{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "717f0170",
   "metadata": {},
   "source": [
    "# Bohlin"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ef3e665",
   "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": "c0f66981",
   "metadata": {},
   "source": [
    "Let's first import some packages:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9ed8aa30",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:01.238956Z",
     "iopub.status.busy": "2026-07-04T20:25:01.238718Z",
     "iopub.status.idle": "2026-07-04T20:25:03.389470Z",
     "shell.execute_reply": "2026-07-04T20:25:03.388354Z"
    }
   },
   "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": "3444bdeb",
   "metadata": {},
   "source": [
    "## Instantiate model class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f738e902",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.391420Z",
     "iopub.status.busy": "2026-07-04T20:25:03.391116Z",
     "iopub.status.idle": "2026-07-04T20:25:03.394180Z",
     "shell.execute_reply": "2026-07-04T20:25:03.393837Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "class Bohlin(LinearReferenceClock):\n",
      "    def postprocess(self, x):\n",
      "        \"\"\"Model returns gestational age in days; convert to weeks.\"\"\"\n",
      "        return x / 7.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def print_entire_class(cls):\n",
    "    source = inspect.getsource(cls)\n",
    "    print(source)\n",
    "\n",
    "print_entire_class(pya.models.Bohlin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ec2cc137",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.395464Z",
     "iopub.status.busy": "2026-07-04T20:25:03.395383Z",
     "iopub.status.idle": "2026-07-04T20:25:03.397276Z",
     "shell.execute_reply": "2026-07-04T20:25:03.396806Z"
    }
   },
   "outputs": [],
   "source": [
    "model = pya.models.Bohlin()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab50c624",
   "metadata": {},
   "source": [
    "## Define clock metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "adac7769",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.398601Z",
     "iopub.status.busy": "2026-07-04T20:25:03.398503Z",
     "iopub.status.idle": "2026-07-04T20:25:03.400886Z",
     "shell.execute_reply": "2026-07-04T20:25:03.400423Z"
    }
   },
   "outputs": [],
   "source": [
    "model.metadata[\"clock_name\"] = \"bohlin\"\n",
    "model.metadata[\"data_type\"] = \"DNA methylation\"  # Paper: The predictor uses cord-blood DNA methylation beta values.\n",
    "model.metadata[\"species\"] = \"Homo sapiens\"  # Paper: The study analyzed newborns from the human MoBa cohort.\n",
    "model.metadata[\"year\"] = 2016\n",
    "model.metadata[\"approved_by_author\"] = \"⌛\"\n",
    "model.metadata[\"citation\"] = \"Bohlin, J., Håberg, S. E., Magnus, P. et al. Prediction of gestational age based on genome-wide differentially methylated regions. Genome Biology 17, 207 (2016).\"\n",
    "model.metadata[\"doi\"] = \"https://doi.org/10.1186/s13059-016-1063-4\"\n",
    "model.metadata[\"notes\"] = \"Official minimum-lambda variant of the Bohlin gestational-age LASSO: pyaging implements the 251-CpG lambda.min model and converts its day-scale output to weeks. The paper/package default one-standard-error variant uses 96 CpGs and has nearly identical predictive performance.\"\n",
    "model.metadata[\"research_only\"] = None\n",
    "model.metadata[\"tissue\"] = [\"cord blood\"]  # Paper: Training used Illumina 450K arrays based on newborn cord-blood DNA.\n",
    "model.metadata[\"predicts\"] = [\"gestational age\"]  # Paper: The model estimates gestational age after birth from cord-blood methylation.\n",
    "model.metadata[\"training_target\"] = [\"gestational age\"]  # Paper: The primary, better-performing model was trained against ultrasound gestational age.\n",
    "model.metadata[\"unit\"] = [\"weeks\"]  # Paper: The coefficient formula is in days and the pyaging implementation divides the result by seven.\n",
    "model.metadata[\"model_type\"] = \"LASSO regression\"  # Paper: Gestational-age prediction used glmnet LASSO regression.\n",
    "model.metadata[\"platform\"] = [\"Illumina 450K\"]  # Paper: MoBa training methylomes were measured on the Illumina HumanMethylation450 platform.\n",
    "model.metadata[\"population\"] = \"newborns\"  # Paper: The model was trained in 1,068 MoBa newborns and tested in 685 additional MoBa newborns.\n",
    "model.metadata[\"journal\"] = \"Genome Biology\"\n",
    "model.metadata[\"last_author\"] = \"Wenche Nystad\"\n",
    "model.metadata[\"n_features\"] = 251\n",
    "model.metadata[\"citations\"] = 237\n",
    "model.metadata[\"citations_date\"] = \"2026-07-05\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "830c4189",
   "metadata": {},
   "source": [
    "## Download clock dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "291f98b7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.402146Z",
     "iopub.status.busy": "2026-07-04T20:25:03.402068Z",
     "iopub.status.idle": "2026-07-04T20:25:03.684827Z",
     "shell.execute_reply": "2026-07-04T20:25:03.683624Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os.system(f\"curl -sL -o coefficients.csv https://raw.githubusercontent.com/Duzhaozhen/OmniAge/c10fbe8cb92957520fbff1d55ae1def0691252e5/OmniAgePy/src/omniage/data/Bohlin_GA.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19851305",
   "metadata": {},
   "source": [
    "## Load features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "70b59676",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.688299Z",
     "iopub.status.busy": "2026-07-04T20:25:03.688026Z",
     "iopub.status.idle": "2026-07-04T20:25:03.697422Z",
     "shell.execute_reply": "2026-07-04T20:25:03.696744Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('coefficients.csv')\n",
    "mask = df['probe'].astype(str).str.lower().isin(['intercept', '(intercept)'])\n",
    "intercept_value = float(df.loc[mask, 'coef'].iloc[0]) if mask.any() else 0.0\n",
    "coef_df = df.loc[~mask].reset_index(drop=True)\n",
    "model.features = coef_df['probe'].tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "117da281",
   "metadata": {},
   "source": [
    "## Load weights into base model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ef29494e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.699299Z",
     "iopub.status.busy": "2026-07-04T20:25:03.699146Z",
     "iopub.status.idle": "2026-07-04T20:25:03.701747Z",
     "shell.execute_reply": "2026-07-04T20:25:03.701259Z"
    }
   },
   "outputs": [],
   "source": [
    "weights = torch.tensor(coef_df['coef'].tolist()).unsqueeze(0).float()\n",
    "intercept = torch.tensor([intercept_value]).float()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "39896fe7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.703008Z",
     "iopub.status.busy": "2026-07-04T20:25:03.702920Z",
     "iopub.status.idle": "2026-07-04T20:25:03.704983Z",
     "shell.execute_reply": "2026-07-04T20:25:03.704640Z"
    }
   },
   "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": "1bededf4",
   "metadata": {},
   "source": [
    "## Load reference values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "27476e4e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.706270Z",
     "iopub.status.busy": "2026-07-04T20:25:03.706195Z",
     "iopub.status.idle": "2026-07-04T20:25:03.707735Z",
     "shell.execute_reply": "2026-07-04T20:25:03.707342Z"
    }
   },
   "outputs": [],
   "source": [
    "model.reference_values = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bc67e24",
   "metadata": {},
   "source": [
    "## Load preprocess and postprocess objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a967bfac",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.708890Z",
     "iopub.status.busy": "2026-07-04T20:25:03.708821Z",
     "iopub.status.idle": "2026-07-04T20:25:03.710357Z",
     "shell.execute_reply": "2026-07-04T20:25:03.710035Z"
    }
   },
   "outputs": [],
   "source": [
    "model.preprocess_name = None\n",
    "model.preprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "466c93e6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.711475Z",
     "iopub.status.busy": "2026-07-04T20:25:03.711405Z",
     "iopub.status.idle": "2026-07-04T20:25:03.713066Z",
     "shell.execute_reply": "2026-07-04T20:25:03.712607Z"
    }
   },
   "outputs": [],
   "source": [
    "model.postprocess_name = 'days_to_weeks'\n",
    "model.postprocess_dependencies = None"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fce69e38",
   "metadata": {},
   "source": [
    "## Check all clock parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "166a9e08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.714484Z",
     "iopub.status.busy": "2026-07-04T20:25:03.714389Z",
     "iopub.status.idle": "2026-07-04T20:25:03.719171Z",
     "shell.execute_reply": "2026-07-04T20:25:03.718310Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Attributes:\n",
      "\n",
      "training: True\n",
      "metadata: {'approved_by_author': '⌛',\n",
      " 'citation': 'Bohlin, Jon, et al. \"Prediction of gestational age based on '\n",
      "             'genome-wide differentially methylated regions.\" Genome Biology '\n",
      "             '17.1 (2016): 207.',\n",
      " 'clock_name': 'bohlin',\n",
      " 'data_type': 'methylation',\n",
      " 'doi': 'https://doi.org/10.1186/s13059-016-1063-4',\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: 'days_to_weeks'\n",
      "postprocess_dependencies: None\n",
      "features: ['cg00099441', 'cg00117869', 'cg00303541', 'cg00602416', 'cg00711496', 'cg00898111', 'cg01091356', 'cg01139861', 'cg01190109', 'cg01281797', 'cg01359999', 'cg01382072', 'cg01470456', 'cg01505275', 'cg01635555', 'cg01697487', 'cg01749742', 'cg01833485', 'cg01847441', 'cg02324006', 'cg02358664', 'cg02378208', 'cg02405476', 'cg02567958', 'cg02576394', 'cg02642822', 'cg02779037', 'cg02853322', 'cg02985694', 'cg03098721']... [Total elements: 251]\n",
      "base_model_features: None\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Structure:\n",
      "\n",
      "base_model: LinearModel(\n",
      "  (linear): Linear(in_features=251, out_features=1, bias=True)\n",
      ")\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "Model Parameters and Weights:\n",
      "\n",
      "base_model.linear.weight: [1.679152488708496, -2.247448444366455, -8.844316482543945, 3.3127026557922363, -1.2231009006500244, 0.2817184627056122, 0.6226192712783813, 1.777445673942566, 9.944578170776367, 25.59168243408203, 4.947725772857666, 3.4168262481689453, -0.4994896650314331, -2.2033567428588867, 5.669358730316162, 1.2348579168319702, 0.014705928973853588, 3.155426263809204, 2.1350390911102295, -12.360982894897461, 1.5933818817138672, -2.458742618560791, 11.212024688720703, -16.415714263916016, -4.602630138397217, -3.348295211791992, -2.3952744007110596, 7.600930213928223, -4.37376594543457, 8.569387435913086]... [Tensor of shape torch.Size([1, 251])]\n",
      "base_model.linear.bias: tensor([277.2421])\n",
      "\n",
      "%==================================== Model Details ====================================%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pya.utils.print_model_details(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d5ab505",
   "metadata": {},
   "source": [
    "## Basic test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "259e4b13",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.722114Z",
     "iopub.status.busy": "2026-07-04T20:25:03.721981Z",
     "iopub.status.idle": "2026-07-04T20:25:03.726325Z",
     "shell.execute_reply": "2026-07-04T20:25:03.725928Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[67.3303],\n",
       "        [36.4561],\n",
       "        [18.0902],\n",
       "        [37.1463],\n",
       "        [50.0283],\n",
       "        [48.1619],\n",
       "        [16.2032],\n",
       "        [38.6205],\n",
       "        [60.7712],\n",
       "        [54.3381]], dtype=torch.float64, grad_fn=<DivBackward0>)"
      ]
     },
     "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": "aee7f34f",
   "metadata": {},
   "source": [
    "## Save torch model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "64ca20a7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.727644Z",
     "iopub.status.busy": "2026-07-04T20:25:03.727567Z",
     "iopub.status.idle": "2026-07-04T20:25:03.731485Z",
     "shell.execute_reply": "2026-07-04T20:25:03.730983Z"
    }
   },
   "outputs": [],
   "source": [
    "torch.save(model, f\"../weights/{model.metadata['clock_name']}.pt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "695ec299",
   "metadata": {},
   "source": [
    "## Clear directory\n",
    "<a id=\"10\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c50bcfdf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-04T20:25:03.733306Z",
     "iopub.status.busy": "2026-07-04T20:25:03.733156Z",
     "iopub.status.idle": "2026-07-04T20:25:03.737057Z",
     "shell.execute_reply": "2026-07-04T20:25:03.736668Z"
    }
   },
   "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
}