preparing for version 0.0.4

daisybio · Jul 19, 2023 · e6b9d1a · e6b9d1a
1 parent 2bea301
commit e6b9d1a
Show file tree

Hide file tree

Showing 7 changed files with 541 additions and 123 deletions.
diff --git a/README.md b/README.md
@@ -116,6 +116,9 @@ data.get_results()
 
 data.get_results_binary()
 
+# get R2 values, coefficients, and coefficient p-values for all models/edges
+data.get_model_stats()
+
 ```
 
 The expected run time for the installation and running the demo dataset on a "normal" desktop computer is around 3~5 minutes.

diff --git a/dev_tests.ipynb b/dev_tests.ipynb
@@ -2,29 +2,20 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 94,
+   "execution_count": 1,
    "id": "d03d274e-6792-4bbf-93bf-b8c7259c1d7f",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "The autoreload extension is already loaded. To reload it, use:\n",
-      "  %reload_ext autoreload\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "%load_ext autoreload\n",
     "%autoreload 2\n",
     "import pandas as pd\n",
-    "from dysregnet.dysregnet import run"
+    "from src.dysregnet.dysregnet import run"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
+   "execution_count": 2,
    "id": "b6408bef-4768-42c3-88e3-21e250102240",
    "metadata": {},
    "outputs": [],
@@ -36,7 +27,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 48,
+   "execution_count": 3,
    "id": "0b876315-e883-450e-95a4-88b7caf76085",
    "metadata": {},
    "outputs": [],
@@ -53,7 +44,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 4,
    "id": "208046d3-5d5f-4ea4-aae7-c575094bc88e",
    "metadata": {},
    "outputs": [
@@ -266,7 +257,7 @@
        "[5 rows x 38 columns]"
       ]
      },
-     "execution_count": 49,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -277,7 +268,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 5,
    "id": "35febb56-1441-4049-a452-450c25afebb1",
    "metadata": {},
    "outputs": [
@@ -490,7 +481,7 @@
        "[5 rows x 22579 columns]"
       ]
      },
-     "execution_count": 50,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -501,7 +492,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 6,
    "id": "9dd72783-5b5d-488d-937e-f1d34535b29a",
    "metadata": {},
    "outputs": [
@@ -569,7 +560,7 @@
        "4       AHR       FOS"
       ]
      },
-     "execution_count": 51,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -580,7 +571,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 82,
+   "execution_count": 7,
    "id": "abda2742-f17d-440c-b5b2-e021d2c26f9c",
    "metadata": {},
    "outputs": [],
@@ -591,15 +582,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 95,
+   "execution_count": 8,
    "id": "62b921d3-27e0-4f54-a77b-d3b35d5eedfb",
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "14979it [00:42, 354.63it/s]\n"
+      "14979it [00:45, 332.12it/s]\n"
      ]
     }
    ],
@@ -614,7 +605,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 96,
+   "execution_count": 9,
    "id": "c3aede6a-7044-456e-affd-3804f54eb9d7",
    "metadata": {},
    "outputs": [
@@ -1044,7 +1035,7 @@
        "[515 rows x 14979 columns]"
       ]
      },
-     "execution_count": 96,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1055,7 +1046,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 97,
+   "execution_count": 10,
    "id": "e719292a-4d97-426a-b1aa-c64e1f0b1837",
    "metadata": {},
    "outputs": [
@@ -1272,7 +1263,7 @@
        "[14979 rows x 9 columns]"
       ]
      },
-     "execution_count": 97,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1283,7 +1274,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 98,
+   "execution_count": 11,
    "id": "3aacbc4e-8f13-48ef-849a-7404f45e9573",
    "metadata": {},
    "outputs": [
@@ -1713,7 +1704,7 @@
        "[515 rows x 14979 columns]"
       ]
      },
-     "execution_count": 98,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }

diff --git a/setup.py b/setup.py
@@ -11,7 +11,7 @@
 
 
 setup(name='dysregnet',
-      version='0.0.3',
+      version='0.0.4',
       description='DysRegNet',
       long_description=README,
       long_description_content_type="text/markdown",
@@ -27,7 +27,8 @@
         "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
         "Topic :: Scientific/Engineering :: Bio-Informatics",
       ],
-      packages=find_packages(),
+      package_dir = {'': 'src'},
+      packages=['dysregnet'],
       include_package_data=True,
       python_requires='>=3.7',
       install_requires=[

diff --git a/dysregnet/__init__.py → src/dysregnet/__init__.py b/dysregnet/__init__.py → src/dysregnet/__init__.py
diff --git a/dysregnet/dysregnet.py → src/dysregnet/dysregnet.py b/dysregnet/dysregnet.py → src/dysregnet/dysregnet.py
@@ -57,7 +57,7 @@ def __init__(self,
                             List of continuous covariates. They should match the name of their columns in meta Dataframe.
 
 
-                    zscoring: boolean, default: True 
+                    zscoring: boolean, default: False 
                          zscoring of expression data (if needed).
 
                     bonferroni_alpha: Float

diff --git a/dysregnet/functions.py → src/dysregnet/functions.py b/dysregnet/functions.py → src/dysregnet/functions.py
@@ -86,6 +86,9 @@ def dyregnet_model(data):
                         # fit the model
                         model = sm.OLS(y_train, x_train)
                         results = model.fit()
+
+                        model_stats[edge] = [results.rsquared] + list(results.params.values) + list(results.pvalues.values)
+
 
                         # get residuals of control
                         resid_control = results.predict(x_train) -  y_train
@@ -162,7 +165,6 @@ def dyregnet_model(data):
 
 
                         edges[edge] = np.round(zscore, 1)
-                        model_stats[edge] = [results.rsquared] + list(results.params.values) + list(results.pvalues.values)