Add reader for Stereo-seq files.

CHANGELOG.md

@@ -14,6 +14,7 @@ and this project adheres to [Semantic Versioning][].
 
 -   (Xenium) support reading multi-polygon selection files from the Xenium Explorer
 -   (ISS) An experimental loader to load elemental ISS data objects, e.g. raw.tif, label.tif and anndata.h5ad
+-   (Stereo-seq) Added reader @LLehner @timtreis @florianingelfinger #70
 
 ### Fixed
 

README.md

@@ -21,6 +21,7 @@ This package contains reader functions to load common spatial omics formats into
 -   MCMICRO (output data)
 -   NanoString CosMx
 -   Steinbock (output data)
+-   STOmics Stereo-seq
 -   Vizgen MERSCOPE (MERFISH)
 
 ## Getting started

docs/api.md

@@ -21,6 +21,7 @@ I/O for the `spatialdata` project.
     mcmicro
     merscope
     steinbock
+    stereoseq
     visium
     visium_hd
     xenium

pyproject.toml

@@ -25,7 +25,7 @@ dependencies = [
     "anndata",
     "numpy",
     "scanpy",
-    "spatialdata",
+    "spatialdata>=0.1.2",
     "scikit-image",
     "h5py",
     "joblib",

src/spatialdata_io/__init__.py

@@ -7,6 +7,7 @@
 from spatialdata_io.readers.mcmicro import mcmicro
 from spatialdata_io.readers.merscope import merscope
 from spatialdata_io.readers.steinbock import steinbock
+from spatialdata_io.readers.stereoseq import stereoseq
 from spatialdata_io.readers.visium import visium
 from spatialdata_io.readers.visium_hd import visium_hd
 from spatialdata_io.readers.xenium import (
@@ -23,6 +24,7 @@
     "cosmx",
     "mcmicro",
     "steinbock",
+    "stereoseq",
     "merscope",
     "xenium_aligned_image",
     "xenium_explorer_selection",

src/spatialdata_io/_constants/_constants.py

@@ -165,6 +165,79 @@ class SteinbockKeys(ModeEnum):
     LABEL_SUFFIX = ".tiff"
 
 
+@unique
+class StereoseqKeys(ModeEnum):
+    """Keys for *Stereo-seq* formatted dataset."""
+
+    # file extensions
+    GEF_FILE = ".gef"
+    TIF_FILE = ".tif"
+    # directories
+    MERGE = "01.merge"
+    COUNT_DIRECTORY = "02.count"
+    REGISTER = "03.register"
+    TISSUECUT = "04.tissuecut"
+    SPATIALCLUSTER = "05.spatialcluster"
+    SATURATION = "06.saturation"
+    CELLCUT = "041.cellcut"
+    CELLCLUSTER = "051.cellcluster"
+    # file identifiers
+    MASK_TIF = "_mask.tif"
+    REGIST_TIF = "_regist.tif"
+    TISSUE_TIF = "_tissue_cut.tif"
+    CELL_CLUSTER_H5AD = "cell.cluster.h5ad"
+    RAW_GEF = ".raw.gef"
+    CELLBIN_GEF = ".cellbin.gef"
+    TISSUECUT_GEF = ".tissuecut.gef"
+    TISSUE_GEM = ".tissue.gem.gz"
+    # transcripts
+    FEATURE_KEY = "gene"
+    GENE_NAME = "geneName"
+    CELL_COUNT = "cellCount"
+    MAX_MID_COUNT = "maxMIDcount"
+    GENE_EXP = "geneExp"
+    EXPRESSION = "expression"
+    EXON = "exon"
+    GENE_ID = "geneID"
+    GENE_EXP_EXON = "geneExpExon"
+    # cells
+    ID = "id"
+    CELL_ID = "cellID"
+    GENE_COUNT = "geneCount"
+    DNBCOUNT = "dnbCount"
+    CELL_AREA = "area"
+    CELL_TYPE_ID = "cellTypeID"
+    CLUSTER_ID = "clusterID"
+    CELL_BIN = "cellBin"
+    CELL_EXON = "cellExon"
+    CELL_DATASET = "cell"
+    GENE_EXON = "geneExon"
+    CELL_BORDER = "cellBorder"
+    CELL_EXP = "cellExp"
+    CELL_EXP_EXON = "cellExpExon"
+    PADDING_VALUE = 32767
+    # metadata
+    COUNT = "count"
+    EXP_COUNT = "expCount"
+    OFFSET = "offset"
+    COORD_X = "x"
+    COORD_Y = "y"
+    SPATIAL_KEY = "spatial"
+    REGION = "cells"
+    REGION_KEY = "region"
+    INSTANCE_KEY = "instance_id"
+    BLOCK_INDEX = "blockIndex"
+    BLOCK_SIZE = "blockSize"
+    CELL_TYPE_LIST = "cellTypeList"
+    # bin data and metadata
+    BIN1 = "bin1"
+    MIN_X = "minX"
+    MIN_Y = "minY"
+    MAX_X = "maxX"
+    MAX_Y = "maxY"
+    RESOLUTION = "resolution"
+
+
 @unique
 class McmicroKeys(ModeEnum):
     """Keys for *Mcmicro* formatted dataset."""

src/spatialdata_io/readers/_utils/_utils.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import os
+from collections.abc import Mapping
 from pathlib import Path
 from typing import Any, Optional, Union
 
@@ -66,3 +67,20 @@ def _read_counts(
 
     adata.uns["spatial"] = {library_id: {"metadata": {}}}  # can overwrite
     return adata, library_id
+
+
+def _initialize_raster_models_kwargs(
+    image_models_kwargs: Mapping[str, Any], labels_models_kwargs: Mapping[str, Any]
+) -> tuple[dict[str, Any], dict[str, Any]]:
+    image_models_kwargs = dict(image_models_kwargs)
+    if "chunks" not in image_models_kwargs:
+        image_models_kwargs["chunks"] = (1, 4096, 4096)
+    if "scale_factors" not in image_models_kwargs:
+        image_models_kwargs["scale_factors"] = [2, 2, 2, 2]
+
+    labels_models_kwargs = dict(labels_models_kwargs)
+    if "chunks" not in labels_models_kwargs:
+        labels_models_kwargs["chunks"] = (4096, 4096)
+    if "scale_factors" not in labels_models_kwargs:
+        labels_models_kwargs["scale_factors"] = [2, 2, 2, 2]
+    return image_models_kwargs, labels_models_kwargs