{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "54261bf3",
   "metadata": {},
   "source": [
    "# EpiTOC3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d4bcfd5",
   "metadata": {},
   "source": [
    "## Index\n",
    "1. [Instantiate model class](#Instantiate-model-class)\n",
    "2. [Define clock metadata](#Define-clock-metadata)\n",
    "3. [Download clock dependencies](#Download-clock-dependencies)\n",
    "5. [Load features](#Load-features)\n",
    "6. [Load weights into base model](#Load-weights-into-base-model)\n",
    "7. [Load reference values](#Load-reference-values)\n",
    "8. [Load preprocess and postprocess objects](#Load-preprocess-and-postprocess-objects)\n",
    "10. [Check all clock parameters](#Check-all-clock-parameters)\n",
    "10. [Basic test](#Basic-test)\n",
    "11. [Save torch model](#Save-torch-model)\n",
    "12. [Clear directory](#Clear-directory)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9923d5ad",
   "metadata": {},
   "source": [
    "Let's first import some packages:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9de93821",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:54.312184Z",
     "iopub.status.busy": "2026-07-04T19:34:54.312023Z",
     "iopub.status.idle": "2026-07-04T19:34:56.218349Z",
     "shell.execute_reply": "2026-07-04T19:34:56.217912Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import inspect\n",
    "import shutil\n",
    "import json\n",
    "import torch\n",
    "import pandas as pd\n",
    "import pyaging as pya"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6194e38f",
   "metadata": {},
   "source": [
    "## Instantiate model class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "56900f70",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.220015Z",
     "iopub.status.busy": "2026-07-04T19:34:56.219825Z",
     "iopub.status.idle": "2026-07-04T19:34:56.222560Z",
     "shell.execute_reply": "2026-07-04T19:34:56.222197Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "class EpiTOC3(epiTOC2):\n",
      "    pass\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def print_entire_class(cls):\n",
    "    source = inspect.getsource(cls)\n",
    "    print(source)\n",
    "\n",
    "print_entire_class(pya.models.EpiTOC3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c9004f70",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.223889Z",
     "iopub.status.busy": "2026-07-04T19:34:56.223803Z",
     "iopub.status.idle": "2026-07-04T19:34:56.225202Z",
     "shell.execute_reply": "2026-07-04T19:34:56.224919Z"
    }
   },
   "outputs": [],
   "source": [
    "model = pya.models.EpiTOC3()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b719befd",
   "metadata": {},
   "source": [
    "## Define clock metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6a8f25ff",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.226300Z",
     "iopub.status.busy": "2026-07-04T19:34:56.226237Z",
     "iopub.status.idle": "2026-07-04T19:34:56.228275Z",
     "shell.execute_reply": "2026-07-04T19:34:56.227931Z"
    }
   },
   "outputs": [],
   "source": [
    "model.metadata[\"clock_name\"] = \"epitoc3\"\n",
    "model.metadata[\"data_type\"] = \"DNA methylation\"  # Paper: The study constructs clocks from DNA methylation measurements.\n",
    "model.metadata[\"species\"] = \"Homo sapiens\"  # Paper: The analyzed samples and clock are human.\n",
    "model.metadata[\"year\"] = 2020\n",
    "model.metadata[\"approved_by_author\"] = \"⌛\"\n",
    "model.metadata[\"citation\"] = \"Teschendorff, Andrew E. \\\"A comparison of epigenetic mitotic-like clocks for cancer risk prediction.\\\" Genome Medicine 12 (2020): 56.\"\n",
    "model.metadata[\"doi\"] = \"https://doi.org/10.1186/s13073-020-00752-3\"\n",
    "model.metadata[\"notes\"] = \"Code-defined 170-CpG extension of the dynamic mitotic model. Official EpiMitClocks data show that all 170 sites are a subset of the 371 stemTOC vivo-mitCpGs derived from fetal/neonatal references, six normal proliferating cell lines, and three adult whole-blood cohorts. The assigned 2020 dynamic-model paper does not name or define epiTOC3.\"\n",
    "model.metadata[\"research_only\"] = None\n",
    "model.metadata[\"tissue\"] = [\"cultured primary human cells\", \"whole blood\", \"multi-tissue\", \"cord blood\"]  # Paper: epiTOC3 uses 170 unmethylated population-doubling-associated CpGs.\n",
    "model.metadata[\"predicts\"] = [\"mitotic age\"]  # Paper: The function directly estimates cumulative stem-cell divisions in a tissue.\n",
    "model.metadata[\"training_target\"] = [\"population doublings\"]  # Paper: The epiTOC3 signature comprises population-doubling-associated CpGs.\n",
    "model.metadata[\"unit\"] = [\"cell divisions per stem cell\"]  # Paper: The returned TNSC is cumulative stem-cell divisions.\n",
    "model.metadata[\"model_type\"] = \"dynamic methylation transmission model\"  # Paper: The function describes epiTOC3 as building on a dynamic DNA methylation-gain model.\n",
    "model.metadata[\"platform\"] = [\"Illumina 450K\", \"Illumina EPIC\"]  # Paper: Input is an Illumina 450k/EPIC DNAm beta matrix.\n",
    "model.metadata[\"population\"] = \"all ages\"  # Paper: The derivation uses fetal/neonatal reference tissues, six normal cell lines, and adult whole-blood cohorts.\n",
    "model.metadata[\"journal\"] = \"Genome Medicine\"\n",
    "model.metadata[\"last_author\"] = \"Andrew E. Teschendorff\"\n",
    "model.metadata[\"n_features\"] = 170\n",
    "model.metadata[\"citations\"] = 155\n",
    "model.metadata[\"citations_date\"] = \"2026-07-05\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd266538",
   "metadata": {},
   "source": [
    "## Download clock dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9fe34941",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.229526Z",
     "iopub.status.busy": "2026-07-04T19:34:56.229447Z",
     "iopub.status.idle": "2026-07-04T19:34:56.508620Z",
     "shell.execute_reply": "2026-07-04T19:34:56.507708Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "supplementary_url = \"https://raw.githubusercontent.com/Duzhaozhen/OmniAge/c10fbe8cb92957520fbff1d55ae1def0691252e5/OmniAgePy/src/omniage/data/epiTOC3.csv\"\n",
    "supplementary_file_name = \"coefficients.csv\"\n",
    "os.system(f\"curl -sL -o {supplementary_file_name} {supplementary_url}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90d7d55b",
   "metadata": {},
   "source": [
    "## Load features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "abcc82f1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.510801Z",
     "iopub.status.busy": "2026-07-04T19:34:56.510636Z",
     "iopub.status.idle": "2026-07-04T19:34:56.516181Z",
     "shell.execute_reply": "2026-07-04T19:34:56.515554Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('coefficients.csv')\n",
    "model.features = df['probe'].astype(str).tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7cf714a",
   "metadata": {},
   "source": [
    "## Load weights into base model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0878d97d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.518286Z",
     "iopub.status.busy": "2026-07-04T19:34:56.518148Z",
     "iopub.status.idle": "2026-07-04T19:34:56.522012Z",
     "shell.execute_reply": "2026-07-04T19:34:56.521391Z"
    }
   },
   "outputs": [],
   "source": [
    "model.delta = torch.tensor(df['delta'].values, dtype=torch.float32).unsqueeze(0)\n",
    "model.beta0 = torch.tensor(df['beta0'].values, dtype=torch.float32).unsqueeze(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "14ce9dc1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.523507Z",
     "iopub.status.busy": "2026-07-04T19:34:56.523405Z",
     "iopub.status.idle": "2026-07-04T19:34:56.525526Z",
     "shell.execute_reply": "2026-07-04T19:34:56.524933Z"
    }
   },
   "outputs": [],
   "source": [
    "model.base_model = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa0e5a5a",
   "metadata": {},
   "source": [
    "## Load reference values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f58a0c2f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.526947Z",
     "iopub.status.busy": "2026-07-04T19:34:56.526843Z",
     "iopub.status.idle": "2026-07-04T19:34:56.528797Z",
     "shell.execute_reply": "2026-07-04T19:34:56.528342Z"
    }
   },
   "outputs": [],
   "source": [
    "model.reference_values = [-1]*len(model.features)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4823572",
   "metadata": {},
   "source": [
    "## Load preprocess and postprocess objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a2d013da",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.530289Z",
     "iopub.status.busy": "2026-07-04T19:34:56.530200Z",
     "iopub.status.idle": "2026-07-04T19:34:56.532403Z",
     "shell.execute_reply": "2026-07-04T19:34:56.531683Z"
    }
   },
   "outputs": [],
   "source": [
    "model.preprocess_name = \"nan_to_zero\"\n",
    "model.preprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ae5ec67a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.533804Z",
     "iopub.status.busy": "2026-07-04T19:34:56.533689Z",
     "iopub.status.idle": "2026-07-04T19:34:56.535772Z",
     "shell.execute_reply": "2026-07-04T19:34:56.535327Z"
    }
   },
   "outputs": [],
   "source": [
    "model.postprocess_name = None\n",
    "model.postprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acdc6c8c",
   "metadata": {},
   "source": [
    "## Check all clock parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "33671fa4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.537042Z",
     "iopub.status.busy": "2026-07-04T19:34:56.536949Z",
     "iopub.status.idle": "2026-07-04T19:34:56.539409Z",
     "shell.execute_reply": "2026-07-04T19:34:56.538914Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Attributes:\n",
      "\n",
      "training: True\n",
      "metadata: {'approved_by_author': '⌛',\n",
      " 'citation': 'Teschendorff, Andrew E. \"A comparison of epigenetic mitotic-like '\n",
      "             'clocks for cancer risk prediction.\" Genome Medicine 12.1 (2020): '\n",
      "             '56.',\n",
      " 'clock_name': 'epitoc3',\n",
      " 'data_type': 'methylation',\n",
      " 'doi': 'https://doi.org/10.1186/s13073-020-00752-3',\n",
      " 'notes': None,\n",
      " 'research_only': None,\n",
      " 'species': 'Homo sapiens',\n",
      " 'version': None,\n",
      " 'year': 2020}\n",
      "reference_values: [-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1]... [Total elements: 170]\n",
      "preprocess_name: 'nan_to_zero'\n",
      "preprocess_dependencies: None\n",
      "postprocess_name: None\n",
      "postprocess_dependencies: None\n",
      "features: ['cg23062112', 'cg18560328', 'cg21229268', 'cg11250773', 'cg12892303', 'cg12781700', 'cg12065366', 'cg17712694', 'cg12799781', 'cg01718742', 'cg10263370', 'cg16491617', 'cg11832210', 'cg07686635', 'cg26510017', 'cg15745900', 'cg15402399', 'cg05009934', 'cg10094616', 'cg06010588', 'cg23847712', 'cg17370163', 'cg13428480', 'cg22620221', 'cg16437728', 'cg25984671', 'cg07882671', 'cg04342955', 'cg14482313', 'cg14537533']... [Total elements: 170]\n",
      "base_model_features: None\n",
      "base_model: None\n",
      "delta: [9.999999747378752e-05, 9.999999747378752e-05, 9.999999747378752e-05, 0.0002500000118743628, 9.999999747378752e-05, 9.999999747378752e-05, 9.999999747378752e-05, 9.999999747378752e-05, 4.999999873689376e-05, 0.0002500000118743628, 0.0002500000118743628, 9.999999747378752e-05, 9.999999747378752e-06, 9.999999747378752e-06, 4.999999873689376e-05, 4.999999873689376e-05, 9.999999747378752e-06, 4.999999873689376e-05, 4.999999873689376e-05, 0.0002500000118743628, 0.0002500000118743628, 0.0002500000118743628, 4.999999873689376e-05, 0.0002500000118743628, 9.999999747378752e-05, 9.999999747378752e-05, 9.999999747378752e-05, 4.999999873689376e-05, 0.0002500000118743628, 9.999999747378752e-05]... [Tensor of shape torch.Size([1, 170])]\n",
      "beta0: [0.05000000074505806, 0.019999999552965164, 0.029999999329447746, 0.0, 0.03999999910593033, 0.03999999910593033, 0.05000000074505806, 0.05000000074505806, 0.029999999329447746, 0.009999999776482582, 0.019999999552965164, 0.029999999329447746, 0.03999999910593033, 0.05000000074505806, 0.019999999552965164, 0.03999999910593033, 0.05000000074505806, 0.03999999910593033, 0.05000000074505806, 0.05000000074505806, 0.05000000074505806, 0.05000000074505806, 0.03999999910593033, 0.05000000074505806, 0.05000000074505806, 0.05000000074505806, 0.019999999552965164, 0.05000000074505806, 0.0, 0.019999999552965164]... [Tensor of shape torch.Size([1, 170])]\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Structure:\n",
      "\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Parameters and Weights:\n",
      "\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pya.utils.print_model_details(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5eaa8b63",
   "metadata": {},
   "source": [
    "## Basic test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3c76bc4f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.540875Z",
     "iopub.status.busy": "2026-07-04T19:34:56.540772Z",
     "iopub.status.idle": "2026-07-04T19:34:56.549368Z",
     "shell.execute_reply": "2026-07-04T19:34:56.548897Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 1921.6122],\n",
       "        [-4698.6198],\n",
       "        [ 1365.4274],\n",
       "        [-4247.0676],\n",
       "        [  -35.5126],\n",
       "        [  116.7216],\n",
       "        [ 7356.7134],\n",
       "        [-3367.8507],\n",
       "        [ 7780.6034],\n",
       "        [-7170.8200]], dtype=torch.float64)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.manual_seed(42)\n",
    "input = torch.randn(10, len(model.features), dtype=float)\n",
    "model.eval()\n",
    "model.to(float)\n",
    "pred = model(input)\n",
    "pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aca67e0a",
   "metadata": {},
   "source": [
    "## Save torch model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c96ff956",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.550699Z",
     "iopub.status.busy": "2026-07-04T19:34:56.550614Z",
     "iopub.status.idle": "2026-07-04T19:34:56.554088Z",
     "shell.execute_reply": "2026-07-04T19:34:56.553609Z"
    }
   },
   "outputs": [],
   "source": [
    "torch.save(model, f\"../weights/{model.metadata['clock_name']}.pt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68b814e5",
   "metadata": {},
   "source": [
    "## Clear directory\n",
    "<a id=\"10\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "20186cd6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:56.555454Z",
     "iopub.status.busy": "2026-07-04T19:34:56.555368Z",
     "iopub.status.idle": "2026-07-04T19:34:56.559114Z",
     "shell.execute_reply": "2026-07-04T19:34:56.558697Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Deleted file: coefficients.csv\n"
     ]
    }
   ],
   "source": [
    "# Function to remove a folder and all its contents\n",
    "def remove_folder(path):\n",
    "    try:\n",
    "        shutil.rmtree(path)\n",
    "        print(f\"Deleted folder: {path}\")\n",
    "    except Exception as e:\n",
    "        print(f\"Error deleting folder {path}: {e}\")\n",
    "\n",
    "# Get a list of all files and folders in the current directory\n",
    "all_items = os.listdir('.')\n",
    "\n",
    "# Loop through the items\n",
    "for item in all_items:\n",
    "    # Check if it's a file and does not end with .ipynb\n",
    "    if os.path.isfile(item) and not item.endswith('.ipynb'):\n",
    "        os.remove(item)\n",
    "        print(f\"Deleted file: {item}\")\n",
    "    # Check if it's a folder\n",
    "    elif os.path.isdir(item):\n",
    "        remove_folder(item)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}