diff --git a/HIVEBreastCancerPredictor.ipynb b/HIVEBreastCancerPredictor.ipynb
new file mode 100644
index 00000000..ae77bedb
--- /dev/null
+++ b/HIVEBreastCancerPredictor.ipynb
@@ -0,0 +1,1867 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "view-in-github",
+        "colab_type": "text"
+      },
+      "source": [
+        "<a href=\"https://colab.research.google.com/github/tukun29/PredictModBreastCancerPredictor/blob/main/HIVEBreastCancerPredictor.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "oVgomI-xvEDZ"
+      },
+      "source": [
+        "These are the imports needed to do the code"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 547
+        },
+        "id": "sOqXE1j0tfM0",
+        "outputId": "df25ff6d-0117-4ae0-d939-9eebf60c1516"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Collecting GEOparse\n",
+            "  Downloading GEOparse-2.0.4-py3-none-any.whl.metadata (6.5 kB)\n",
+            "Requirement already satisfied: numpy>=1.7 in /usr/local/lib/python3.12/dist-packages (from GEOparse) (2.0.2)\n",
+            "Requirement already satisfied: pandas>=0.17 in /usr/local/lib/python3.12/dist-packages (from GEOparse) (2.2.2)\n",
+            "Requirement already satisfied: requests>=2.21.0 in /usr/local/lib/python3.12/dist-packages (from GEOparse) (2.32.4)\n",
+            "Requirement already satisfied: tqdm>=4.31.1 in /usr/local/lib/python3.12/dist-packages (from GEOparse) (4.67.3)\n",
+            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.12/dist-packages (from pandas>=0.17->GEOparse) (2.9.0.post0)\n",
+            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.12/dist-packages (from pandas>=0.17->GEOparse) (2025.2)\n",
+            "Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.12/dist-packages (from pandas>=0.17->GEOparse) (2026.1)\n",
+            "Requirement already satisfied: charset_normalizer<4,>=2 in /usr/local/lib/python3.12/dist-packages (from requests>=2.21.0->GEOparse) (3.4.7)\n",
+            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.12/dist-packages (from requests>=2.21.0->GEOparse) (3.13)\n",
+            "Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.12/dist-packages (from requests>=2.21.0->GEOparse) (2.5.0)\n",
+            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.12/dist-packages (from requests>=2.21.0->GEOparse) (2026.4.22)\n",
+            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.12/dist-packages (from python-dateutil>=2.8.2->pandas>=0.17->GEOparse) (1.17.0)\n",
+            "Downloading GEOparse-2.0.4-py3-none-any.whl (29 kB)\n",
+            "Installing collected packages: GEOparse\n",
+            "Successfully installed GEOparse-2.0.4\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<IPython.core.display.HTML object>"
+            ],
+            "text/html": [
+              "\n",
+              "     <input type=\"file\" id=\"files-69fbd33b-0453-4f98-9041-841c0de63dfd\" name=\"files[]\" multiple disabled\n",
+              "        style=\"border:none\" />\n",
+              "     <output id=\"result-69fbd33b-0453-4f98-9041-841c0de63dfd\">\n",
+              "      Upload widget is only available when the cell has been executed in the\n",
+              "      current browser session. Please rerun this cell to enable.\n",
+              "      </output>\n",
+              "      <script>// Copyright 2017 Google LLC\n",
+              "//\n",
+              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
+              "// you may not use this file except in compliance with the License.\n",
+              "// You may obtain a copy of the License at\n",
+              "//\n",
+              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
+              "//\n",
+              "// Unless required by applicable law or agreed to in writing, software\n",
+              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
+              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
+              "// See the License for the specific language governing permissions and\n",
+              "// limitations under the License.\n",
+              "\n",
+              "/**\n",
+              " * @fileoverview Helpers for google.colab Python module.\n",
+              " */\n",
+              "(function(scope) {\n",
+              "function span(text, styleAttributes = {}) {\n",
+              "  const element = document.createElement('span');\n",
+              "  element.textContent = text;\n",
+              "  for (const key of Object.keys(styleAttributes)) {\n",
+              "    element.style[key] = styleAttributes[key];\n",
+              "  }\n",
+              "  return element;\n",
+              "}\n",
+              "\n",
+              "// Max number of bytes which will be uploaded at a time.\n",
+              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
+              "\n",
+              "function _uploadFiles(inputId, outputId) {\n",
+              "  const steps = uploadFilesStep(inputId, outputId);\n",
+              "  const outputElement = document.getElementById(outputId);\n",
+              "  // Cache steps on the outputElement to make it available for the next call\n",
+              "  // to uploadFilesContinue from Python.\n",
+              "  outputElement.steps = steps;\n",
+              "\n",
+              "  return _uploadFilesContinue(outputId);\n",
+              "}\n",
+              "\n",
+              "// This is roughly an async generator (not supported in the browser yet),\n",
+              "// where there are multiple asynchronous steps and the Python side is going\n",
+              "// to poll for completion of each step.\n",
+              "// This uses a Promise to block the python side on completion of each step,\n",
+              "// then passes the result of the previous step as the input to the next step.\n",
+              "function _uploadFilesContinue(outputId) {\n",
+              "  const outputElement = document.getElementById(outputId);\n",
+              "  const steps = outputElement.steps;\n",
+              "\n",
+              "  const next = steps.next(outputElement.lastPromiseValue);\n",
+              "  return Promise.resolve(next.value.promise).then((value) => {\n",
+              "    // Cache the last promise value to make it available to the next\n",
+              "    // step of the generator.\n",
+              "    outputElement.lastPromiseValue = value;\n",
+              "    return next.value.response;\n",
+              "  });\n",
+              "}\n",
+              "\n",
+              "/**\n",
+              " * Generator function which is called between each async step of the upload\n",
+              " * process.\n",
+              " * @param {string} inputId Element ID of the input file picker element.\n",
+              " * @param {string} outputId Element ID of the output display.\n",
+              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
+              " */\n",
+              "function* uploadFilesStep(inputId, outputId) {\n",
+              "  const inputElement = document.getElementById(inputId);\n",
+              "  inputElement.disabled = false;\n",
+              "\n",
+              "  const outputElement = document.getElementById(outputId);\n",
+              "  outputElement.innerHTML = '';\n",
+              "\n",
+              "  const pickedPromise = new Promise((resolve) => {\n",
+              "    inputElement.addEventListener('change', (e) => {\n",
+              "      resolve(e.target.files);\n",
+              "    });\n",
+              "  });\n",
+              "\n",
+              "  const cancel = document.createElement('button');\n",
+              "  inputElement.parentElement.appendChild(cancel);\n",
+              "  cancel.textContent = 'Cancel upload';\n",
+              "  const cancelPromise = new Promise((resolve) => {\n",
+              "    cancel.onclick = () => {\n",
+              "      resolve(null);\n",
+              "    };\n",
+              "  });\n",
+              "\n",
+              "  // Wait for the user to pick the files.\n",
+              "  const files = yield {\n",
+              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
+              "    response: {\n",
+              "      action: 'starting',\n",
+              "    }\n",
+              "  };\n",
+              "\n",
+              "  cancel.remove();\n",
+              "\n",
+              "  // Disable the input element since further picks are not allowed.\n",
+              "  inputElement.disabled = true;\n",
+              "\n",
+              "  if (!files) {\n",
+              "    return {\n",
+              "      response: {\n",
+              "        action: 'complete',\n",
+              "      }\n",
+              "    };\n",
+              "  }\n",
+              "\n",
+              "  for (const file of files) {\n",
+              "    const li = document.createElement('li');\n",
+              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
+              "    li.append(span(\n",
+              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
+              "        `last modified: ${\n",
+              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
+              "                                    'n/a'} - `));\n",
+              "    const percent = span('0% done');\n",
+              "    li.appendChild(percent);\n",
+              "\n",
+              "    outputElement.appendChild(li);\n",
+              "\n",
+              "    const fileDataPromise = new Promise((resolve) => {\n",
+              "      const reader = new FileReader();\n",
+              "      reader.onload = (e) => {\n",
+              "        resolve(e.target.result);\n",
+              "      };\n",
+              "      reader.readAsArrayBuffer(file);\n",
+              "    });\n",
+              "    // Wait for the data to be ready.\n",
+              "    let fileData = yield {\n",
+              "      promise: fileDataPromise,\n",
+              "      response: {\n",
+              "        action: 'continue',\n",
+              "      }\n",
+              "    };\n",
+              "\n",
+              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
+              "    let position = 0;\n",
+              "    do {\n",
+              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
+              "      const chunk = new Uint8Array(fileData, position, length);\n",
+              "      position += length;\n",
+              "\n",
+              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
+              "      yield {\n",
+              "        response: {\n",
+              "          action: 'append',\n",
+              "          file: file.name,\n",
+              "          data: base64,\n",
+              "        },\n",
+              "      };\n",
+              "\n",
+              "      let percentDone = fileData.byteLength === 0 ?\n",
+              "          100 :\n",
+              "          Math.round((position / fileData.byteLength) * 100);\n",
+              "      percent.textContent = `${percentDone}% done`;\n",
+              "\n",
+              "    } while (position < fileData.byteLength);\n",
+              "  }\n",
+              "\n",
+              "  // All done.\n",
+              "  yield {\n",
+              "    response: {\n",
+              "      action: 'complete',\n",
+              "    }\n",
+              "  };\n",
+              "}\n",
+              "\n",
+              "scope.google = scope.google || {};\n",
+              "scope.google.colab = scope.google.colab || {};\n",
+              "scope.google.colab._files = {\n",
+              "  _uploadFiles,\n",
+              "  _uploadFilesContinue,\n",
+              "};\n",
+              "})(self);\n",
+              "</script> "
+            ]
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Saving GSE25055_family.soft.gz to GSE25055_family.soft.gz\n"
+          ]
+        },
+        {
+          "output_type": "error",
+          "ename": "NameError",
+          "evalue": "name 'sklearn' is not defined",
+          "traceback": [
+            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
+            "\u001b[0;32m/tmp/ipykernel_19112/1601073121.py\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mgoogle\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolab\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mfiles\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0muploaded\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfiles\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 16\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msklearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mversion\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+            "\u001b[0;31mNameError\u001b[0m: name 'sklearn' is not defined"
+          ]
+        }
+      ],
+      "source": [
+        "### Imports / Setup\n",
+        "\n",
+        "import numpy as np\n",
+        "import pandas as pd\n",
+        "\n",
+        "!pip install GEOparse\n",
+        "import GEOparse\n",
+        "\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "from sklearn.preprocessing import StandardScaler\n",
+        "from sklearn.linear_model import LogisticRegression\n",
+        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score\n",
+        "\n",
+        "from google.colab import files\n",
+        "uploaded = files.upload()\n",
+        "print(sklearn.version)\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "8G57eOASvBxc"
+      },
+      "source": [
+        "Now that the soft file has been uploaded, it gets saved under pmid_soft as the variable"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": true,
+        "id": "_CGN8qVuu_5b",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "580754f8-9c97-46bb-c22a-bbf05ce8487f"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "15-Apr-2026 14:20:36 INFO GEOparse - Parsing GSE25055_family.soft.gz: \n",
+            "INFO:GEOparse:Parsing GSE25055_family.soft.gz: \n",
+            "15-Apr-2026 14:20:36 DEBUG GEOparse - DATABASE: GeoMiame\n",
+            "DEBUG:GEOparse:DATABASE: GeoMiame\n",
+            "15-Apr-2026 14:20:36 DEBUG GEOparse - SERIES: GSE25055\n",
+            "DEBUG:GEOparse:SERIES: GSE25055\n",
+            "15-Apr-2026 14:20:36 DEBUG GEOparse - PLATFORM: GPL96\n",
+            "DEBUG:GEOparse:PLATFORM: GPL96\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615096\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615096\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615097\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615097\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615098\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615098\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615099\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615099\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615100\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615100\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615101\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615101\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615102\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615102\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615103\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615103\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615104\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615104\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615105\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615105\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615106\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615106\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615107\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615107\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615108\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615108\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615109\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615109\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615110\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615110\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615111\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615111\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615112\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615112\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615113\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615113\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615114\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615114\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615115\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615115\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615116\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615116\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615117\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615117\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615118\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615118\n",
+            "15-Apr-2026 14:20:37 DEBUG GEOparse - SAMPLE: GSM615119\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615119\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615120\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615120\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615121\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615121\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615122\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615122\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615123\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615123\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615124\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615124\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615125\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615125\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615126\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615126\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615127\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615127\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615128\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615128\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615129\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615129\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615130\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615130\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615131\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615131\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615132\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615132\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615133\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615133\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615134\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615134\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615135\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615135\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615136\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615136\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615137\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615137\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615138\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615138\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615139\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615139\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615140\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615140\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615141\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615141\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615142\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615142\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615143\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615143\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615144\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615144\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615145\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615145\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615146\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615146\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615147\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615147\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615148\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615148\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615149\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615149\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615150\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615150\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615151\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615151\n",
+            "15-Apr-2026 14:20:38 DEBUG GEOparse - SAMPLE: GSM615152\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615152\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615153\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615153\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615154\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615154\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615155\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615155\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615156\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615156\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615157\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615157\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615158\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615158\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615159\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615159\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615160\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615160\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615161\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615161\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615162\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615162\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615163\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615163\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615164\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615164\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615165\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615165\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615166\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615166\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615167\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615167\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615168\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615168\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615169\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615169\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615170\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615170\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615171\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615171\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615172\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615172\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615173\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615173\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615174\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615174\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615175\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615175\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615176\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615176\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615177\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615177\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615178\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615178\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615179\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615179\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615180\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615180\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615181\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615181\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615182\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615182\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615183\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615183\n",
+            "15-Apr-2026 14:20:39 DEBUG GEOparse - SAMPLE: GSM615184\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615184\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615185\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615185\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615186\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615186\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615187\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615187\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615188\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615188\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615189\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615189\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615190\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615190\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615191\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615191\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615192\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615192\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615193\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615193\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615194\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615194\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615195\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615195\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615196\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615196\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615197\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615197\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615198\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615198\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615199\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615199\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615200\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615200\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615201\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615201\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615202\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615202\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615203\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615203\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615204\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615204\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615205\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615205\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615206\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615206\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615207\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615207\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615208\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615208\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615209\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615209\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615210\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615210\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615211\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615211\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615212\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615212\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615213\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615213\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615214\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615214\n",
+            "15-Apr-2026 14:20:40 DEBUG GEOparse - SAMPLE: GSM615215\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615215\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615216\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615216\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615217\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615217\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615218\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615218\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615219\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615219\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615220\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615220\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615221\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615221\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615222\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615222\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615223\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615223\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615224\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615224\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615225\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615225\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615226\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615226\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615227\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615227\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615228\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615228\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615229\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615229\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615230\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615230\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615231\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615231\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615232\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615232\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615233\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615233\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615234\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615234\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615235\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615235\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615236\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615236\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615237\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615237\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615238\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615238\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615239\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615239\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615240\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615240\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615241\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615241\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615242\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615242\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615243\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615243\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615244\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615244\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615245\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615245\n",
+            "15-Apr-2026 14:20:41 DEBUG GEOparse - SAMPLE: GSM615246\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615246\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615247\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615247\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615248\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615248\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615249\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615249\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615250\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615250\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615251\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615251\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615252\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615252\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615253\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615253\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615254\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615254\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615255\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615255\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615256\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615256\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615257\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615257\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615258\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615258\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615259\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615259\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615260\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615260\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615261\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615261\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615262\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615262\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615263\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615263\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615264\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615264\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615265\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615265\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615266\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615266\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615267\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615267\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615268\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615268\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615269\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615269\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615270\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615270\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615271\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615271\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615272\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615272\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615273\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615273\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615274\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615274\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615275\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615275\n",
+            "15-Apr-2026 14:20:42 DEBUG GEOparse - SAMPLE: GSM615276\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615276\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615277\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615277\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615278\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615278\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615279\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615279\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615280\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615280\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615281\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615281\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615282\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615282\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615283\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615283\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615284\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615284\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615285\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615285\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615286\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615286\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615287\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615287\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615288\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615288\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615289\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615289\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615290\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615290\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615291\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615291\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615292\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615292\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615293\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615293\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615294\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615294\n",
+            "15-Apr-2026 14:20:43 DEBUG GEOparse - SAMPLE: GSM615295\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615295\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615296\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615296\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615297\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615297\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615298\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615298\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615299\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615299\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615300\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615300\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615301\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615301\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615302\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615302\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615303\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615303\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615304\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615304\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615305\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615305\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615306\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615306\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615307\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615307\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615308\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615308\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615309\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615309\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615310\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615310\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615311\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615311\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615312\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615312\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615313\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615313\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615314\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615314\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615315\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615315\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615316\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615316\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615317\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615317\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615318\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615318\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615319\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615319\n",
+            "15-Apr-2026 14:20:44 DEBUG GEOparse - SAMPLE: GSM615320\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615320\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615321\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615321\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615322\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615322\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615323\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615323\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615324\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615324\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615325\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615325\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615326\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615326\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615327\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615327\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615328\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615328\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615329\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615329\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615330\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615330\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615331\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615331\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615332\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615332\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615333\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615333\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615334\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615334\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615335\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615335\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615336\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615336\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615337\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615337\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615338\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615338\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615339\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615339\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615340\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615340\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615341\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615341\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615342\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615342\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615343\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615343\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615344\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615344\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615345\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615345\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615346\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615346\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615347\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615347\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615348\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615348\n",
+            "15-Apr-2026 14:20:45 DEBUG GEOparse - SAMPLE: GSM615349\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615349\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615350\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615350\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615351\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615351\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615352\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615352\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615353\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615353\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615354\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615354\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615355\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615355\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615356\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615356\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615357\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615357\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615358\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615358\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615359\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615359\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615360\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615360\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615361\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615361\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615362\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615362\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615363\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615363\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615364\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615364\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615365\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615365\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615366\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615366\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615367\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615367\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615368\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615368\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615369\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615369\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615370\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615370\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615371\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615371\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615372\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615372\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615373\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615373\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615374\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615374\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615375\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615375\n",
+            "15-Apr-2026 14:20:46 DEBUG GEOparse - SAMPLE: GSM615376\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615376\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615377\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615377\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615378\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615378\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615379\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615379\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615380\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615380\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615381\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615381\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615382\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615382\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615383\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615383\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615384\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615384\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615385\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615385\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615386\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615386\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615387\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615387\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615388\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615388\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615389\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615389\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615390\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615390\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615391\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615391\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615392\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615392\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615393\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615393\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615394\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615394\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615395\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615395\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615396\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615396\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615397\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615397\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615398\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615398\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615399\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615399\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615400\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615400\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615401\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615401\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615402\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615402\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615403\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615403\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615404\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615404\n",
+            "15-Apr-2026 14:20:47 DEBUG GEOparse - SAMPLE: GSM615405\n",
+            "DEBUG:GEOparse:SAMPLE: GSM615405\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "loaded\n",
+            "<class 'GEOparse.GEOTypes.GSE'>\n",
+            "310\n",
+            "dict_keys(['GPL96'])\n"
+          ]
+        }
+      ],
+      "source": [
+        "pmid_soft = GEOparse.get_GEO(filepath=\"GSE25055_family.soft.gz\")\n",
+        "print(\"loaded\")\n",
+        "\n",
+        "#additioal checks\n",
+        "print(type(pmid_soft))\n",
+        "print(len(pmid_soft.gsms))\n",
+        "print(pmid_soft.gpls.keys())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "EyddC3b-3j60"
+      },
+      "source": [
+        "Extracting the Meta-data: \"data about data\" gives us the type of data points sitting in the SOFT file\n",
+        "The code prints the columns that exist in the metadata some of the data is useful some of it is not. Seeing a list helps us identify that\n",
+        "The SHAPE of the metadata tells us its dimensions. In this case: 310, 25 indicates there are 310 rows and 35 colums. 35 characterisics for 310 total samples. We should also see the DRFS binary property show up becyase that property actuall gives us the resulting 1 or 0 that tells the code whether the paitient actually made it through treatment or relapsed."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "JTRSN-la3jUi",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "eb0c2253-59da-4fab-c2e3-15fad39503e8"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Metadata shape: (310, 61)\n",
+            "\n",
+            "Metadata columns:\n",
+            "['sample_id', 'title', 'geo_accession', 'status', 'submission_date', 'last_update_date', 'type', 'channel_count', 'source_name_ch1', 'organism_ch1', 'taxid_ch1', 'characteristics_ch1', 'source', 'age_years', 'er_status_ihc', 'pr_status_ihc', 'her2_status', 'er_status_ihc_esr1_for_indeterminate', 'clinical_t_stage', 'clinical_nodal_status', 'clinical_ajcc_stage', 'grade', 'pathologic_response_pcr_rd', 'pathologic_response_rcb_class', 'drfs_1_event_0_censored', 'drfs_even_time_years', 'esr1_status', 'erbb2_status', 'set_class', 'chemosensitivity_prediction', 'ggi_class', 'pam50_class', 'dlda30_prediction', 'rcb_0_i_prediction', 'tissue', 'treatment_protocol_ch1', 'molecule_ch1', 'extract_protocol_ch1', 'label_ch1', 'label_protocol_ch1', 'hyb_protocol', 'scan_protocol', 'data_processing', 'probe_intensities_were_quantified_with_microarray_suite_version_5.0_(mas_5.0)_using_affymetrix_default_analysis_settings.__cel_files_were_normalized_using_global_scaling_with_a_trimmed_mean_target_intensity_of_each_array_arbitrarily_set_to_600_using_the_mas5_algorithm_from_the_simpleaffy_package_(http', 'platform_id', 'contact_name', 'contact_email', 'contact_phone', 'contact_institute', 'contact_address', 'contact_city', 'contact_state', 'contact_zip/postal_code', 'contact_country', 'contact_web_link', 'supplementary_file', 'ftp', 'relation', 'reanalyzed_by', 'series_id', 'data_row_count']\n"
+          ]
+        }
+      ],
+      "source": [
+        "# EXTRACT METADATA\n",
+        "metadata_list = []\n",
+        "\n",
+        "for sample_name, sample in pmid_soft.gsms.items():\n",
+        "    row = {\"sample_id\": sample_name}\n",
+        "\n",
+        "    for key, value in sample.metadata.items():\n",
+        "        # Keep full metadata as readable strings\n",
+        "        if isinstance(value, list):\n",
+        "            row[key] = \" | \".join(map(str, value))\n",
+        "\n",
+        "            # ALSO split things like \"drfs_1_event_0_censored: 1\"\n",
+        "            for item in value:\n",
+        "                if \":\" in item:\n",
+        "                    k, v = item.split(\":\", 1)\n",
+        "                    clean_key = k.strip().lower().replace(\" \", \"_\")\n",
+        "                    row[clean_key] = v.strip()\n",
+        "        else:\n",
+        "            row[key] = value\n",
+        "\n",
+        "    metadata_list.append(row)\n",
+        "\n",
+        "metadata_df = pd.DataFrame(metadata_list)\n",
+        "\n",
+        "print(\"Metadata shape:\", metadata_df.shape)\n",
+        "print(\"\\nMetadata columns:\")\n",
+        "print(metadata_df.columns.tolist())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "9_RuqUfj5D1q"
+      },
+      "source": [
+        "Now we are going to identify the numeric (useful) data from the SOFT.file. Instead of looking at the metadate we go to the \"table\" layer/component of the file which the code will search through based on the VALUE parameter to pull the numeric data. The VALUE parmeter is the standardized parameter for GEO\n",
+        "files."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "RXg8Pzn25Ffe",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "5a3a9658-36f4-4dff-ee10-d76af25bd725"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Extracting expression matrix...\n",
+            "Original shape (genes x samples): (22283, 310)\n",
+            "Transposed shape (samples x probes): (310, 22283)\n",
+            "ID_REF     1007_s_at   1053_at    117_at     121_at  1255_g_at    1294_at  \\\n",
+            "name                                                                        \n",
+            "GSM615096  12.359267  8.021741  8.294244  10.840949   7.874062   8.341292   \n",
+            "GSM615097  12.613243  7.842947  5.474758  11.715159   7.347955   9.419005   \n",
+            "GSM615098  11.551397  8.776267  8.197121  10.715077   6.523880   9.297479   \n",
+            "GSM615099  11.828700  9.523877  7.907263  10.750475   6.478643   8.698516   \n",
+            "GSM615100  11.983987  6.836237  8.625010  10.827725   7.336754   9.599059   \n",
+            "...              ...       ...       ...        ...        ...        ...   \n",
+            "GSM615401  12.135865  8.633192  9.783710   9.830411   7.357560   8.217632   \n",
+            "GSM615402  11.982492  9.451202  8.210670   9.354246   7.178217   9.082518   \n",
+            "GSM615403  11.721887  8.955527  8.980073  10.727569   8.242348   9.691482   \n",
+            "GSM615404  11.066133  9.489278  7.945081   9.552854   7.353828  10.022531   \n",
+            "GSM615405  10.703124  8.813052  9.025983   9.693816   5.531112   9.214619   \n",
+            "\n",
+            "ID_REF      1316_at   1320_at  1405_i_at   1431_at  ...  AFFX-r2-Ec-bioD-3_at  \\\n",
+            "name                                                ...                         \n",
+            "GSM615096  8.256335  6.485966   6.495728  7.063123  ...             15.640044   \n",
+            "GSM615097  7.983681  7.330512   7.332244  7.033859  ...             16.719929   \n",
+            "GSM615098  7.820261  6.423794  10.066291  6.661108  ...             14.850736   \n",
+            "GSM615099  7.526617  7.276040  10.399704  6.642930  ...             14.414238   \n",
+            "GSM615100  8.312127  6.832232   7.090151  7.081613  ...             15.066009   \n",
+            "...             ...       ...        ...       ...  ...                   ...   \n",
+            "GSM615401  8.313859  5.304950   8.262117  7.488756  ...             14.103782   \n",
+            "GSM615402  7.144019  6.389862   8.596580  6.407605  ...             14.308051   \n",
+            "GSM615403  7.817483  5.722656  11.104730  8.290744  ...             15.862593   \n",
+            "GSM615404  7.460670  4.903692   9.808574  7.383959  ...             14.668847   \n",
+            "GSM615405  7.460981  4.446914   9.367801  6.786354  ...             14.701467   \n",
+            "\n",
+            "ID_REF     AFFX-r2-Ec-bioD-5_at  AFFX-r2-Hs18SrRNA-3_s_at  \\\n",
+            "name                                                        \n",
+            "GSM615096             15.441187                 11.190421   \n",
+            "GSM615097             16.383235                 10.750292   \n",
+            "GSM615098             14.637426                 11.580917   \n",
+            "GSM615099             14.218011                 10.275100   \n",
+            "GSM615100             14.816849                 10.840278   \n",
+            "...                         ...                       ...   \n",
+            "GSM615401             13.826508                 11.742811   \n",
+            "GSM615402             13.858847                 10.993654   \n",
+            "GSM615403             15.411223                  9.693377   \n",
+            "GSM615404             14.260912                 15.270607   \n",
+            "GSM615405             14.235730                 11.370290   \n",
+            "\n",
+            "ID_REF     AFFX-r2-Hs18SrRNA-5_at  AFFX-r2-Hs18SrRNA-M_x_at  \\\n",
+            "name                                                          \n",
+            "GSM615096               11.955084                 12.139089   \n",
+            "GSM615097               11.610319                 12.080350   \n",
+            "GSM615098               13.075420                 12.347660   \n",
+            "GSM615099               11.467574                 10.788742   \n",
+            "GSM615100               12.989398                 11.851791   \n",
+            "...                           ...                       ...   \n",
+            "GSM615401               14.297823                 10.871473   \n",
+            "GSM615402               12.314814                  9.767024   \n",
+            "GSM615403               12.189645                  8.891483   \n",
+            "GSM615404               15.506028                 14.867092   \n",
+            "GSM615405               12.392458                 10.071009   \n",
+            "\n",
+            "ID_REF     AFFX-r2-Hs28SrRNA-3_at  AFFX-r2-Hs28SrRNA-5_at  \\\n",
+            "name                                                        \n",
+            "GSM615096               14.074549                9.047564   \n",
+            "GSM615097               13.681936                8.765575   \n",
+            "GSM615098               13.676062                9.450242   \n",
+            "GSM615099               13.395810                8.466331   \n",
+            "GSM615100               13.943810                8.721290   \n",
+            "...                           ...                     ...   \n",
+            "GSM615401               11.313766                9.321655   \n",
+            "GSM615402               10.714735                8.832938   \n",
+            "GSM615403               10.494081                6.370288   \n",
+            "GSM615404               12.315510                9.259754   \n",
+            "GSM615405               11.244264                9.335462   \n",
+            "\n",
+            "ID_REF     AFFX-r2-Hs28SrRNA-M_at  AFFX-r2-P1-cre-3_at  AFFX-r2-P1-cre-5_at  \n",
+            "name                                                                         \n",
+            "GSM615096               11.274337            17.356685            17.004035  \n",
+            "GSM615097               11.287301            18.404677            18.143576  \n",
+            "GSM615098               11.308490            16.304500            15.988321  \n",
+            "GSM615099               10.208604            16.190139            15.807992  \n",
+            "GSM615100               11.110014            16.767231            16.485521  \n",
+            "...                           ...                  ...                  ...  \n",
+            "GSM615401                9.334666            15.501419            15.074929  \n",
+            "GSM615402                9.378723            15.488476            15.190513  \n",
+            "GSM615403                9.124964            16.847545            16.685688  \n",
+            "GSM615404               10.416107            15.839781            15.501607  \n",
+            "GSM615405                8.978068            15.661208            15.337064  \n",
+            "\n",
+            "[310 rows x 22283 columns]\n"
+          ]
+        }
+      ],
+      "source": [
+        "print(\"\\nExtracting expression matrix...\")\n",
+        "expression_df = pmid_soft.pivot_samples(\"VALUE\")\n",
+        "\n",
+        "print(\"Original shape (genes x samples):\", expression_df.shape)\n",
+        "\n",
+        "# GEOparse usually gives rows = probes, columns = samples\n",
+        "# We want rows = samples, columns = probes\n",
+        "expression_df = expression_df.T\n",
+        "\n",
+        "print(\"Transposed shape (samples x probes):\", expression_df.shape)\n",
+        "print(expression_df)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "N9H9ij73_Hw5"
+      },
+      "source": [
+        "Mapping the data back and renaimig things to be a bit more clearer and biologically meaningful to the average person (me)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "eRJhrC5b_Ira",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "591036cf-9029-4efa-83fa-48da8076004b"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Annotation columns:\n",
+            "['ID', 'GB_ACC', 'SPOT_ID', 'Species Scientific Name', 'Annotation Date', 'Sequence Type', 'Sequence Source', 'Target Description', 'Representative Public ID', 'Gene Title', 'Gene Symbol', 'ENTREZ_GENE_ID', 'RefSeq Transcript ID', 'Gene Ontology Biological Process', 'Gene Ontology Cellular Component', 'Gene Ontology Molecular Function']\n"
+          ]
+        }
+      ],
+      "source": [
+        "# MAP PROBE IDs TO GENE SYMBOLS\n",
+        "\n",
+        "gpl = pmid_soft.gpls[\"GPL96\"]\n",
+        "annotation = gpl.table\n",
+        "\n",
+        "print(\"Annotation columns:\")\n",
+        "print(annotation.columns.tolist())\n",
+        "\n",
+        "# Some rows may have missing gene symbols\n",
+        "annotation = annotation[[\"ID\", \"Gene Symbol\"]].copy()\n",
+        "annotation[\"Gene Symbol\"] = annotation[\"Gene Symbol\"].fillna(\"\").astype(str).str.strip()\n",
+        "\n",
+        "id_to_gene = dict(zip(annotation[\"ID\"], annotation[\"Gene Symbol\"]))\n",
+        "\n",
+        "# Replace probe IDs with gene symbols when possible\n",
+        "new_columns = []\n",
+        "for col in expression_df.columns:\n",
+        "    gene = id_to_gene.get(col, \"\")\n",
+        "    if gene == \"\" or gene == \"---\":\n",
+        "        new_columns.append(col)   # keep probe ID if no gene symbol\n",
+        "    else:\n",
+        "        new_columns.append(gene)\n",
+        "\n",
+        "expression_df.columns = new_columns"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3oEvs7sHOFCd"
+      },
+      "source": [
+        "Now that the drfs event is visible we have to clean and integrate it to the matrix\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": true,
+        "id": "nJJ2hWI2OFfW",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "a5f7f38a-5454-4209-ebee-a47dd464b0f1"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Original shape (genes x samples): (22283, 310)\n",
+            "Transposed shape (samples x probes): (310, 22283)\n",
+            "\n",
+            "Expression sample IDs:\n",
+            "Index(['GSM615096', 'GSM615097', 'GSM615098', 'GSM615099', 'GSM615100',\n",
+            "       'GSM615101', 'GSM615102', 'GSM615103', 'GSM615104', 'GSM615105'],\n",
+            "      dtype='object', name='name')\n"
+          ]
+        }
+      ],
+      "source": [
+        "# REBUILD EXPRESSION MATRIX\n",
+        "expression_df = pmid_soft.pivot_samples(\"VALUE\")\n",
+        "\n",
+        "print(\"Original shape (genes x samples):\", expression_df.shape)\n",
+        "\n",
+        "expression_df = expression_df.T\n",
+        "\n",
+        "print(\"Transposed shape (samples x probes):\", expression_df.shape)\n",
+        "print(\"\\nExpression sample IDs:\")\n",
+        "print(expression_df.index[:10])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "GrEYHa_DRi9U"
+      },
+      "outputs": [],
+      "source": [
+        "# CLEAN EXPRESSION INDEX\n",
+        "expression_df.index = expression_df.index.astype(str).str.strip()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "FiQMzDdSRq7T",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "44227b2d-a8f0-417e-c2d7-389f293eace9"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Columns containing 'drfs':\n",
+            "['drfs_1_event_0_censored', 'drfs_even_time_years']\n"
+          ]
+        }
+      ],
+      "source": [
+        "# FIND DRFS COLUMNS\n",
+        "drfs_cols = [col for col in metadata_df.columns if \"drfs\" in col.lower()]\n",
+        "print(\"\\nColumns containing 'drfs':\")\n",
+        "print(drfs_cols)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "collapsed": true,
+        "id": "0gx4hw89Rx4G",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "920e9ea2-2a1c-4f42-8b0e-ab2918bfebb7"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "DRFS value counts:\n",
+            "drfs_1_event_0_censored\n",
+            "0    244\n",
+            "1     66\n",
+            "Name: count, dtype: int64\n"
+          ]
+        }
+      ],
+      "source": [
+        "# CLEAN DRFS LABEL\n",
+        "metadata_df[\"drfs_1_event_0_censored\"] = pd.to_numeric(\n",
+        "    metadata_df[\"drfs_1_event_0_censored\"],\n",
+        "    errors=\"coerce\"\n",
+        ")\n",
+        "\n",
+        "print(\"\\nDRFS value counts:\")\n",
+        "print(metadata_df[\"drfs_1_event_0_censored\"].value_counts(dropna=False))"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "rJkhQxdAR4eL",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "941e26ef-369c-4349-abe2-015861cf2364"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Column 'geo_accession' has 310 matching sample IDs.\n",
+            "Example matches: ['GSM615264', 'GSM615392', 'GSM615391', 'GSM615194', 'GSM615389', 'GSM615198', 'GSM615327', 'GSM615377', 'GSM615163', 'GSM615385']\n",
+            "\n",
+            "Chosen matching column: geo_accession\n"
+          ]
+        }
+      ],
+      "source": [
+        "# SEARCH FOR THE CORRECT MATCHING ID COLUMN\n",
+        "matching_col = None\n",
+        "\n",
+        "for col in metadata_df.columns:\n",
+        "    values = metadata_df[col].astype(str).str.strip()\n",
+        "    overlap = len(set(values) & set(expression_df.index))\n",
+        "\n",
+        "    if overlap > 0:\n",
+        "        print(f\"\\nColumn '{col}' has {overlap} matching sample IDs.\")\n",
+        "\n",
+        "        # print a few example matches\n",
+        "        matches = list(set(values) & set(expression_df.index))[:10]\n",
+        "        print(\"Example matches:\", matches)\n",
+        "\n",
+        "        if matching_col is None:\n",
+        "            matching_col = col\n",
+        "\n",
+        "print(\"\\nChosen matching column:\", matching_col)\n",
+        "\n",
+        "\n",
+        "# STOP IF NO MATCH FOUND\n",
+        "if matching_col is None:\n",
+        "    raise ValueError(\"No metadata column matches expression_df sample IDs. Need to inspect metadata_df manually.\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "tqXgve51SGUp",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "16a7b314-e466-4d37-ee0f-7ea096b1d188"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Labels shape: (310, 1)\n",
+            "           drfs_1_event_0_censored\n",
+            "match_id                          \n",
+            "GSM615096                        1\n",
+            "GSM615097                        0\n",
+            "GSM615098                        0\n",
+            "GSM615099                        0\n",
+            "GSM615100                        1\n",
+            "\n",
+            "Label sample IDs:\n",
+            "Index(['GSM615096', 'GSM615097', 'GSM615098', 'GSM615099', 'GSM615100',\n",
+            "       'GSM615101', 'GSM615102', 'GSM615103', 'GSM615104', 'GSM615105'],\n",
+            "      dtype='object', name='match_id')\n"
+          ]
+        }
+      ],
+      "source": [
+        "# BUILD LABEL TABLE USING THE MATCHING COLUMN\n",
+        "labels = metadata_df[[matching_col, \"drfs_1_event_0_censored\"]].dropna()\n",
+        "labels = labels.rename(columns={matching_col: \"match_id\"})\n",
+        "labels[\"match_id\"] = labels[\"match_id\"].astype(str).str.strip()\n",
+        "labels = labels.set_index(\"match_id\")\n",
+        "\n",
+        "print(\"\\nLabels shape:\", labels.shape)\n",
+        "print(labels.head())\n",
+        "\n",
+        "print(\"\\nLabel sample IDs:\")\n",
+        "print(labels.index[:10])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "Y8Ukz3l4SJmz",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "75f789e8-e32c-4514-b707-e6495aad69ad"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Labels shape: (310, 1)\n",
+            "           drfs_1_event_0_censored\n",
+            "match_id                          \n",
+            "GSM615096                        1\n",
+            "GSM615097                        0\n",
+            "GSM615098                        0\n",
+            "GSM615099                        0\n",
+            "GSM615100                        1\n",
+            "\n",
+            "Label sample IDs:\n",
+            "Index(['GSM615096', 'GSM615097', 'GSM615098', 'GSM615099', 'GSM615100',\n",
+            "       'GSM615101', 'GSM615102', 'GSM615103', 'GSM615104', 'GSM615105'],\n",
+            "      dtype='object', name='match_id')\n"
+          ]
+        }
+      ],
+      "source": [
+        "# BUILD LABEL TABLE USING THE MATCHING COLUMN\n",
+        "labels = metadata_df[[matching_col, \"drfs_1_event_0_censored\"]].dropna()\n",
+        "labels = labels.rename(columns={matching_col: \"match_id\"})\n",
+        "labels[\"match_id\"] = labels[\"match_id\"].astype(str).str.strip()\n",
+        "labels = labels.set_index(\"match_id\")\n",
+        "\n",
+        "print(\"\\nLabels shape:\", labels.shape)\n",
+        "print(labels.head())\n",
+        "\n",
+        "print(\"\\nLabel sample IDs:\")\n",
+        "print(labels.index[:10])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "ABzC2V3HSWba",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "108ad650-60ee-427a-e35b-f0e984d96c41"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Number of matching samples: 310\n",
+            "First few matching sample IDs:\n",
+            "Index(['GSM615096', 'GSM615097', 'GSM615098', 'GSM615099', 'GSM615100',\n",
+            "       'GSM615101', 'GSM615102', 'GSM615103', 'GSM615104', 'GSM615105'],\n",
+            "      dtype='object')\n"
+          ]
+        }
+      ],
+      "source": [
+        "# FIND OVERLAPPING SAMPLES\n",
+        "common_samples = expression_df.index.intersection(labels.index)\n",
+        "\n",
+        "print(\"\\nNumber of matching samples:\", len(common_samples))\n",
+        "print(\"First few matching sample IDs:\")\n",
+        "print(common_samples[:10])\n",
+        "\n",
+        "\n",
+        "# KEEP ONLY MATCHED SAMPLES\n",
+        "expression_df_matched = expression_df.loc[common_samples]\n",
+        "labels_matched = labels.loc[common_samples]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "BPqA-qXdSaDR",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "2021cf48-d02c-43c3-a1ca-397f407df645"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "Final dataset shape: (310, 22284)\n",
+            "           1007_s_at   1053_at    117_at     121_at  1255_g_at   1294_at  \\\n",
+            "GSM615096  12.359267  8.021741  8.294244  10.840949   7.874062  8.341292   \n",
+            "GSM615097  12.613243  7.842947  5.474758  11.715159   7.347955  9.419005   \n",
+            "GSM615098  11.551397  8.776267  8.197121  10.715077   6.523880  9.297479   \n",
+            "GSM615099  11.828700  9.523877  7.907263  10.750475   6.478643  8.698516   \n",
+            "GSM615100  11.983987  6.836237  8.625010  10.827725   7.336754  9.599059   \n",
+            "\n",
+            "            1316_at   1320_at  1405_i_at   1431_at  ...  AFFX-r2-Ec-bioD-5_at  \\\n",
+            "GSM615096  8.256335  6.485966   6.495728  7.063123  ...             15.441187   \n",
+            "GSM615097  7.983681  7.330512   7.332244  7.033859  ...             16.383235   \n",
+            "GSM615098  7.820261  6.423794  10.066291  6.661108  ...             14.637426   \n",
+            "GSM615099  7.526617  7.276040  10.399704  6.642930  ...             14.218011   \n",
+            "GSM615100  8.312127  6.832232   7.090151  7.081613  ...             14.816849   \n",
+            "\n",
+            "           AFFX-r2-Hs18SrRNA-3_s_at  AFFX-r2-Hs18SrRNA-5_at  \\\n",
+            "GSM615096                 11.190421               11.955084   \n",
+            "GSM615097                 10.750292               11.610319   \n",
+            "GSM615098                 11.580917               13.075420   \n",
+            "GSM615099                 10.275100               11.467574   \n",
+            "GSM615100                 10.840278               12.989398   \n",
+            "\n",
+            "           AFFX-r2-Hs18SrRNA-M_x_at  AFFX-r2-Hs28SrRNA-3_at  \\\n",
+            "GSM615096                 12.139089               14.074549   \n",
+            "GSM615097                 12.080350               13.681936   \n",
+            "GSM615098                 12.347660               13.676062   \n",
+            "GSM615099                 10.788742               13.395810   \n",
+            "GSM615100                 11.851791               13.943810   \n",
+            "\n",
+            "           AFFX-r2-Hs28SrRNA-5_at  AFFX-r2-Hs28SrRNA-M_at  \\\n",
+            "GSM615096                9.047564               11.274337   \n",
+            "GSM615097                8.765575               11.287301   \n",
+            "GSM615098                9.450242               11.308490   \n",
+            "GSM615099                8.466331               10.208604   \n",
+            "GSM615100                8.721290               11.110014   \n",
+            "\n",
+            "           AFFX-r2-P1-cre-3_at  AFFX-r2-P1-cre-5_at  drfs_1_event_0_censored  \n",
+            "GSM615096            17.356685            17.004035                        1  \n",
+            "GSM615097            18.404677            18.143576                        0  \n",
+            "GSM615098            16.304500            15.988321                        0  \n",
+            "GSM615099            16.190139            15.807992                        0  \n",
+            "GSM615100            16.767231            16.485521                        1  \n",
+            "\n",
+            "[5 rows x 22284 columns]\n"
+          ]
+        }
+      ],
+      "source": [
+        "# MERGE EXPRESSION + LABELS\n",
+        "final_df = expression_df_matched.merge(\n",
+        "    labels_matched,\n",
+        "    left_index=True,\n",
+        "    right_index=True\n",
+        ")\n",
+        "\n",
+        "print(\"\\nFinal dataset shape:\", final_df.shape)\n",
+        "print(final_df.head())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "1U1jZ8obTuGu"
+      },
+      "source": [
+        "Now the dataset if FINALLYYY properly created and can be used to train a model."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "0eT4W1IXUlLi",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 211
+        },
+        "outputId": "606277ce-d41f-4e0c-aeda-bf9e77dbb2bf"
+      },
+      "outputs": [
+        {
+          "output_type": "error",
+          "ename": "NameError",
+          "evalue": "name 'final_df' is not defined",
+          "traceback": [
+            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
+            "\u001b[0;32m/tmp/ipykernel_35429/4143188667.py\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# SPLIT FEATURES + TARGET\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfinal_df\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"drfs_1_event_0_censored\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfinal_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"drfs_1_event_0_censored\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"X shape:\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;31mNameError\u001b[0m: name 'final_df' is not defined"
+          ]
+        }
+      ],
+      "source": [
+        "# SPLIT FEATURES + TARGET\n",
+        "X = final_df.drop(columns=[\"drfs_1_event_0_censored\"])\n",
+        "y = final_df[\"drfs_1_event_0_censored\"]\n",
+        "\n",
+        "print(\"X shape:\", X.shape)\n",
+        "print(\"y shape:\", y.shape)\n",
+        "\n",
+        "print(\"\\nClass distribution:\")\n",
+        "print(y.value_counts())\n",
+        "\n",
+        "\n",
+        "# FEATURE REDUCTION\n",
+        "gene_variance = X.var(axis=0)\n",
+        "top_genes = gene_variance.sort_values(ascending=False).head(1000).index\n",
+        "X = X[top_genes]\n",
+        "\n",
+        "print(\"\\nReduced feature count:\", X.shape[1])\n",
+        "\n",
+        "\n",
+        "# NORMALIZATION\n",
+        "scaler = StandardScaler()\n",
+        "X_scaled = scaler.fit_transform(X)\n",
+        "\n",
+        "\n",
+        "# TRAIN / TEST SPLIT\n",
+        "X_train, X_test, y_train, y_test = train_test_split(\n",
+        "    X_scaled, y,\n",
+        "    test_size=0.2,\n",
+        "    stratify=y,\n",
+        "    random_state=42\n",
+        ")\n",
+        "\n",
+        "print(\"\\nTraining set shape:\", X_train.shape)\n",
+        "print(\"Test set shape:\", X_test.shape)\n",
+        "\n",
+        "\n",
+        "# MODEL\n",
+        "model = LogisticRegression(\n",
+        "    C=100,\n",
+        "    penalty=11,\n",
+        "    solver='liblinear',\n",
+        "    max_iter=500,\n",
+        "    class_weight='balanced'\n",
+        ")\n",
+        "\n",
+        "model.fit(X_train, y_train)\n",
+        "\n",
+        "\n",
+        "# EVALUATION\n",
+        "y_pred = model.predict(X_test)\n",
+        "y_prob = model.predict_proba(X_test)[:, 1]\n",
+        "\n",
+        "metrics = {\n",
+        "    \"accuracy\": accuracy_score(y_test, y_pred),\n",
+        "    \"precision\": precision_score(y_test, y_pred),\n",
+        "    \"recall\": recall_score(y_test, y_pred),\n",
+        "    \"f1\": f1_score(y_test, y_pred),\n",
+        "    \"auc\": roc_auc_score(y_test, y_prob)\n",
+        "}\n",
+        "\n",
+        "print(\"\\nModel Performance:\")\n",
+        "for k, v in metrics.items():\n",
+        "    print(f\"{k}: {v:.4f}\")\n",
+        "\n",
+        "\n",
+        "# FEATURE IMPORTANCE\n",
+        "feature_importance = pd.Series(\n",
+        "    model.coef_[0],\n",
+        "    index=X.columns\n",
+        ").sort_values(key=abs, ascending=False)\n",
+        "\n",
+        "print(\"\\nTop Features:\")\n",
+        "print(feature_importance.head(10))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "d44K4a_6WA5e"
+      },
+      "source": [
+        "Now we visualize the results so that the model can be improved"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "sOt92J-GWHHS",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 1000
+        },
+        "outputId": "5b104002-0384-4521-aa86-3d1f7bb802d2"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 600x500 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 600x500 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 640x480 with 2 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "import matplotlib.pyplot as plt\n",
+        "from sklearn.metrics import (\n",
+        "    roc_curve,\n",
+        "    precision_recall_curve,\n",
+        "    confusion_matrix,\n",
+        "    ConfusionMatrixDisplay\n",
+        ")\n",
+        "\n",
+        "# ROC CURVE\n",
+        "fpr, tpr, _ = roc_curve(y_test, y_prob)\n",
+        "\n",
+        "plt.figure(figsize=(6, 5))\n",
+        "plt.plot(fpr, tpr, label=f\"AUC = {roc_auc_score(y_test, y_prob):.3f}\")\n",
+        "plt.plot([0, 1], [0, 1], linestyle='--')\n",
+        "plt.xlabel(\"False Positive Rate\")\n",
+        "plt.ylabel(\"True Positive Rate\")\n",
+        "plt.title(\"ROC Curve - Logistic Regression\")\n",
+        "plt.legend()\n",
+        "plt.grid(True)\n",
+        "plt.show()\n",
+        "\n",
+        "# PRECISION-RECALL CURVE\n",
+        "precision_vals, recall_vals, _ = precision_recall_curve(y_test, y_prob)\n",
+        "\n",
+        "plt.figure(figsize=(6, 5))\n",
+        "plt.plot(recall_vals, precision_vals)\n",
+        "plt.xlabel(\"Recall\")\n",
+        "plt.ylabel(\"Precision\")\n",
+        "plt.title(\"Precision-Recall Curve - Logistic Regression\")\n",
+        "plt.grid(True)\n",
+        "plt.show()\n",
+        "\n",
+        "\n",
+        "# CONFUSION MATRIX\n",
+        "cm = confusion_matrix(y_test, y_pred)\n",
+        "\n",
+        "disp = ConfusionMatrixDisplay(confusion_matrix=cm)\n",
+        "disp.plot()\n",
+        "plt.title(\"Confusion Matrix - Logistic Regression\")\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "# GETTING THE BIOLOGICAL GENES\n",
+        "\n",
+        "top_features = feature_importance.head(10)\n",
+        "\n",
+        "def clean_gene_symbol(gene):\n",
+        "    if pd.isna(gene) or gene == \"---\":\n",
+        "        return \"Unknown\"\n",
+        "    return gene.split(\"///\")[0].strip()\n",
+        "\n",
+        "mapped_genes = [\n",
+        "    clean_gene_symbol(id_to_gene.get(probe, \"Unknown\"))\n",
+        "    for probe in top_features.index\n",
+        "]\n",
+        "\n",
+        "df_top = pd.DataFrame({\n",
+        "    \"Probe\": top_features.index,\n",
+        "    \"Gene\": mapped_genes,\n",
+        "    \"Coefficient\": top_features.values\n",
+        "})\n",
+        "\n",
+        "print(df_top)"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "DPrhsGr7ffhp",
+        "outputId": "71f3060b-2f0d-46ec-e825-ebc2602126d4"
+      },
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "         Probe       Gene  Coefficient\n",
+            "0    204540_at     EEF1A2     1.518858\n",
+            "1  211737_x_at        PTN    -1.278893\n",
+            "2  218885_s_at    GALNT12    -1.155978\n",
+            "3    220084_at  C14orf105    -1.058550\n",
+            "4    207147_at       DLX2     1.039317\n",
+            "5    202989_at       RGS1    -1.022044\n",
+            "6  215112_x_at     MCF2L2     1.000583\n",
+            "7  211078_s_at       STK3     0.987127\n",
+            "8    217595_at      GSPT1     0.922447\n",
+            "9    205916_at     S100A7     0.906573\n"
+          ]
+        }
+      ]
+    }
+  ],
+  "metadata": {
+    "colab": {
+      "provenance": [],
+      "authorship_tag": "ABX9TyOdYPqxwiEHebctk4HhrcaP",
+      "include_colab_link": true
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/PMIDExampleData.pdf b/PMIDExampleData.pdf
new file mode 100644
index 00000000..f7c3095d
Binary files /dev/null and b/PMIDExampleData.pdf differ
diff --git a/READMEforBreastCancerModel.pdf b/READMEforBreastCancerModel.pdf
new file mode 100644
index 00000000..96d064e5
Binary files /dev/null and b/READMEforBreastCancerModel.pdf differ
diff --git a/bco_draft (1).json b/bco_draft (1).json
new file mode 100644
index 00000000..840ff20e
--- /dev/null
+++ b/bco_draft (1).json	
@@ -0,0 +1,240 @@
+{
+  "object_id": "https://example.org/bco/breast-cancer-drfs-gse25055",
+  "spec_version": "https://w3id.org/ieee/ieee-2791-schema/2791object.json",
+  "etag": "draft-1",
+  "provenance_domain": {
+    "name": "Breast Cancer DRFS Prediction using GEO GSE25055",
+    "version": "1.0",
+    "review_status": "unreviewed",
+    "derived_from": [],
+    "created": "2026-04-27T00:00:00Z",
+    "modified": "2026-04-27T00:00:00Z",
+    "contributors": [
+      {
+        "name": "SAMPURNA CHAKRAVORTY",
+        "affiliation": "University of Maryland - College Park",
+        "email": "sampurnachakravorty@gmail.com",
+        "contribution": [
+          "createdBy"
+        ],
+        "orcid": ""
+      }
+    ],
+    "license": "https://spdx.org/licenses/CC-BY-4.0.html"
+  },
+  "usability_domain": [
+    "This workflow processes gene expression data from GEO dataset GSE25055 to predict distant relapse-free survival (DRFS) outcomes in breast cancer patients.",
+    "The pipeline extracts metadata and expression values from SOFT files, maps probe IDs to gene symbols, and constructs a labeled dataset.",
+    "A logistic regression model is trained on normalized gene expression data using variance-based feature selection.",
+    "This workflow is intended for exploratory bioinformatics analysis and predictive modeling of cancer outcomes."
+  ],
+  "description_domain": {
+    "keywords": [
+      "breast cancer",
+      "gene expression",
+      "GEO",
+      "machine learning",
+      "logistic regression",
+      "DRFS"
+    ],
+    "pipeline_steps": [
+      {
+        "step_number": 1,
+        "name": "Upload GEO SOFT file",
+        "description": "Upload GSE25055 SOFT file into the environment"
+      },
+      {
+        "step_number": 2,
+        "name": "Parse dataset",
+        "description": "Parse SOFT file using GEOparse"
+      },
+      {
+        "step_number": 3,
+        "name": "Extract metadata",
+        "description": "Extract sample-level metadata"
+      },
+      {
+        "step_number": 4,
+        "name": "Extract expression matrix",
+        "description": "Retrieve VALUE matrix from GEO dataset"
+      },
+      {
+        "step_number": 5,
+        "name": "Transpose matrix",
+        "description": "Convert to samples x features format"
+      },
+      {
+        "step_number": 6,
+        "name": "Map probes to genes",
+        "description": "Use GPL96 annotation to map probe IDs to gene symbols"
+      },
+      {
+        "step_number": 7,
+        "name": "Clean DRFS labels",
+        "description": "Extract and clean DRFS outcome variable"
+      },
+      {
+        "step_number": 8,
+        "name": "Match sample IDs",
+        "description": "Align metadata and expression data"
+      },
+      {
+        "step_number": 9,
+        "name": "Merge dataset",
+        "description": "Combine features and labels"
+      },
+      {
+        "step_number": 10,
+        "name": "Feature selection",
+        "description": "Select top 1000 genes by variance"
+      },
+      {
+        "step_number": 11,
+        "name": "Normalization",
+        "description": "Standardize features using StandardScaler"
+      },
+      {
+        "step_number": 12,
+        "name": "Train-test split",
+        "description": "Split dataset into 80/20 stratified sets"
+      },
+      {
+        "step_number": 13,
+        "name": "Train model",
+        "description": "Train logistic regression classifier"
+      },
+      {
+        "step_number": 14,
+        "name": "Evaluate model",
+        "description": "Compute accuracy, precision, recall, F1, ROC-AUC"
+      },
+      {
+        "step_number": 15,
+        "name": "Feature importance",
+        "description": "Extract most predictive genes"
+      },
+      {
+        "step_number": 16,
+        "name": "Visualization",
+        "description": "Generate ROC, PR curve, and confusion matrix"
+      }
+    ]
+  },
+  "execution_domain": {
+    "script": [
+      {
+        "uri": "https://colab.research.google.com/",
+        "access_time": "2026-04-27T00:00:00Z"
+      }
+    ],
+    "script_driver": "python",
+    "software_prerequisites": [
+      {
+        "name": "Python",
+        "version": "3.x"
+      },
+      {
+        "name": "numpy"
+      },
+      {
+        "name": "pandas"
+      },
+      {
+        "name": "GEOparse"
+      },
+      {
+        "name": "scikit-learn"
+      },
+      {
+        "name": "matplotlib"
+      }
+    ],
+    "external_data_endpoints": [
+      {
+        "name": "GEO",
+        "url": "https://www.ncbi.nlm.nih.gov/geo/"
+      }
+    ],
+    "environment_variables": {}
+  },
+  "parametric_domain": [
+    {
+      "param": "test_size",
+      "value": "0.2"
+    },
+    {
+      "param": "random_state",
+      "value": "42"
+    },
+    {
+      "param": "feature_selection",
+      "value": "top 1000 genes by variance"
+    },
+    {
+      "param": "C",
+      "value": "100"
+    },
+    {
+      "param": "penalty",
+      "value": "l1"
+    },
+    {
+      "param": "solver",
+      "value": "liblinear"
+    },
+    {
+      "param": "max_iter",
+      "value": "500"
+    },
+    {
+      "param": "class_weight",
+      "value": "balanced"
+    }
+  ],
+  "io_domain": {
+    "input_subdomain": [
+      {
+        "filename": "GSE25055_family.soft.gz",
+        "uri": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE25055"
+      },
+      {
+        "filename": "GPL96 annotation",
+        "uri": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GPL96"
+      }
+    ],
+    "output_subdomain": [
+      {
+        "filename": "trained_model",
+        "mediatype": "application/octet-stream"
+      },
+      {
+        "filename": "metrics",
+        "mediatype": "application/json"
+      },
+      {
+        "filename": "roc_curve",
+        "mediatype": "image/png"
+      },
+      {
+        "filename": "precision_recall_curve",
+        "mediatype": "image/png"
+      },
+      {
+        "filename": "confusion_matrix",
+        "mediatype": "image/png"
+      },
+      {
+        "filename": "top_genes",
+        "mediatype": "text/csv"
+      }
+    ]
+  },
+  "error_domain": {
+    "empirical_error": {
+      "comment": "Model performance varies depending on dataset split and feature selection"
+    },
+    "algorithmic_error": {
+      "comment": "Potential overfitting due to high dimensionality of gene expression data"
+    }
+  }
+}
\ No newline at end of file