xenium tutorial

xenium tutorial#

import scispy as scis
import pandas as pd
import seaborn as sns
import scanpy as sc
import anndata as ad
from anndata import AnnData
import spatialdata as sd
from spatialdata import SpatialData
from spatialdata import polygon_query
from matplotlib import pyplot as plt

load raw data and perform scanpy analysis#

sdata = scis.io.load_xenium('xenium_output_dir')
scis.pp.run_scanpy(sdata)

annotate your spatial cells using a single-cell reference#

scref = sc.read_h5ad('scref.h5ad')
scis.pp.scvi_annotate(sdata['table'], 
                      scref, 
                      label_ref='celltype', 
                      label_key='celltype')

add anatomical regions#

scis.tl.add_to_shapes(sdata, 
                      shape_key='shapes', 
                      scale_factor=1/0.2125, 
                      target_coordinates='global', 
                      shape_file='coordinates.csv')

sdata['shapes'].name = sdata['shapes'].name.astype('category')

f, ax = plt.subplots(figsize=(8, 5))
sdata.pl.render_images(cmap='Blues').pl.show(ax=ax)
sdata.pl.render_shapes(elements='shapes', color='name', fill_alpha=0.8).pl.show("global",ax=ax)
ax.set_title('my_sample')
plt.show()

sdata['table'].obs['anatomy'] = 'unknown'
for i, shape in sdata['shapes'].iterrows():
    try:
        sd_shape = polygon_query(sdata, polygon=shape.geometry, target_coordinate_system="global")
        sdata['table'].obs.anatomy[sdata['table'].obs['cell_id'].isin(sd_shape.table.obs['cell_id'])] = shape['name']
    except:
        print(shape['name'], " contains invalid coordinates")

save sdata object#

sdata.write("sdata_1")

#import spatialdata_xenium_explorer
#spatialdata_xenium_explorer.write("explorer_1", sdata, shapes_key='cell_boundaries', points_key='transcripts', gene_column='feature_name', layer='counts', polygon_max_vertices=40, mode='-it')

concatenate sdata objects#

adatas = []
samples = ['sdata_1','sdata_2','sdata_3','sdata_4']
for s in samples:
    sdata = SpatialData.read(s)
    adatas += [sdata.table]

adata = ad.concat(adatas)

visualize cell type proportions#

scis.tl.scis_prop(adata, 
                  celltype='celltype', 
                  zone='anatomy', 
                  top=10, 
                  replicate='sample', 
                  condition='genotype', 
                  condition_order=['KO','CTRL'],
                  figsize=(6,4))

run pseudobulk analysis#

df = scis.tl.run_pseudobulk(sdata['table'], 
                            replicate_key='sample', 
                            cond_key='genotype', 
                            cond_1='KO', 
                            cond_2='CTRL', 
                            groups_key='celltype')

# plot results as seaborn heatmap
pivlfc = pd.pivot_table(df, values=["log2FoldChange"], index=["index"], columns=['celltype'], fill_value=0)
sns.clustermap(pivlfc, cmap="vlag", figsize=(2, 8))
Contents