{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "21791569",
   "metadata": {},
   "source": [
    "# VidalBralo"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93a90ac5",
   "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": "dcd636f4",
   "metadata": {},
   "source": [
    "Let's first import some packages:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2350ba0e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:30.379901Z",
     "iopub.status.busy": "2026-07-04T19:10:30.379559Z",
     "iopub.status.idle": "2026-07-04T19:10:37.575026Z",
     "shell.execute_reply": "2026-07-04T19:10:37.574076Z"
    }
   },
   "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": "345bcbc8",
   "metadata": {},
   "source": [
    "## Instantiate model class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "26516fe2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.578279Z",
     "iopub.status.busy": "2026-07-04T19:10:37.577926Z",
     "iopub.status.idle": "2026-07-04T19:10:37.583421Z",
     "shell.execute_reply": "2026-07-04T19:10:37.582386Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "class VidalBralo(pyagingModel):\n",
      "    def __init__(self):\n",
      "        super().__init__()\n",
      "\n",
      "    def preprocess(self, x):\n",
      "        if self.reference_values is None:\n",
      "            return x\n",
      "        if isinstance(self.reference_values, torch.Tensor):\n",
      "            reference = self.reference_values.to(device=x.device, dtype=x.dtype)\n",
      "        else:\n",
      "            reference = torch.tensor(self.reference_values, device=x.device, dtype=x.dtype)\n",
      "        return torch.where(torch.isnan(x), reference, x)\n",
      "\n",
      "    def postprocess(self, x):\n",
      "        return x\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def print_entire_class(cls):\n",
    "    source = inspect.getsource(cls)\n",
    "    print(source)\n",
    "\n",
    "print_entire_class(pya.models.VidalBralo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c8a99eb0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.585260Z",
     "iopub.status.busy": "2026-07-04T19:10:37.585073Z",
     "iopub.status.idle": "2026-07-04T19:10:37.588151Z",
     "shell.execute_reply": "2026-07-04T19:10:37.586971Z"
    }
   },
   "outputs": [],
   "source": [
    "model = pya.models.VidalBralo()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f96ce673",
   "metadata": {},
   "source": [
    "## Define clock metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "61667f70",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.592761Z",
     "iopub.status.busy": "2026-07-04T19:10:37.592286Z",
     "iopub.status.idle": "2026-07-04T19:10:37.599245Z",
     "shell.execute_reply": "2026-07-04T19:10:37.598192Z"
    }
   },
   "outputs": [],
   "source": [
    "model.metadata[\"clock_name\"] = \"vidalbralo\"\n",
    "model.metadata[\"data_type\"] = \"DNA methylation\"  # Paper: DNA methylation age measures\n",
    "model.metadata[\"species\"] = \"Homo sapiens\"  # Paper: 390 healthy Caucasian donors\n",
    "model.metadata[\"year\"] = 2016\n",
    "model.metadata[\"approved_by_author\"] = \"⌛\"\n",
    "model.metadata[\"citation\"] = \"Vidal-Bralo, Laura, Yolanda Lopez-Golan, and Antonio Gonzalez. \\\"Simplified assay for epigenetic age estimation in whole blood of adults.\\\" Frontiers in Genetics 7 (2016): 126.\"\n",
    "model.metadata[\"doi\"] = \"https://doi.org/10.3389/fgene.2016.00126\"\n",
    "model.metadata[\"notes\"] = \"Eight-CpG whole-blood chronological-age estimator selected by forward stepwise regression in 390 adults and calibrated by multiple linear regression; CpGs were chosen for compatibility with a single multiplex MS-SNuPE assay.\"\n",
    "model.metadata[\"research_only\"] = None\n",
    "model.metadata[\"tissue\"] = [\"whole blood\"]  # Paper: whole blood (WB)\n",
    "model.metadata[\"predicts\"] = [\"chronological age\"]  # Paper: estimate age from blood DNA\n",
    "model.metadata[\"training_target\"] = [\"chronological age\"]  # Paper: forward stepwise linear regression with age\n",
    "model.metadata[\"unit\"] = [\"years\"]  # Paper: Age range; MAD ... years\n",
    "model.metadata[\"model_type\"] = \"linear regression\"  # Paper: multiple linear regression parameters of the 8 CpG DmAM\n",
    "model.metadata[\"platform\"] = [\"Illumina 27K\"]  # Paper: training ... obtained with the Illumina Human Methylation 27K BeadChip\n",
    "model.metadata[\"population\"] = \"adults\"  # Paper: 390 healthy subjects older than 20 years\n",
    "model.metadata[\"journal\"] = \"Frontiers in Genetics\"\n",
    "model.metadata[\"last_author\"] = \"Antonio Gonzalez\"\n",
    "model.metadata[\"n_features\"] = 8\n",
    "model.metadata[\"citations\"] = 145\n",
    "model.metadata[\"citations_date\"] = \"2026-07-05\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b49ab90b",
   "metadata": {},
   "source": [
    "## Download clock dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "63c7ad3c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.601499Z",
     "iopub.status.busy": "2026-07-04T19:10:37.601292Z",
     "iopub.status.idle": "2026-07-04T19:10:37.604147Z",
     "shell.execute_reply": "2026-07-04T19:10:37.603043Z"
    }
   },
   "outputs": [],
   "source": [
    "# The 8-CpG DmAM coefficients are given directly in Table 2 of\n",
    "# Vidal-Bralo et al. 2016 (Front. Genet. 7:126); they are defined inline below,\n",
    "# so no external download is required."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "357edd3a",
   "metadata": {},
   "source": [
    "## Load features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "55100340",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.607284Z",
     "iopub.status.busy": "2026-07-04T19:10:37.607099Z",
     "iopub.status.idle": "2026-07-04T19:10:37.615611Z",
     "shell.execute_reply": "2026-07-04T19:10:37.614591Z"
    }
   },
   "outputs": [],
   "source": [
    "# Multiple linear regression parameters of the 8 CpG DmAM (Vidal-Bralo et al. 2016, Table 2)\n",
    "coef_df = pd.DataFrame({\n",
    "    'CpG': ['cg16386080', 'cg24768561', 'cg19761273', 'cg25809905',\n",
    "            'cg09809672', 'cg02228185', 'cg17471102', 'cg10917602'],\n",
    "    'coefficient': [59.5, 33.9, -44.0, -19.7, -22.8, -16.8, -17.7, -11.4],\n",
    "})\n",
    "model.features = coef_df['CpG'].tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9a39d6c",
   "metadata": {},
   "source": [
    "## Load weights into base model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a1ec53d8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.617562Z",
     "iopub.status.busy": "2026-07-04T19:10:37.617386Z",
     "iopub.status.idle": "2026-07-04T19:10:37.625704Z",
     "shell.execute_reply": "2026-07-04T19:10:37.624815Z"
    }
   },
   "outputs": [],
   "source": [
    "weights = torch.tensor(coef_df['coefficient'].tolist()).unsqueeze(0).float()\n",
    "intercept = torch.tensor([84.7]).float()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "8d1f4eaa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.628313Z",
     "iopub.status.busy": "2026-07-04T19:10:37.628021Z",
     "iopub.status.idle": "2026-07-04T19:10:37.635702Z",
     "shell.execute_reply": "2026-07-04T19:10:37.634965Z"
    }
   },
   "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": "128eb0a2",
   "metadata": {},
   "source": [
    "## Load reference values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "355c423c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.639378Z",
     "iopub.status.busy": "2026-07-04T19:10:37.639018Z",
     "iopub.status.idle": "2026-07-04T19:10:37.643684Z",
     "shell.execute_reply": "2026-07-04T19:10:37.642477Z"
    }
   },
   "outputs": [],
   "source": [
    "model.reference_values = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "824ad0a5",
   "metadata": {},
   "source": [
    "## Load preprocess and postprocess objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "48fbb8c8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.645964Z",
     "iopub.status.busy": "2026-07-04T19:10:37.645805Z",
     "iopub.status.idle": "2026-07-04T19:10:37.649077Z",
     "shell.execute_reply": "2026-07-04T19:10:37.647615Z"
    }
   },
   "outputs": [],
   "source": [
    "model.preprocess_name = None\n",
    "model.preprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4b4a1619",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.652166Z",
     "iopub.status.busy": "2026-07-04T19:10:37.652004Z",
     "iopub.status.idle": "2026-07-04T19:10:37.655693Z",
     "shell.execute_reply": "2026-07-04T19:10:37.654613Z"
    }
   },
   "outputs": [],
   "source": [
    "model.postprocess_name = None\n",
    "model.postprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0faec4a7",
   "metadata": {},
   "source": [
    "## Check all clock parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "dac05a3d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.660253Z",
     "iopub.status.busy": "2026-07-04T19:10:37.659795Z",
     "iopub.status.idle": "2026-07-04T19:10:37.687764Z",
     "shell.execute_reply": "2026-07-04T19:10:37.686787Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Attributes:\n",
      "\n",
      "training: True\n",
      "metadata: {'approved_by_author': '⌛',\n",
      " 'citation': 'Vidal-Bralo, Laura, Yolanda Lopez-Golan, and Antonio Gonzalez. '\n",
      "             '\"Simplified assay for epigenetic age estimation in whole blood '\n",
      "             'of adults.\" Frontiers in genetics 7 (2016): 126.',\n",
      " 'clock_name': 'vidalbralo',\n",
      " 'data_type': 'methylation',\n",
      " 'doi': 'https://doi.org/10.3389/fgene.2016.00126',\n",
      " 'notes': None,\n",
      " 'research_only': None,\n",
      " 'species': 'Homo sapiens',\n",
      " 'version': None,\n",
      " 'year': 2016}\n",
      "reference_values: None\n",
      "preprocess_name: None\n",
      "preprocess_dependencies: None\n",
      "postprocess_name: None\n",
      "postprocess_dependencies: None\n",
      "features: ['cg16386080',\n",
      " 'cg24768561',\n",
      " 'cg19761273',\n",
      " 'cg25809905',\n",
      " 'cg09809672',\n",
      " 'cg02228185',\n",
      " 'cg17471102',\n",
      " 'cg10917602']\n",
      "base_model_features: None\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Structure:\n",
      "\n",
      "base_model: LinearModel(\n",
      "  (linear): Linear(in_features=8, out_features=1, bias=True)\n",
      ")\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Parameters and Weights:\n",
      "\n",
      "base_model.linear.weight: tensor([[ 59.5000,  33.9000, -44.0000, -19.7000, -22.8000, -16.8000, -17.7000,\n",
      "         -11.4000]])\n",
      "base_model.linear.bias: tensor([84.7000])\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pya.utils.print_model_details(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fca3f6f",
   "metadata": {},
   "source": [
    "## Basic test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6297584f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.689696Z",
     "iopub.status.busy": "2026-07-04T19:10:37.689530Z",
     "iopub.status.idle": "2026-07-04T19:10:37.705864Z",
     "shell.execute_reply": "2026-07-04T19:10:37.704964Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[126.0523],\n",
       "        [104.1466],\n",
       "        [ 75.2432],\n",
       "        [ 58.5473],\n",
       "        [160.9427],\n",
       "        [ 27.2660],\n",
       "        [137.7398],\n",
       "        [262.1763],\n",
       "        [125.6747],\n",
       "        [171.3440]], 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": "cbbe2378",
   "metadata": {},
   "source": [
    "## Save torch model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7ba9465a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.707817Z",
     "iopub.status.busy": "2026-07-04T19:10:37.707673Z",
     "iopub.status.idle": "2026-07-04T19:10:37.714323Z",
     "shell.execute_reply": "2026-07-04T19:10:37.713380Z"
    }
   },
   "outputs": [],
   "source": [
    "torch.save(model, f\"../weights/{model.metadata['clock_name']}.pt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "619ea422",
   "metadata": {},
   "source": [
    "## Clear directory\n",
    "<a id=\"10\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "bfb7a130",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T19:10:37.716060Z",
     "iopub.status.busy": "2026-07-04T19:10:37.715875Z",
     "iopub.status.idle": "2026-07-04T19:10:37.720797Z",
     "shell.execute_reply": "2026-07-04T19:10:37.720150Z"
    }
   },
   "outputs": [],
   "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
}