Custom Segmentation

In [1]:

Copied!

%load_ext autoreload
%autoreload 2
%env ANYWIDGET_HMR=1
%load_ext autoreload
%autoreload 2
%env ANYWIDGET_HMR=1

env: ANYWIDGET_HMR=1

In [2]:

Copied!

import celldega as dega
import os
dega.__version__
import celldega as dega
import os
dega.__version__

Out[2]:

'0.4.0'

Download data if needed¶

Data source: https://www.10xgenomics.com/datasets/xenium-prime-ffpe-human-skin

In [3]:

Copied!

# ! curl -o data/Xenium_Prime_Human_Skin_FFPE_outs.zip https://cf.10xgenomics.com/samples/xenium/3.0.0/Xenium_Prime_Human_Skin_FFPE/Xenium_Prime_Human_Skin_FFPE_outs.zip
# ! unzip data/Xenium_Prime_Human_Skin_FFPE_outs.zip -d data/Xenium_Prime_Human_Skin_FFPE_outs
# ! curl -o data/Xenium_Prime_Human_Skin_FFPE_outs.zip https://cf.10xgenomics.com/samples/xenium/3.0.0/Xenium_Prime_Human_Skin_FFPE/Xenium_Prime_Human_Skin_FFPE_outs.zip
# ! unzip data/Xenium_Prime_Human_Skin_FFPE_outs.zip -d data/Xenium_Prime_Human_Skin_FFPE_outs

Add custom segmentation¶

In [ ]:

Copied!

path_landscape_files="data/xenium_landscape_files/Xenium_Prime_Human_Skin_FFPE_outs"
path_segmentation_files="data/processed_data/xenium_skin/cellpose2/"

dega.pre.add_custom_segmentation(path_landscape_files=path_landscape_files,
                                 path_segmentation_files=path_segmentation_files)
path_landscape_files="data/xenium_landscape_files/Xenium_Prime_Human_Skin_FFPE_outs"
path_segmentation_files="data/processed_data/xenium_skin/cellpose2/"

dega.pre.add_custom_segmentation(path_landscape_files=path_landscape_files,
                                 path_segmentation_files=path_segmentation_files)

========Write meta gene files========
cbg is a dense DataFrame. Proceeding with dense operations.
Calculating mean expression
Calculating variance
All meta gene files are succesfully saved.
data/xenium_landscape_files/Xenium_Prime_Human_Skin_FFPE_outs/cbg_cellpose2

========Write gene-specific parquet files========
Processing gene 0: A2ML1
Processing gene 100: ADIPOR1
Processing gene 200: ANKRD40
Processing gene 300: ATF2
Processing gene 400: BDNF
Processing gene 500: CALCRL
Processing gene 600: CCL20
Processing gene 700: CD72
Processing gene 800: CFC1
Processing gene 900: CNKSR3
Processing gene 1000: CSNK1A1
Processing gene 1100: CYTH2
Processing gene 1200: DMKN
Processing gene 1300: DeprecatedCodeword_0994
Processing gene 1400: DeprecatedCodeword_15617
Processing gene 1500: DeprecatedCodeword_4666
Processing gene 1600: ECD
Processing gene 1700: EPCAM
Processing gene 1800: FANCC
Processing gene 1900: FMO3
Processing gene 2000: GALNS
Processing gene 2100: GPATCH11
Processing gene 2200: H3F3B
Processing gene 2300: HOXB9
Processing gene 2400: IFNW1
Processing gene 2500: INCA1
Processing gene 2600: JAM2
Processing gene 2700: KIR3DL1
Processing gene 2800: LILRA6
Processing gene 2900: MALL
Processing gene 3000: MEST
Processing gene 3100: MTCH2
Processing gene 3200: NCSTN
Processing gene 3300: NORAD
Processing gene 3400: NXPH2
Processing gene 3500: NegControlCodeword_18746
Processing gene 3600: NegControlCodeword_18846
Processing gene 3700: NegControlCodeword_18946
Processing gene 3800: NegControlCodeword_19046
Processing gene 3900: NegControlCodeword_19146
Processing gene 4000: NegControlCodeword_19246
Processing gene 4100: P2RX1
Processing gene 4200: PDE6H
Processing gene 4300: PKIA
Processing gene 4400: PPARD
Processing gene 4500: PRXL2A
Processing gene 4600: RABL2B
Processing gene 4700: RGN
Processing gene 4800: RUBCN
Processing gene 4900: SERPINA9
Processing gene 5000: SLC17A8
Processing gene 5100: SMC1A
Processing gene 5200: SPATS2L
Processing gene 5300: STX7
Processing gene 5400: TENT5B
Processing gene 5500: TMEM130
Processing gene 5600: TPX2
Processing gene 5700: TUFM
Processing gene 5800: UnassignedCodeword_0100
Processing gene 5900: UnassignedCodeword_0579
Processing gene 6000: UnassignedCodeword_0998
Processing gene 6100: UnassignedCodeword_10427
Processing gene 6200: UnassignedCodeword_10846
Processing gene 6300: UnassignedCodeword_11242
Processing gene 6400: UnassignedCodeword_11661
Processing gene 6500: UnassignedCodeword_12151
Processing gene 6600: UnassignedCodeword_12526
Processing gene 6700: UnassignedCodeword_12880
Processing gene 6800: UnassignedCodeword_13294
Processing gene 6900: UnassignedCodeword_13728
Processing gene 7000: UnassignedCodeword_14115
Processing gene 7100: UnassignedCodeword_14486
Processing gene 7200: UnassignedCodeword_1487
Processing gene 7300: UnassignedCodeword_15237
Processing gene 7400: UnassignedCodeword_15662
Processing gene 7500: UnassignedCodeword_16071
Processing gene 7600: UnassignedCodeword_16476
Processing gene 7700: UnassignedCodeword_16830
Processing gene 7800: UnassignedCodeword_17241
Processing gene 7900: UnassignedCodeword_17672
Processing gene 8000: UnassignedCodeword_18196
Processing gene 8100: UnassignedCodeword_1886
Processing gene 8200: UnassignedCodeword_2384
Processing gene 8300: UnassignedCodeword_2846
Processing gene 8400: UnassignedCodeword_3332
Processing gene 8500: UnassignedCodeword_3845
Processing gene 8600: UnassignedCodeword_4262
Processing gene 8700: UnassignedCodeword_4726
Processing gene 8800: UnassignedCodeword_5195
Processing gene 8900: UnassignedCodeword_5693
Processing gene 9000: UnassignedCodeword_6174
Processing gene 9100: UnassignedCodeword_6629
Processing gene 9200: UnassignedCodeword_7161
Processing gene 9300: UnassignedCodeword_7569
Processing gene 9400: UnassignedCodeword_8046
Processing gene 9500: UnassignedCodeword_8534
Processing gene 9600: UnassignedCodeword_9011
Processing gene 9700: UnassignedCodeword_9525
Processing gene 9800: UnassignedCodeword_9939
Processing gene 9900: XPO1
Processing gene 10000: ZNF687
All gene-specific parquet files are succesfully saved.

========Make meta cells in pixel space========
Done.

========Create clusters and meta clusters files========
Cell clusters and meta cluster files created successfully.

========Create cell boundary spatial tiles========

Processing coarse tiles: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 9/9 [00:15<00:00,  1.77s/it]

Done.

========Create cluster gene expression (df_sig)========
Cluster-specific gene expression signatures saved successfully.

========Save landscape parameters========
Done.

Visualize Custom Segmentation Landscape files in Celldega¶

In [5]:

Copied!





base_url = 'https://raw.githubusercontent.com/broadinstitute/celldega_custom_cellpose2_Xenium_Prime_Human_Skin_FFPE_outs/main/Xenium_Prime_Human_Skin_FFPE_outs'

landscape_ist = dega.viz.Landscape(
    technology='Xenium',
    base_url = base_url,
    segmentation='cellpose2'
)

landscape_ist
base_url = 'https://raw.githubusercontent.com/broadinstitute/celldega_custom_cellpose2_Xenium_Prime_Human_Skin_FFPE_outs/main/Xenium_Prime_Human_Skin_FFPE_outs'

landscape_ist = dega.viz.Landscape(
    technology='Xenium',
    base_url = base_url,
    segmentation='cellpose2'
)

landscape_ist

Out[5]:

Landscape(base_url='https://raw.githubusercontent.com/broadinstitute/celldega_custom_cellpose2_Xenium_Prime_Hu…

In [ ]: