qptif to sdata

config/ihc_prism/config_v0.toml

@@ -0,0 +1,11 @@
+[reader]
+type = "qptiff"
+
+[reader.channels_renaming]
+'DAPI MSI' = 'DAPI'
+'Opal 780 MSI' = 'CK'
+'Opal 480 MSI' = 'CD3'
+'Cy5 MSI' = 'CD31'
+'FITC' = 'FAP'
+'Cy3' = 'CD68'
+'Texas Red' = 'CD20'

pyproject.toml

@@ -0,0 +1,47 @@
+[tool.poetry]
+name = "sopa"
+version = "0.0.1"
+description = "Spatial-omics preprocessing and analysis"
+authors = ["Quentin Blampey <[email protected]>"]
+packages = [{ include = "sopa" }]
+
+[tool.poetry.dependencies]
+python = ">=3.9,<3.11"
+botocore = "^1.31.47"
+spatialdata = "^0.0.12"
+
+[tool.poetry.group.dev.dependencies]
+black = "^22.8.0"
+isort = "^5.10.1"
+pytest = "^7.1.3"
+ipykernel = "^6.25.2"
+
+[build-system]
+requires = ["poetry-core>=1.0.0"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+python_files = "test_*.py"
+
+[tool.black]
+line-length = 100
+include = '\.pyi?$'
+exclude = '''
+/(
+      \.eggs         # exclude a few common directories in the
+    | \.git          # root of the project
+    | \.hg
+    | \.mypy_cache
+    | \.tox
+    | \.venv
+    | _build
+    | buck-out
+    | build
+    | dist
+  )/
+'''
+
+[tool.isort]
+profile = "black"
+skip_glob = ["*/__init__.py"]

setup.py

@@ -0,0 +1,4 @@
+from setuptools import setup
+
+if __name__ == "__main__":
+    setup()

sopa/io/explorer/categories.py

@@ -1,5 +1,5 @@
+import anndata
 import numpy as np
-import pandas as pd
 import zarr
 
 
@@ -15,7 +15,7 @@ def add_group(root: zarr.Group, index: int, values: np.ndarray, categories: list
     group.array("indptr", indptr, dtype="uint32", chunks=(len(indptr),))
 
 
-def write_groups(path: str, df: pd.DataFrame):
+def write_groups(path: str, adata: anndata.AnnData):
     ATTRS = {
         "major_version": 1,
         "minor_version": 0,
@@ -24,18 +24,19 @@ def write_groups(path: str, df: pd.DataFrame):
         "group_names": [],
     }
 
+    categorical_columns = [
+        name for name, cat in adata.obs.dtypes.items() if cat == "category"
+    ]
+
     with zarr.ZipStore(path, mode="w") as store:
         g = zarr.group(store=store)
         cell_groups = g.create_group("cell_groups")
 
-        i = 0
-        for name in df.columns:
-            if df[name].dtype == "category":
-                categories = list(df[name].cat.categories)
-                ATTRS["grouping_names"].append(name)
-                ATTRS["group_names"].append(categories)
+        for i, name in enumerate(categorical_columns):
+            categories = list(adata.obs[name].cat.categories)
+            ATTRS["grouping_names"].append(name)
+            ATTRS["group_names"].append(categories)
 
-                add_group(cell_groups, i, df[name], categories)
-                i += 1
+            add_group(cell_groups, i, adata.obs[name], categories)
 
         cell_groups.attrs.put(ATTRS)

sopa/io/qptif.py

@@ -3,52 +3,55 @@
 import shutil
 from pathlib import Path
 
+import dask.array as da
 import tifffile as tf
-
-from .explorer import write_ome_tif
+import toml
+import xarray as xr
+from spatialdata import SpatialData
+from spatialdata.models import Image2DModel
+from spatialdata.transformations import Identity
 
 
 def get_channel_name(description):
     return re.search(r"<Name>(.*?)</Name>", description).group(1)
 
 
-def read_series(path: Path) -> list[tf.TiffPageSeries]:
-    with tf.TiffFile(path) as tif:
-        return list(reversed(sorted(tif.series[0], key=lambda p: p.size)))
-
+def read_qptiff(
+    path: Path, channels_renaming: dict | None = None, image_models_kwargs: dict | None = None
+) -> SpatialData:
+    image_models_kwargs = {} if image_models_kwargs is None else image_models_kwargs
+    if "chunks" not in image_models_kwargs:
+        image_models_kwargs["chunks"] = (1, 4096, 4096)
 
-def write_zarr(
-    path: Path,
-    series: list[tf.TiffPageSeries],
-    names: list[str],
-    overwrite: bool = True,
-) -> None:
-    import dask.array as da
-    import xarray as xr
+    with tf.TiffFile(path) as tif:
+        page_series = tif.series[0]
+        names = [get_channel_name(page.description) for page in page_series]
 
-    dask_array = da.asarray(series[0].asarray())
-    xarr = xr.DataArray(
-        dask_array, dims=list(series[0]._axes.lower()), coords={"c": names}
-    )
-    ds = xr.Dataset({"image": xarr})
+        if channels_renaming is not None:
+            names = [channels_renaming[name] for name in names]
 
-    if path.exists():
-        assert overwrite, f"Path {path} exists and overwrite is False"
-        shutil.rmtree(path)
+        image_name = Path(path).absolute().stem
+        image = Image2DModel.parse(
+            da.from_array(page_series.asarray(), chunks=image_models_kwargs["chunks"]),
+            dims=list(page_series._axes.lower()),
+            transformations={"pixels": Identity()},
+            c_coords=names,
+            **image_models_kwargs,
+        )
 
-    print("Saving xarray")
-    ds.to_zarr(path)
+        return SpatialData(images={image_name: image})
 
 
 def main(args):
-    path, output = Path(args.path), Path(args.output)
+    path = Path(args.path)
+    output = path.with_suffix(".zarr")
 
     assert not output.exists(), f"Output path {output} already exists"
 
-    series = read_series(path)
-    names = [get_channel_name(page.description) for page in series[0]._pages]
+    config = toml.load(args.config)
 
-    write_ome_tif(output, series, names)
+    sdata = read_qptiff(path, channels_renaming=config["reader"]["channels_renaming"])
+    sdata.write(output)
 
 
 if __name__ == "__main__":
@@ -61,11 +64,11 @@ def main(args):
         help="Path to the qptiff file",
     )
     parser.add_argument(
-        "-o",
-        "--output",
+        "-c",
+        "--config",
         type=str,
         required=True,
-        help="Path to the morphology.ome.tif file",
+        help="Path to the config file",
     )
 
     main(parser.parse_args())