DeepProfiler data stat analysis (heatmap and umap)

d33bs · Jan 19, 2023 · 0f2e400 · 0f2e400
1 parent e71bb1f
commit 0f2e400
Show file tree

Hide file tree

Showing 24 changed files with 4,952 additions and 115 deletions.
diff --git a/.gitignore b/.gitignore
@@ -36,3 +36,4 @@ example_code.ipynb
 
 # ignore deprecated notebooks
 5_analyze_data/notebooks/UMAP_analysis/deprecated_notebooks
+
diff --git a/...alyze_data/data/norm_fs_embeddings.csv.gz → ...ze_data/data/norm_fs_embeddings_cp.csv.gz b/...alyze_data/data/norm_fs_embeddings.csv.gz → ...ze_data/data/norm_fs_embeddings_cp.csv.gz
diff --git a/5_analyze_data/data/norm_fs_embeddings_dp_cyto.csv.gz b/5_analyze_data/data/norm_fs_embeddings_dp_cyto.csv.gz
diff --git a/5_analyze_data/data/norm_fs_embeddings_dp_nuc.csv.gz b/5_analyze_data/data/norm_fs_embeddings_dp_nuc.csv.gz
diff --git a/5_analyze_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc.png b/5_analyze_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc.png
diff --git a/...e_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc_dp_cyto.png b/...e_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc_dp_cyto.png
diff --git a/...ze_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc_dp_nuc.png b/...ze_data/notebooks/Heatmap_analysis/figures/correlation_clustermap_sc_dp_nuc.png
diff --git a/5_analyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc.png b/5_analyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc.png
diff --git a/...lyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc_dp_cyto.png b/...lyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc_dp_cyto.png
diff --git a/...alyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc_dp_nuc.png b/...alyze_data/notebooks/Heatmap_analysis/figures/correlation_heatmap_sc_dp_nuc.png
diff --git a/5_analyze_data/notebooks/Heatmap_analysis/nf1_heatmap.ipynb b/5_analyze_data/notebooks/Heatmap_analysis/nf1_heatmap.ipynb
diff --git a/5_analyze_data/notebooks/Heatmap_analysis/nf1_heatmap.py b/5_analyze_data/notebooks/Heatmap_analysis/nf1_heatmap.py
@@ -5,18 +5,22 @@
 
 # ## Import libraries
 
-# In[8]:
+# In[1]:
 
 
 import matplotlib.pyplot as plt
 import pathlib
 import pandas as pd
 import seaborn as sb
 
+import sys
+sys.path.append("../UMAP_analysis/")
+import UMAPutils as utils
+
 
 # ## Read in NF1 data `csv`
 
-# In[9]:
+# In[2]:
 
 
 norm_fs_data = pathlib.Path("../../../4_processing_features/data/nf1_sc_norm_fs_cellprofiler.csv.gz")
@@ -27,49 +31,18 @@
 data.head()
 
 
-# ## Helper function to split `csv` into metadata and features
-
-# In[10]:
-
-
-def split_data(pycytominer_output: pd.DataFrame):
-    """
-    split pycytominer output to return metadata dataframe
-
-    Parameters
-    ----------
-    pycytominer_output : pd.DataFrame
-        dataframe with pycytominer output
-
-    Returns
-    -------
-    pd.Dataframe, np.array
-        metadata dataframe, feature_data
-    """
-    # split metadata from features
-    metadata_cols = [
-        col_name
-        for col_name in pycytominer_output.columns.tolist()
-        if "Metadata" in col_name
-    ]
-    metadata_dataframe = pycytominer_output[metadata_cols]
-    feature_data = pycytominer_output[pycytominer_output.columns.difference(metadata_cols)]
-
-    return metadata_dataframe, feature_data
-
-
 # ## Split NF1 data `csv`
 
-# In[11]:
+# In[3]:
 
 
-metadata_dataframe, feature_data = split_data(data)
+metadata_dataframe, feature_data = utils.split_data(data)
 feature_data
 
 
 # ## Transpose the NF1 dataframe
 
-# In[12]:
+# In[4]:
 
 
 data_trans = feature_data.transpose()
@@ -78,7 +51,7 @@ def split_data(pycytominer_output: pd.DataFrame):
 
 # ## Create correlation heatmap
 
-# In[13]:
+# In[5]:
 
 
 data_trans_heatmap = sb.heatmap(data_trans.corr())
@@ -91,7 +64,7 @@ def split_data(pycytominer_output: pd.DataFrame):
 
 # ## Create clustermap with correlation heatmap
 
-# In[14]:
+# In[6]:
 
 
 sb.clustermap(data_trans.corr(), 

diff --git a/...analysis/get_2D_umap_embeddings_nf1.ipynb → ...lysis/CP_get_2D_umap_embeddings_nf1.ipynb b/...analysis/get_2D_umap_embeddings_nf1.ipynb → ...lysis/CP_get_2D_umap_embeddings_nf1.ipynb
@@ -5,7 +5,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# Get 2D UMAP embeddings for NF1 data"
+    "# Get 2D UMAP embeddings for NF1 CellProfiler data"
    ]
   },
   {
@@ -510,7 +510,7 @@
     }
    ],
    "source": [
-    "save_path = pathlib.Path('../../data/norm_fs_embeddings.csv.gz')\n",
+    "save_path = pathlib.Path('../../data/norm_fs_embeddings_cp.csv.gz')\n",
     "\n",
     "norm_fs_embeddings_data = utils.merge_metadata_embeddings(metadata_dataframe, embeddings, save_path)\n",
     "norm_fs_embeddings_data"
@@ -533,7 +533,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 06:04:18) \n[GCC 10.3.0]"
+   "version": "3.8.13"
   },
   "orig_nbformat": 4,
   "vscode": {