{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d98099fd",
   "metadata": {},
   "source": [
    "# StemTOCvitro"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e768463",
   "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": "3c93398b",
   "metadata": {},
   "source": [
    "Let's first import some packages:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "68253f85",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:41.988486Z",
     "iopub.status.busy": "2026-07-04T19:34:41.988302Z",
     "iopub.status.idle": "2026-07-04T19:34:43.875297Z",
     "shell.execute_reply": "2026-07-04T19:34:43.874841Z"
    }
   },
   "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": "f89eb2f3",
   "metadata": {},
   "source": [
    "## Instantiate model class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a960596b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:43.876910Z",
     "iopub.status.busy": "2026-07-04T19:34:43.876746Z",
     "iopub.status.idle": "2026-07-04T19:34:43.879335Z",
     "shell.execute_reply": "2026-07-04T19:34:43.879008Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "class StemTOCvitro(stemTOC):\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.StemTOCvitro)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "302a7c18",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:43.880651Z",
     "iopub.status.busy": "2026-07-04T19:34:43.880568Z",
     "iopub.status.idle": "2026-07-04T19:34:43.882183Z",
     "shell.execute_reply": "2026-07-04T19:34:43.881906Z"
    }
   },
   "outputs": [],
   "source": [
    "model = pya.models.StemTOCvitro()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d13953b",
   "metadata": {},
   "source": [
    "## Define clock metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "35a15e9a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:43.883351Z",
     "iopub.status.busy": "2026-07-04T19:34:43.883283Z",
     "iopub.status.idle": "2026-07-04T19:34:43.885411Z",
     "shell.execute_reply": "2026-07-04T19:34:43.885027Z"
    }
   },
   "outputs": [],
   "source": [
    "model.metadata[\"clock_name\"] = \"stemtocvitro\"\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\"] = 2024\n",
    "model.metadata[\"approved_by_author\"] = \"⌛\"\n",
    "model.metadata[\"citation\"] = \"Zhu, Tianlei, et al. \\\"An improved epigenetic counter to track mitotic age in cells.\\\" Nature Communications 15 (2024): 4211.\"\n",
    "model.metadata[\"doi\"] = \"https://doi.org/10.1038/s41467-024-48649-8\"\n",
    "model.metadata[\"notes\"] = \"In-vitro precursor of stemTOC based on the 95th percentile across 629 population-doubling-associated CpGs.\"\n",
    "model.metadata[\"research_only\"] = None\n",
    "model.metadata[\"tissue\"] = [\"multi-tissue\", \"cultured human cells\"]  # Paper: Feature selection used fetal/neonatal reference tissues and cultured normal cell types; stemTOC also used blood cohorts.\n",
    "model.metadata[\"predicts\"] = [\"mitotic age\"]  # Paper: The paper describes the returned score as relative mitotic age.\n",
    "model.metadata[\"training_target\"] = [\"population doublings\"]  # Paper: Candidate CpGs were selected for methylation gain with population doublings and, where applicable, age in blood.\n",
    "model.metadata[\"unit\"] = [\"beta value\"]  # Paper: The score is the 95% upper quantile of CpG DNAm beta values.\n",
    "model.metadata[\"model_type\"] = \"95th-percentile methylation aggregation\"  # Paper: The score is computed as the 95% upper quantile across selected CpGs.\n",
    "model.metadata[\"platform\"] = [\"Illumina 450K\", \"Illumina EPIC\"]  # Paper: Derivation datasets included 450K fetal samples and EPIC cultured-cell samples.\n",
    "model.metadata[\"population\"] = \"prenatal and newborn\"  # Paper: The derivation cohorts are fetal/neonatal tissues, normal cell cultures, and for stemTOC adult blood cohorts.\n",
    "model.metadata[\"journal\"] = \"Nature Communications\"\n",
    "model.metadata[\"last_author\"] = \"Andrew E. Teschendorff\"\n",
    "model.metadata[\"n_features\"] = 629\n",
    "model.metadata[\"citations\"] = 24\n",
    "model.metadata[\"citations_date\"] = \"2026-07-05\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94b38fbd",
   "metadata": {},
   "source": [
    "## Download clock dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f8ffb424",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:43.886594Z",
     "iopub.status.busy": "2026-07-04T19:34:43.886525Z",
     "iopub.status.idle": "2026-07-04T19:34:44.162319Z",
     "shell.execute_reply": "2026-07-04T19:34:44.161610Z"
    }
   },
   "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/StemTOCvitro.csv\"\n",
    "supplementary_file_name = \"coefficients.csv\"\n",
    "os.system(f\"curl -sL -o {supplementary_file_name} {supplementary_url}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46faed2e",
   "metadata": {},
   "source": [
    "## Load features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d7cc4659",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.164789Z",
     "iopub.status.busy": "2026-07-04T19:34:44.164593Z",
     "iopub.status.idle": "2026-07-04T19:34:44.172169Z",
     "shell.execute_reply": "2026-07-04T19:34:44.171496Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('coefficients.csv')\n",
    "if str(df.columns[0]).startswith('Unnamed'):\n",
    "    df = df.iloc[:, 1:]\n",
    "model.features = df['probe'].tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1bd735e7",
   "metadata": {},
   "source": [
    "## Load weights into base model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "86132fef",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.174045Z",
     "iopub.status.busy": "2026-07-04T19:34:44.173887Z",
     "iopub.status.idle": "2026-07-04T19:34:44.176727Z",
     "shell.execute_reply": "2026-07-04T19:34:44.175989Z"
    }
   },
   "outputs": [],
   "source": [
    "weights = torch.tensor([1.0]).unsqueeze(0)\n",
    "intercept = torch.tensor([0.0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "556797de",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.178441Z",
     "iopub.status.busy": "2026-07-04T19:34:44.178281Z",
     "iopub.status.idle": "2026-07-04T19:34:44.181691Z",
     "shell.execute_reply": "2026-07-04T19:34:44.181030Z"
    }
   },
   "outputs": [],
   "source": [
    "base_model = pya.models.LinearModel(input_dim=len(model.features))\n",
    "\n",
    "base_model.linear.weight.data = weights.float()\n",
    "base_model.linear.bias.data = intercept.float()\n",
    "\n",
    "model.base_model = base_model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d5f9e4f",
   "metadata": {},
   "source": [
    "## Load reference values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "21a23bd7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.183360Z",
     "iopub.status.busy": "2026-07-04T19:34:44.183239Z",
     "iopub.status.idle": "2026-07-04T19:34:44.185467Z",
     "shell.execute_reply": "2026-07-04T19:34:44.184918Z"
    }
   },
   "outputs": [],
   "source": [
    "model.reference_values = [-1]*len(model.features)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a207b9ef",
   "metadata": {},
   "source": [
    "## Load preprocess and postprocess objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "88c05f98",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.187290Z",
     "iopub.status.busy": "2026-07-04T19:34:44.187160Z",
     "iopub.status.idle": "2026-07-04T19:34:44.189586Z",
     "shell.execute_reply": "2026-07-04T19:34:44.189134Z"
    }
   },
   "outputs": [],
   "source": [
    "model.preprocess_name = \"0.95 quantile\"\n",
    "model.preprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1636e3dd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.191384Z",
     "iopub.status.busy": "2026-07-04T19:34:44.191247Z",
     "iopub.status.idle": "2026-07-04T19:34:44.193184Z",
     "shell.execute_reply": "2026-07-04T19:34:44.192701Z"
    }
   },
   "outputs": [],
   "source": [
    "model.postprocess_name = None\n",
    "model.postprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2fe33f5",
   "metadata": {},
   "source": [
    "## Check all clock parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ad9e9b16",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.194748Z",
     "iopub.status.busy": "2026-07-04T19:34:44.194636Z",
     "iopub.status.idle": "2026-07-04T19:34:44.198601Z",
     "shell.execute_reply": "2026-07-04T19:34:44.198060Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Attributes:\n",
      "\n",
      "training: True\n",
      "metadata: {'approved_by_author': '⌛',\n",
      " 'citation': 'Zhu, Tianyu, et al. \"A pan-tissue DNA methylation atlas enables '\n",
      "             'in silico decomposition of human tissue methylomes at cell-type '\n",
      "             'resolution.\" Nature Communications 15 (2024).',\n",
      " 'clock_name': 'stemtocvitro',\n",
      " 'data_type': 'methylation',\n",
      " 'doi': 'https://doi.org/10.1038/s41467-024-48649-8',\n",
      " 'notes': None,\n",
      " 'research_only': None,\n",
      " 'species': 'Homo sapiens',\n",
      " 'version': None,\n",
      " 'year': 2024}\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: 629]\n",
      "preprocess_name: '0.95 quantile'\n",
      "preprocess_dependencies: None\n",
      "postprocess_name: None\n",
      "postprocess_dependencies: None\n",
      "features: ['cg20327258', 'cg23062112', 'cg07365816', 'cg08659394', 'cg14575559', 'cg13294856', 'cg18560328', 'cg21229268', 'cg20078466', 'cg23091824', 'cg03865257', 'cg16443455', 'cg11250773', 'cg10687823', 'cg12892303', 'cg12781700', 'cg13583934', 'cg12065366', 'cg25203031', 'cg17404915', 'cg22299454', 'cg10698404', 'cg09097345', 'cg05404701', 'cg24010885', 'cg17712694', 'cg20707222', 'cg08901752', 'cg11297107', 'cg17876581']... [Total elements: 629]\n",
      "base_model_features: None\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Structure:\n",
      "\n",
      "base_model: LinearModel(\n",
      "  (linear): Linear(in_features=629, out_features=1, bias=True)\n",
      ")\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Parameters and Weights:\n",
      "\n",
      "base_model.linear.weight: tensor([[1.]])\n",
      "base_model.linear.bias: tensor([0.])\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pya.utils.print_model_details(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f6d1733",
   "metadata": {},
   "source": [
    "## Basic test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a2461127",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.200005Z",
     "iopub.status.busy": "2026-07-04T19:34:44.199909Z",
     "iopub.status.idle": "2026-07-04T19:34:44.217490Z",
     "shell.execute_reply": "2026-07-04T19:34:44.217008Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[1.8129],\n",
       "        [1.7694],\n",
       "        [1.5825],\n",
       "        [1.5507],\n",
       "        [1.6496],\n",
       "        [1.5994],\n",
       "        [1.6658],\n",
       "        [1.6827],\n",
       "        [1.6054],\n",
       "        [1.7836]], dtype=torch.float64, grad_fn=<AddmmBackward0>)"
      ]
     },
     "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": "81ce6867",
   "metadata": {},
   "source": [
    "## Save torch model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e397d341",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.218795Z",
     "iopub.status.busy": "2026-07-04T19:34:44.218698Z",
     "iopub.status.idle": "2026-07-04T19:34:44.222110Z",
     "shell.execute_reply": "2026-07-04T19:34:44.221816Z"
    }
   },
   "outputs": [],
   "source": [
    "torch.save(model, f\"../weights/{model.metadata['clock_name']}.pt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "742514a1",
   "metadata": {},
   "source": [
    "## Clear directory\n",
    "<a id=\"10\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a5ac5c01",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:34:44.223540Z",
     "iopub.status.busy": "2026-07-04T19:34:44.223471Z",
     "iopub.status.idle": "2026-07-04T19:34:44.226987Z",
     "shell.execute_reply": "2026-07-04T19:34:44.226615Z"
    }
   },
   "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
}