Add basic tests for the main class

Author: Marius1311

tests/conftest.py

@@ -1,5 +1,11 @@
+from importlib.resources import files
+
 import numpy as np
+import pandas as pd
 import pytest
+import scanpy as sc
+
+from cellmapper.cellmapper import CellMapper
 
 
 @pytest.fixture
@@ -20,3 +26,105 @@ def small_data():
     x = np.array([[0, 0], [1, 0], [0, 1], [1, 1], [2, 2]], dtype=np.float64)
     y = np.array([[0, 0], [1, 0], [0, 1], [1, 1], [2, 2]], dtype=np.float64)
     return x, y
+
+
+@pytest.fixture
+def precomputed_leiden():
+    """Fixture to load precomputed leiden clustering."""
+    data_path = files("tests.data") / "precomputed_leiden.csv"
+    leiden_cl = pd.read_csv(str(data_path), index_col=0)["leiden"]
+
+    return leiden_cl
+
+
+@pytest.fixture
+def adata_pbmc3k(precomputed_leiden):
+    adata = sc.datasets.pbmc3k()
+
+    # basic cell and gene filtering
+    sc.pp.filter_cells(adata, min_genes=200)
+    sc.pp.filter_genes(adata, min_cells=3)
+
+    # Saving count data
+    adata.layers["counts"] = adata.X.copy()
+
+    # Normalizing to median total counts
+    sc.pp.normalize_total(adata)
+    # Logarithmize the data
+    sc.pp.log1p(adata)
+
+    # compute hvgs
+    sc.pp.highly_variable_genes(adata, n_top_genes=2000)
+
+    # PCA
+    sc.tl.pca(adata, mask_var="highly_variable")
+
+    # Load precomputed leiden clustering
+    adata.obs["leiden"] = precomputed_leiden.astype("str").astype("category")
+
+    return adata
+
+
+@pytest.fixture
+def query_ref_adata(adata_pbmc3k):
+    # Define the number of query cells and genes
+    n_query_cells = 500
+    n_query_genes = 300
+    n_ref_cells = adata_pbmc3k.n_obs - n_query_cells
+
+    # Create modality annotations in the AnnData object
+    adata_pbmc3k.obs["modality"] = (
+        np.repeat("query", repeats=n_query_cells).tolist() + np.repeat("ref", repeats=n_ref_cells).tolist()
+    )
+    adata_pbmc3k.obs["modality"] = adata_pbmc3k.obs["modality"].astype("category")
+
+    # use these annotations to create query and reference AnnData objects
+    query = adata_pbmc3k[adata_pbmc3k.obs["modality"] == "query"].copy()
+    ref = adata_pbmc3k[adata_pbmc3k.obs["modality"] == "ref"].copy()
+
+    # Subset genes in the query AnnData object
+    query_genes = (
+        query.var[query.var["highly_variable"]].sort_values("means", ascending=False).head(n_query_genes).index.tolist()
+    )
+    query = query[:, query_genes].copy()
+
+    return query, ref
+
+
+@pytest.fixture
+def cmap(query_ref_adata):
+    query, ref = query_ref_adata
+
+    # Create a CellMapper object
+    cmap = CellMapper(
+        query=query,
+        ref=ref,
+    )
+
+    # Compute neighbors and mapping matrix
+    cmap.compute_neighbors(n_neighbors=30, use_rep="X_pca", method="sklearn")
+    cmap.compute_mappping_matrix(method="gaussian")
+
+    return cmap
+
+
+@pytest.fixture
+def expected_label_transfer_metrics():
+    return {
+        "accuracy": 0.954,
+        "precision": 0.955,
+        "recall": 0.954,
+        "f1_weighted": 0.951,
+        "f1_macro": 0.899,
+        "excluded_fraction": 0.0,
+    }
+
+
+@pytest.fixture
+def expected_expression_transfer_metrics():
+    return {
+        "method": "pearson",
+        "average_correlation": 0.376,
+        "n_genes": 300,
+        "n_valid_genes": 300,
+    }