Merge branch 'dev' for 0.11.1

Dockerfile

@@ -21,7 +21,7 @@ RUN pip install --no-cache-dir --upgrade pip wheel && \
 # use version from argument (--build-arg version=0.11.0), or a default:
 ARG version="0.11.0"
 RUN pip install --no-cache-dir pyscenic==$version && \
-    pip install --no-cache-dir scanpy==1.7.0
+    pip install --no-cache-dir scanpy==1.7.2
 
 
 FROM python:3.7.9-slim AS build-image
@@ -42,3 +42,5 @@ COPY --from=compile-image /opt/venv /opt/venv
 # Make sure we use the virtualenv:
 ENV PATH="/opt/venv/bin:$PATH"
 
+EXPOSE 8787
+

README.rst

@@ -18,6 +18,14 @@ in no time. The latter is achieved via the dask_ framework for distributed compu
 News and releases
 -----------------
 
+0.11.1 | 2021-02-11
+^^^^^^^^^^^^^^^^^^^
+
+* Fix bug in motif url construction (#275)
+* Fix for export2loom with sparse dataframe (#278)
+* Fix sklearn t-SNE import (#285)
+* Updates to Docker image (expose port 8787 for Dask dashboard)
+
 0.11.0 | 2021-02-10
 ^^^^^^^^^^^^^^^^^^^
 

requirements.txt

@@ -23,3 +23,4 @@ scipy
 fsspec
 requests
 aiohttp
+scikit-learn>=0.22.2

requirements_docker.txt

@@ -1,5 +1,5 @@
 adjustText==0.7.3
-aiohttp==3.7.3
+aiohttp==3.7.4
 anndata==0.7.5
 ansiwrap==0.8.4
 appdirs==1.4.4
@@ -81,7 +81,7 @@ pathspec==0.8.1
 patsy==0.5.1
 pexpect==4.8.0
 pickleshare==0.7.5
-Pillow==8.1.0
+Pillow==8.1.1
 prompt-toolkit==3.0.15
 psutil==5.8.0
 ptyprocess==0.7.0
@@ -122,7 +122,7 @@ traitlets==5.0.5
 typed-ast==1.4.2
 typing-extensions==3.7.4.3
 umap-learn==0.5.1
-urllib3==1.26.3
+urllib3==1.26.4
 wcwidth==0.2.5
 webencodings==0.5.1
 yarl==1.6.3

src/pyscenic/aucell.py

@@ -60,10 +60,7 @@ def derive_auc_threshold(ex_mtx: pd.DataFrame) -> pd.DataFrame:
         that when using this value as the AUC threshold for 99% of the cells all ranked genes used for AUC calculation will
         have had a detected expression in the single-cell experiment.
     """
-    binary_mtx = ex_mtx.copy()
-    binary_mtx[binary_mtx != 0] = 1.0
-    n_genes = len(binary_mtx.columns)
-    return binary_mtx.sum(axis=1).quantile([.01, .05, .10, .50, 1])/n_genes
+    return pd.Series(np.count_nonzero(ex_mtx, axis=1)).quantile([.01, .05, .10, .50, 1])/ex_mtx.shape[1]
 
 
 enrichment = enrichment4cells

src/pyscenic/export.py

@@ -4,7 +4,7 @@
 import numpy as np
 import pandas as pd
 import loompy as lp
-from sklearn.manifold.t_sne import TSNE
+from sklearn.manifold import TSNE
 from .aucell import aucell
 from .genesig import Regulon
 from typing import List, Mapping, Union, Sequence, Optional
@@ -95,9 +95,7 @@ def export2loom(ex_mtx: pd.DataFrame, regulons: List[Regulon], out_fname: str,
         embeddings_Y = pd.merge(embeddings_Y, embedding['_Y'].to_frame().rename(columns={'_Y': str(embedding_id)}), left_index=True, right_index=True)
 
     # Calculate the number of genes per cell.
-    binary_mtx = ex_mtx.copy()
-    binary_mtx[binary_mtx != 0] = 1.0
-    ngenes = binary_mtx.sum(axis=1).astype(int)
+    ngenes = np.count_nonzero(ex_mtx, axis=1)
 
     # Encode genes in regulons as "binary" membership matrix.
     genes = np.array(ex_mtx.columns)
@@ -127,7 +125,7 @@ def create_structure_array(df):
     default_embedding.columns=['_X', '_Y']
     column_attrs = {
         "CellID": ex_mtx.index.values.astype('str'),
-        "nGene": ngenes.values,
+        "nGene": ngenes,
         "Embedding": create_structure_array(default_embedding),
         "RegulonsAUC": create_structure_array(auc_mtx),
         "Clusterings": create_structure_array(clusterings),

src/pyscenic/utils.py

@@ -323,7 +323,7 @@ def add_motif_url(df: pd.DataFrame, base_url: str):
     :param base_url:
     :return:
     """
-    df[("Enrichment", COLUMN_NAME_MOTIF_URL)] = list(map(partial(urljoin, base=base_url), df.index.get_level_values(COLUMN_NAME_MOTIF_ID)))
+    df[("Enrichment", COLUMN_NAME_MOTIF_URL)] = list(map(partial(urljoin, base_url), df.index.get_level_values(COLUMN_NAME_MOTIF_ID)))
     return df