jitter baseline patch (#838)

* jitter baseline patch

* celltype_random_embedding_w/o_jitter

* Split jitter and no jitter

* import

---------

Co-authored-by: Scott Gigante <84813314+scottgigante-immunai@users.noreply.github.com>
Former-commit-id: d4a00be

Author: danielStrobl

openproblems/tasks/_batch_integration/_common/methods/baseline.py

@@ -35,14 +35,15 @@ def _randomize_graph(adata, partition=None):
     return adata
 
 
-def _random_embedding(partition):
+def _random_embedding(partition, jitter=0.01):
     from sklearn.preprocessing import LabelEncoder
     from sklearn.preprocessing import OneHotEncoder
 
     embedding = OneHotEncoder().fit_transform(
         LabelEncoder().fit_transform(partition)[:, None]
     )
-    embedding = embedding + np.random.uniform(-0.01, 0.01, embedding.shape)
+    if jitter is not None:
+        embedding = embedding + np.random.uniform(-1 * jitter, jitter, embedding.shape)
     return embedding
 
 

openproblems/tasks/_batch_integration/batch_integration_embed/methods/__init__.py

@@ -34,6 +34,7 @@
 from ...batch_integration_graph.methods.scvi import scvi_full_unscaled
 from ...batch_integration_graph.methods.scvi import scvi_hvg_unscaled
 from .baseline import celltype_random_embedding
+from .baseline import celltype_random_embedding_jitter
 from .baseline import no_integration_batch
 from .scalex import scalex_full
 from .scalex import scalex_hvg

openproblems/tasks/_batch_integration/batch_integration_embed/methods/baseline.py

@@ -16,11 +16,20 @@
 )
 
 
+@_baseline_method(
+    method_name="Random Embedding by Celltype (with jitter)",
+)
+def celltype_random_embedding_jitter(adata, test=False):
+    adata.obsm["X_emb"] = _random_embedding(partition=adata.obs["labels"], jitter=0.01)
+    adata.uns["method_code_version"] = check_version("openproblems")
+    return adata
+
+
 @_baseline_method(
     method_name="Random Embedding by Celltype",
 )
 def celltype_random_embedding(adata, test=False):
-    adata.obsm["X_emb"] = _random_embedding(partition=adata.obs["labels"])
+    adata.obsm["X_emb"] = _random_embedding(partition=adata.obs["labels"], jitter=None)
     adata.uns["method_code_version"] = check_version("openproblems")
     return adata