0.439758	337.336627	carcinoma	90de
47345-2	Fibroblast	point16	0.724977	556.126784	carcinoma	90de
47346-2	Fibroblast	point16	0.473780	363.435001	carcinoma	90de

3309 rows × 6 columns

We now want to visualize the results and plot the expression of CD98 by distance to the closest Epithelial cell. In addition we want to differentiate between two expression trends by specifying a covariate.

sq.pl.var_by_distance(
    adata=adata,
    design_matrix_key="design_matrix",
    var="CD98",
    anchor_key="Epithelial",
    covariate="donor",
    line_palette=["blue", "orange"],
    show_scatter=False,
    figsize=(5, 4),
)