LASSO_Coefficients.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1b95f400",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "01-Sep-2024 11:58:18 DEBUG utils - Directory ./ already exists. Skipping.\n",
      "01-Sep-2024 11:58:18 INFO GEOparse - File already exist: using local version.\n",
      "01-Sep-2024 11:58:18 INFO GEOparse - Parsing ./GSE11691_family.soft.gz: \n",
      "01-Sep-2024 11:58:18 DEBUG GEOparse - DATABASE: GeoMiame\n",
      "01-Sep-2024 11:58:18 DEBUG GEOparse - SERIES: GSE11691\n",
      "01-Sep-2024 11:58:18 DEBUG GEOparse - PLATFORM: GPL96\n",
      "01-Sep-2024 11:58:19 DEBUG GEOparse - SAMPLE: GSM296875\n",
      "01-Sep-2024 11:58:19 DEBUG GEOparse - SAMPLE: GSM296876\n",
      "01-Sep-2024 11:58:19 DEBUG GEOparse - SAMPLE: GSM296877\n",
      "01-Sep-2024 11:58:19 DEBUG GEOparse - SAMPLE: GSM296878\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296879\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296880\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296881\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296882\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296883\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296884\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296885\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296886\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296887\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296888\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296889\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296890\n",
      "01-Sep-2024 11:58:20 DEBUG GEOparse - SAMPLE: GSM296891\n",
      "01-Sep-2024 11:58:21 DEBUG GEOparse - SAMPLE: GSM296892\n"
     ]
    }
   ],
   "source": [
    "import GEOparse\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Step 1: Load and preprocess the GSE dataset\n",
    "gse = GEOparse.get_GEO(\"GSE11691\")\n",
    "expression_data = pd.DataFrame({\n",
    "    gsm: gse.gsms[gsm].table.set_index('ID_REF')['VALUE']\n",
    "    for gsm in gse.gsms.keys()\n",
    "}).T\n",
    "\n",
    "# Step 2: Standardize the data\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(expression_data)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9833daf3",
   "metadata": {},
   "source": [
    "### Apply LASSO to select relevant genes and Use a specific alpha value to ensure some features are selected"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "15209b6d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Silhouette Score: 0.30727010949657463\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Step 3: Apply LASSO to select relevant genes\n",
    "\n",
    "alpha_value = 0.01  # You can adjust this value\n",
    "lasso = Lasso(alpha=alpha_value)\n",
    "lasso.fit(X_scaled, np.random.choice([0, 1], size=expression_data.shape[0]))\n",
    "\n",
    "lasso_coef = lasso.coef_\n",
    "\n",
    "# Ensure some features are selected\n",
    "if np.sum(lasso_coef != 0) == 0:\n",
    "    print(\"No features were selected. Consider lowering the alpha value.\")\n",
    "else:\n",
    "    # Select genes with non-zero coefficients\n",
    "    selected_genes = expression_data.columns[lasso_coef != 0]\n",
    "    X_selected = X_scaled[:, lasso_coef != 0]\n",
    "\n",
    "    # Step 4: Clustering on selected features\n",
    "    kmeans = KMeans(n_clusters=3, random_state=42)\n",
    "    clusters = kmeans.fit_predict(X_selected)\n",
    "\n",
    "    # Step 5: Evaluate and visualize clusters\n",
    "    # Silhouette Score\n",
    "    silhouette_avg = silhouette_score(X_selected, clusters)\n",
    "    print(f\"Silhouette Score: {silhouette_avg}\")\n",
    "\n",
    "    # PCA for 2D visualization\n",
    "    pca = PCA(n_components=2)\n",
    "    X_pca = pca.fit_transform(X_selected)\n",
    "\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    sns.scatterplot(x=X_pca[:, 0], y=X_pca[:, 1], hue=clusters, palette='viridis')\n",
    "    plt.title('Clustering Results (PCA Visualization)')\n",
    "    plt.xlabel('PCA Component 1')\n",
    "    plt.ylabel('PCA Component 2')\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c1a3420d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}