Don't recompute X_emb and neighborhood graph for baseline datasets

openproblems/tasks/_batch_integration/batch_integration_embed/api.py

@@ -23,4 +23,5 @@ def sample_method(adata):
     """Create sample method output for testing metrics in this task."""
 
     adata.obsm["X_emb"] = adata.obsm["X_uni_pca"]
+    adata.uns["is_baseline"] = False
     return adata

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/ari.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import embedding_to_graph
 
 """
 The Rand index compares the overlap of two clusterings;
@@ -16,7 +17,4 @@
 
 @metric(**graph_metrics.ari.metadata)
 def ari(adata):
-    from scanpy.pp import neighbors
-
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.ari(adata)
+    return graph_metrics.ari(embedding_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/cc_score.py

@@ -1,4 +1,5 @@
 from .....tools.decorators import metric
+from .utils import get_split
 
 """
 The cell-cycle conservation score evaluates how well the cell-cycle effect can be
@@ -24,13 +25,12 @@
     maximize=True,
     image="openproblems-r-pytorch",
 )
-def cc_score(adata, test=False):
-    from ._utils import _get_split
+def cc_score(adata):
     from scib.metrics import cell_cycle
 
     try:
         cc = cell_cycle(
-            *_get_split(adata), "batch", embed="X_emb", organism=adata.uns["organism"]
+            *get_split(adata), "batch", embed="X_emb", organism=adata.uns["organism"]
         )
 
     except ValueError:

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/graph_connectivity.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import embedding_to_graph
 
 """
 The graph connectivity metric assesses whether the kNN graph representation,
@@ -22,7 +23,4 @@
 
 @metric(**graph_metrics.graph_connectivity.metadata)
 def graph_connectivity(adata):
-    from scanpy.pp import neighbors
-
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.graph_connectivity(adata)
+    return graph_metrics.graph_connectivity(embedding_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/iso_label_f1.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import embedding_to_graph
 
 """
 We developed two isolated label scores to evaluate how well the data integration methods
@@ -27,7 +28,4 @@
 
 @metric(**graph_metrics.isolated_labels_f1.metadata)
 def isolated_labels_f1(adata):
-    from scanpy.pp import neighbors
-
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.isolated_labels_f1(adata)
+    return graph_metrics.isolated_labels_f1(embedding_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/nmi.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import embedding_to_graph
 
 """NMI compares the overlap of two clusterings.
 We used NMI to compare the cell-type labels with Louvain clusters computed on
@@ -15,7 +16,4 @@
 
 @metric(**graph_metrics.nmi.metadata)
 def nmi(adata):
-    from scanpy.pp import neighbors
-
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.nmi(adata)
+    return graph_metrics.nmi(embedding_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/pcr.py

@@ -1,4 +1,5 @@
 from .....tools.decorators import metric
+from .utils import get_split
 
 """
 Principal component regression, derived from PCA, has previously been used to quantify
@@ -22,7 +23,6 @@
     image="openproblems-r-pytorch",
 )
 def pcr(adata):
-    from ._utils import _get_split
     from scib.metrics import pcr_comparison
 
-    return pcr_comparison(*_get_split(adata), "batch", embed="X_emb")
+    return pcr_comparison(*get_split(adata), "batch", embed="X_emb")

openproblems/tasks/_batch_integration/batch_integration_embed/metrics/utils.py

@@ -0,0 +1,16 @@
+def embedding_to_graph(adata):
+    import scanpy as sc
+
+    if adata.uns["is_baseline"] and "neighbors" in adata.uns:
+        # precomputed; do nothing
+        return adata
+
+    sc.pp.neighbors(adata, use_rep="X_emb")
+    return adata
+
+
+def get_split(adata):
+    uni = adata
+    uni.obsm["X_pca"] = uni.obsm["X_uni_pca"]
+    uni.X = uni.layers["log_normalized"]
+    return (uni, adata)

openproblems/tasks/_batch_integration/batch_integration_feature/api.py

@@ -29,4 +29,5 @@ def sample_dataset():
 def sample_method(adata):
     """Create sample method output for testing metrics in this task."""
     adata.X = adata.X.multiply(2)
+    adata.uns["is_baseline"] = False
     return adata

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/ari.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import feature_to_graph
 
 """
 The Rand index compares the overlap of two clusterings;
@@ -16,9 +17,4 @@
 
 @metric(**graph_metrics.ari.metadata)
 def ari(adata):
-    from scanpy.pp import neighbors
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.ari(adata)
+    return graph_metrics.ari(feature_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/cc_score.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 The cell-cycle conservation score evaluates how well the cell-cycle effect can be
@@ -20,8 +21,5 @@
 
 
 @metric(**embed_metrics.cc_score.metadata)
-def cc_score(adata, test=False):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.cc_score(adata)
+def cc_score(adata):
+    return embed_metrics.cc_score(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/graph_connectivity.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import feature_to_graph
 
 """
 The graph connectivity metric assesses whether the kNN graph representation,
@@ -22,9 +23,4 @@
 
 @metric(**graph_metrics.graph_connectivity.metadata)
 def graph_connectivity(adata):
-    from scanpy.pp import neighbors
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.graph_connectivity(adata)
+    return graph_metrics.graph_connectivity(feature_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/iso_label_f1.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import feature_to_graph
 
 """
 We developed two isolated label scores to evaluate how well the data integration methods
@@ -27,9 +28,4 @@
 
 @metric(**graph_metrics.isolated_labels_f1.metadata)
 def isolated_labels_f1(adata):
-    from scanpy.pp import neighbors
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.isolated_labels_f1(adata)
+    return graph_metrics.isolated_labels_f1(feature_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/iso_label_sil.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 Isolated cell labels are defined as the labels present in the least number
@@ -15,7 +16,4 @@
 
 @metric(**embed_metrics.isolated_labels_sil.metadata)
 def isolated_labels_sil(adata):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.isolated_labels_sil(adata)
+    return embed_metrics.isolated_labels_sil(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/kBET.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 The kBET algorithm (v.0.99.6, release 4c9dafa) determines whether the label composition
@@ -27,7 +28,4 @@
 
 @metric(**embed_metrics.kBET.metadata)
 def kBET(adata):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.kBET(adata)
+    return embed_metrics.kBET(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/nmi.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_graph import metrics as graph_metrics
+from .utils import feature_to_graph
 
 """NMI compares the overlap of two clusterings.
 We used NMI to compare the cell-type labels with Louvain clusters computed on
@@ -15,9 +16,4 @@
 
 @metric(**graph_metrics.nmi.metadata)
 def nmi(adata):
-    from scanpy.pp import neighbors
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    neighbors(adata, use_rep="X_emb")
-    return graph_metrics.nmi(adata)
+    return graph_metrics.nmi(feature_to_graph(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/pcr.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 Principal component regression, derived from PCA, has previously been used to quantify
@@ -18,7 +19,4 @@
 
 @metric(**embed_metrics.pcr.metadata)
 def pcr(adata):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.pcr(adata)
+    return embed_metrics.pcr(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/sil_batch.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 We consider the absolute silhouette width, s(i), on
@@ -24,7 +25,4 @@
 
 @metric(**embed_metrics.silhouette_batch.metadata)
 def silhouette_batch(adata):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.silhouette_batch(adata)
+    return embed_metrics.silhouette_batch(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/silhouette.py

@@ -1,5 +1,6 @@
 from .....tools.decorators import metric
 from ...batch_integration_embed import metrics as embed_metrics
+from .utils import feature_to_embedding
 
 """
 For the bio-conservation score, the ASW was computed on cell identity labels and
@@ -12,7 +13,4 @@
 
 @metric(**embed_metrics.silhouette.metadata)
 def silhouette(adata):
-    from scanpy.tl import pca
-
-    adata.obsm["X_emb"] = pca(adata.X)
-    return embed_metrics.silhouette(adata)
+    return embed_metrics.silhouette(feature_to_embedding(adata))

openproblems/tasks/_batch_integration/batch_integration_feature/metrics/utils.py

@@ -0,0 +1,18 @@
+from ...batch_integration_embed.metrics.utils import embedding_to_graph
+
+
+def feature_to_embedding(adata):
+    import scanpy as sc
+
+    if adata.uns["is_baseline"] and "X_emb" in adata.obsm:
+        # precomputed; do nothing
+        return adata
+
+    adata.obsm["X_emb"] = sc.pp.pca(adata.X)
+    return adata
+
+
+def feature_to_graph(adata):
+    adata = feature_to_embedding(adata)
+    adata = embedding_to_graph(adata)
+    return adata

openproblems/tools/decorators.py

@@ -85,8 +85,9 @@ def decorator(func):
         @functools.wraps(func)
         def apply_method(adata: anndata.AnnData, *args, **kwargs):
             log.debug("Running {} method".format(func.__name__))
+            adata = func(adata, *args, **kwargs)
             adata.uns["is_baseline"] = is_baseline
-            return func(adata, *args, **kwargs)
+            return adata
 
         apply_method.metadata = dict(
             method_name=method_name,