examples.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "16cb838f-d40c-4158-bdce-e134e4da2152",
   "metadata": {},
   "outputs": [],
   "source": [
    "from artemis import artemis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "befb760d-b9df-4a44-82ef-e7fefe9b0296",
   "metadata": {},
   "source": [
    "# Basic example\n",
    "Pre-requisities\n",
    "```\n",
    "artemis\n",
    "numpy\n",
    "scipy\n",
    "pandas\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5b8527f1-6514-400e-aae5-7bf05ea82b52",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy import sparse\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "024f7699-6371-4b05-a07f-553734370016",
   "metadata": {},
   "source": [
    "### Generate data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5b83b145-0f98-4f5c-9c7a-c921f39f2d33",
   "metadata": {},
   "outputs": [],
   "source": [
    "# set the random seed\n",
    "np.random.seed(1)\n",
    "\n",
    "# Assign simulation parameters\n",
    "n = 10000 # Number of samples (n_samples)\n",
    "p = 100 # Number of features (genes)\n",
    "p_X = 0.05 # Probability of a variant in a given gene\n",
    "\n",
    "# Generate features\n",
    "X = 1*(np.random.rand(n,p)<=p_X)\n",
    "\n",
    "# Assign gene names (could be anything, here we opt for 'gene_0', 'gene_1' etc.)\n",
    "gene_names = [f'gene_{i}' for i in range(p)]\n",
    "\n",
    "# Generate phenotype labels from the first p0 features\n",
    "p0 = 5\n",
    "y = (X[:,:p0].sum(axis=1)>=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "806c8765-e2b1-4faa-abf5-840f57dcecc1",
   "metadata": {},
   "source": [
    "### Perform inference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4e2f5044-fe20-4c42-81ab-59c462487a42",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:00:02.644887\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Gene PIPs'}, xlabel='gene', ylabel='PIP value'>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Initialise the variational inference module\n",
    "vb = artemis.VB(\n",
    "    sparse.csr_matrix(X), # converts the numpy array to a sparse matrix\n",
    "    gene_names, # name of genes corresponding to each column of the sparse matrix\n",
    "    y # phenotype labels (should be a binary vector of length n_samples)\n",
    ")\n",
    "\n",
    "# Run variational inference (n_rep=number of iterations)\n",
    "vb.run(n_rep=3)\n",
    "\n",
    "# Plot the top 10 values\n",
    "vb.qpip_df[:10].plot(kind='bar', xlabel='gene', ylabel='PIP value', title='Gene PIPs')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9dcc653c-dd13-41a2-92d2-c2cab4c749a9",
   "metadata": {},
   "source": [
    "# XOR example\n",
    "We also explore what happens when the phenotype label depends on the first two features through an XOR operation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "42db6586-5499-422d-bdc1-4d015823e49f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate new data using an XOR process\n",
    "y = (X[:,:2].sum(axis=1)==1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2c7f19f6-2339-444d-8b10-54ac5c20964a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:00:02.624217\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Gene PIPs'}, xlabel='gene', ylabel='PIP value'>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Run ARTEMIS again\n",
    "vb = artemis.VB(\n",
    "    sparse.csr_matrix(X), \n",
    "    gene_names, \n",
    "    y\n",
    ")\n",
    "vb.run(n_rep=3)\n",
    "vb.qpip_df[:10].plot(kind='bar', xlabel='gene', ylabel='PIP value', title='Gene PIPs')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "051282c8-ff7f-4d5b-9fe3-7355ed44ed14",
   "metadata": {},
   "source": [
    "# Example pipeline\n",
    "- Here, we resample cases and controls according to a stratified sampling routine\n",
    "- For each sample of cases and controls we perform variational Bayes inference in the model\n",
    "- This generates a set of PIPs for each gene for a given sub-sample\n",
    "- PIPs are then averaged to obtain a single measure of Bayesian importance for each gene\n",
    "\n",
    "Processing of this part may take a few minutes (~5mins on this test)\n",
    "\n",
    "The **first five** genes are the only ones associated to the phenotype from the synthetic data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "206532d0-996a-4539-9405-37ce35f4a67e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gene_0</th>\n",
       "      <th>gene_1</th>\n",
       "      <th>gene_2</th>\n",
       "      <th>gene_3</th>\n",
       "      <th>gene_4</th>\n",
       "      <th>gene_5</th>\n",
       "      <th>gene_6</th>\n",
       "      <th>gene_7</th>\n",
       "      <th>gene_8</th>\n",
       "      <th>gene_9</th>\n",
       "      <th>...</th>\n",
       "      <th>gene_990</th>\n",
       "      <th>gene_991</th>\n",
       "      <th>gene_992</th>\n",
       "      <th>gene_993</th>\n",
       "      <th>gene_994</th>\n",
       "      <th>gene_995</th>\n",
       "      <th>gene_996</th>\n",
       "      <th>gene_997</th>\n",
       "      <th>gene_998</th>\n",
       "      <th>gene_999</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5000 rows × 1000 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      gene_0  gene_1  gene_2  gene_3  gene_4  gene_5  gene_6  gene_7  gene_8  \\\n",
       "0          0       0       1       0       0       0       0       0       0   \n",
       "1          0       0       0       0       0       0       0       0       0   \n",
       "2          0       0       0       0       0       0       0       0       0   \n",
       "3          0       0       0       0       0       0       0       0       0   \n",
       "4          0       0       0       0       0       0       0       0       0   \n",
       "...      ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "4995       0       0       0       0       0       0       0       0       0   \n",
       "4996       0       0       0       0       0       0       0       0       0   \n",
       "4997       0       0       0       0       0       0       0       0       0   \n",
       "4998       0       0       0       0       0       0       0       0       0   \n",
       "4999       0       0       0       0       0       0       0       0       0   \n",
       "\n",
       "      gene_9  ...  gene_990  gene_991  gene_992  gene_993  gene_994  gene_995  \\\n",
       "0          0  ...         0         0         0         0         0         0   \n",
       "1          0  ...         0         0         0         0         0         0   \n",
       "2          0  ...         0         0         0         0         0         0   \n",
       "3          0  ...         0         0         0         0         0         0   \n",
       "4          0  ...         0         0         0         0         0         0   \n",
       "...      ...  ...       ...       ...       ...       ...       ...       ...   \n",
       "4995       0  ...         0         0         0         0         0         0   \n",
       "4996       0  ...         0         0         0         0         0         0   \n",
       "4997       0  ...         0         0         0         0         0         0   \n",
       "4998       0  ...         0         0         0         0         0         0   \n",
       "4999       0  ...         0         0         0         0         0         0   \n",
       "\n",
       "      gene_996  gene_997  gene_998  gene_999  \n",
       "0            0         0         0         0  \n",
       "1            0         0         0         0  \n",
       "2            0         0         0         0  \n",
       "3            0         0         0         0  \n",
       "4            0         0         0         0  \n",
       "...        ...       ...       ...       ...  \n",
       "4995         0         0         0         0  \n",
       "4996         0         0         0         0  \n",
       "4997         0         0         0         0  \n",
       "4998         0         0         0         0  \n",
       "4999         0         0         0         0  \n",
       "\n",
       "[5000 rows x 1000 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cov_df = pd.read_csv('synthetic_data/cov.csv.gz', index_col=0)\n",
    "X_df = pd.read_csv('synthetic_data/X.csv.gz', index_col=0)\n",
    "y_df = pd.read_csv('synthetic_data/y.csv.gz', index_col=0).pheno\n",
    "X_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "79673e45-5bb5-4672-b9aa-3354c32eb9b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running cohort sample: 1 of 10\n",
      "0:00:26.977839\n",
      "Running cohort sample: 2 of 10\n",
      "0:00:26.732903\n",
      "Running cohort sample: 3 of 10\n",
      "0:00:26.880314\n",
      "Running cohort sample: 4 of 10\n",
      "0:00:27.026876\n",
      "Running cohort sample: 5 of 10\n",
      "0:00:26.978402\n",
      "Running cohort sample: 6 of 10\n",
      "0:00:26.941626\n",
      "Running cohort sample: 7 of 10\n",
      "0:00:26.777696\n",
      "Running cohort sample: 8 of 10\n",
      "0:00:26.928300\n",
      "Running cohort sample: 9 of 10\n",
      "0:00:26.967927\n",
      "Running cohort sample: 10 of 10\n",
      "0:00:26.766663\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pip</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>gene_1</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_3</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_4</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_2</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_0</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_213</th>\n",
       "      <td>0.029380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_110</th>\n",
       "      <td>0.011586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_284</th>\n",
       "      <td>0.015564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_794</th>\n",
       "      <td>0.018438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_174</th>\n",
       "      <td>0.022750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_150</th>\n",
       "      <td>0.033756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_976</th>\n",
       "      <td>0.010334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_880</th>\n",
       "      <td>0.006237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_848</th>\n",
       "      <td>0.019230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_114</th>\n",
       "      <td>0.010466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_813</th>\n",
       "      <td>0.013172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_75</th>\n",
       "      <td>0.004340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_665</th>\n",
       "      <td>0.002319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_106</th>\n",
       "      <td>0.001288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene_589</th>\n",
       "      <td>0.000755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               pip\n",
       "gene_1    1.000000\n",
       "gene_3    1.000000\n",
       "gene_4    1.000000\n",
       "gene_2    1.000000\n",
       "gene_0    1.000000\n",
       "gene_213  0.029380\n",
       "gene_110  0.011586\n",
       "gene_284  0.015564\n",
       "gene_794  0.018438\n",
       "gene_174  0.022750\n",
       "gene_150  0.033756\n",
       "gene_976  0.010334\n",
       "gene_880  0.006237\n",
       "gene_848  0.019230\n",
       "gene_114  0.010466\n",
       "gene_813  0.013172\n",
       "gene_75   0.004340\n",
       "gene_665  0.002319\n",
       "gene_106  0.001288\n",
       "gene_589  0.000755"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n_samplings = 10\n",
    "pip_dfs = {}\n",
    "\n",
    "for subsample in range(n_samplings):\n",
    "\n",
    "    print(f'Running cohort sample: {subsample+1} of {n_samplings}')\n",
    "    \n",
    "    age_sex = cov_df.copy()\n",
    "    cases = y_df[y_df==1].index\n",
    "    controls = y_df[y_df==0].index\n",
    "    \n",
    "    # Bin age covariate into 4 bins\n",
    "    age_sex['age'] = pd.qcut(age_sex['age'], 4)\n",
    "    \n",
    "    # stratify resample cases and controls by age and sex\n",
    "    stratified_df = artemis.stratified_sample( age_sex.loc[controls], age_sex.loc[cases], ['age','sex'])\n",
    "    stratified_df['case'] = stratified_df.index.map(lambda x: x in cases)*1\n",
    "    \n",
    "    # Run ARTEMIS with new cohort\n",
    "    X_df_in = X_df.loc[stratified_df.index]\n",
    "    y_df_in = stratified_df.case\n",
    "    \n",
    "    vb = artemis.VB(\n",
    "        sparse.csc_matrix(X_df_in.values), \n",
    "        X_df_in.columns, \n",
    "        y_df_in,\n",
    "        verbose=True\n",
    "    )\n",
    "    vb.run(n_rep=3)\n",
    "    pip_dfs[f'iter_{subsample}'] = vb.qpip_df\n",
    "\n",
    "pip_df = pd.concat(pip_dfs, axis=1).mean(axis=1).rename('pip').to_frame()\n",
    "\n",
    "pip_df[:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "336ed2c6-d2ec-478f-8c71-29358089d772",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Gene PIPs'}, xlabel='gene', ylabel='PIP value'>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pip_df[:20].plot(kind='bar', xlabel='gene', ylabel='PIP value', title='Gene PIPs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "145a22fd-a1a5-423c-aefe-ee12adbd09cd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "art-test",
   "language": "python",
   "name": "art_test"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}