Feature/hucira

pertpy/data/__init__.py

@@ -1,4 +1,5 @@
 from pertpy.data._datasets import (
+    MS_CSF_tutorial_data,
     adamson_2016_pilot,
     adamson_2016_upr_epistasis,
     adamson_2016_upr_perturb_seq,
@@ -25,6 +26,7 @@
     gehring_2019,
     haber_2017_regions,
     hagai_2018,
+    human_cytokine_dict,
     kang_2018,
     mcfarland_2020,
     norman_2019,
@@ -84,8 +86,10 @@
     "gehring_2019",
     "haber_2017_regions",
     "hagai_2018",
+    "human_cytokine_dict",
     "kang_2018",
     "mcfarland_2020",
+    "MS_CSF_tutorial_data",
     "norman_2019",
     "norman_2019_raw",
     "papalexi_2021",

pertpy/data/_datasets.py

@@ -1,5 +1,6 @@
 from pathlib import Path
 
+import pandas as pd
 import scanpy as sc
 from anndata import AnnData
 from mudata import MuData
@@ -1598,3 +1599,70 @@ def hagai_2018() -> AnnData:  # pragma: no cover
     adata = sc.read_h5ad(output_file_path)
 
     return adata
+
+
+def human_cytokine_dict(exclude_well_biased_genes=True) -> pd.DataFrame:
+    r"""Human Cytokine Dictionary curated from PBMC allows you to infer differential cytokine activity.
+
+    The Human Cytokine Dictionary was created from single-cell RNA-seq of 9,697,974 human peripheral blood mononuclear cells (PBMC) from 12 donors stimulated in vitro with 87 different cytokines. The object is a dataframe representing cytokine activity as differentially expressed genes after cytokine perturbation.
+
+    References:
+        Oesinghaus, Lukas and Becker, S{\"o}ren and Vornholz, Larsen
+        .... bla bla coming
+
+    Returns:
+        Pandas DataFrame
+
+    """
+    output_file_name = "human_cytokine_dict.csv"
+    output_file_path = settings.datasetdir / output_file_name
+    if not Path(output_file_path).exists():
+        _download(
+            url="https://cdn.parsebiosciences.com/gigalab/10m/DEGs.csv",
+            output_file_name=output_file_name,
+            output_path=settings.datasetdir,
+            is_zip=False,
+        )
+
+    cytokine_dict = pd.read_csv(output_file_path, index_col=0)
+    revision_cytokines = ["TGF-beta1", "IL-18", "C3a"]
+    cytokine_dict = cytokine_dict[~cytokine_dict["cytokine"].isin(revision_cytokines)]
+    cytokine_dict = cytokine_dict.reset_index(drop=True)
+
+    if exclude_well_biased_genes:
+        cytokine_dict = cytokine_dict.loc[~cytokine_dict.well_biased]
+
+    return cytokine_dict
+
+
+def MS_CSF_tutorial_data(save_dir="", force_download=False):
+    """Multiple Sklerosis Dataset (blood and cerebrospinal fluid) for hucira tutorial.
+
+    Download and load the MS dataset automatically.
+    Xu, Chenling (2021). MS_CSF.h5ad. figshare. Dataset. https://doi.org/10.6084/m9.figshare.14356661.v1
+
+    Parameters
+    ----------
+    save_dir : str
+        Directory where the file will be saved.
+    force_download : bool
+        Allows user to force a fresh download from CellxGene
+
+    Returns:
+    -------
+    adata : AnnData
+        MS adata object.
+    """
+    output_file_name = "MS_CSF.h5ad"
+    output_file_path = settings.datasetdir / output_file_name
+
+    if force_download or not output_file_path.exists():
+        _download(
+            url="https://figshare.com/ndownloader/files/27405182",
+            output_file_name=output_file_name,
+            output_path=settings.datasetdir,
+            is_zip=False,
+        )
+
+    adata = sc.read_h5ad(output_file_path)
+    return adata

pertpy/tools/__init__.py

@@ -7,6 +7,7 @@
 from pertpy.tools._distances._distance_tests import DistanceTest
 from pertpy.tools._distances._distances import Distance
 from pertpy.tools._enrichment import Enrichment
+from pertpy.tools._hucira import Hucira
 from pertpy.tools._milo import Milo
 from pertpy.tools._mixscape import Mixscape
 from pertpy.tools._perturbation_space._clustering import ClusteringSpace
@@ -68,6 +69,7 @@ def __dir__():
     "DistanceTest",
     "Distance",
     "Enrichment",
+    "hucira",
     "Milo",
     "Mixscape",
     "ClusteringSpace",