{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e65aaa5f",
   "metadata": {},
   "source": [
    "# Bayesian hierarchical calibration models\n",
    "While some calibrations can be established _once and for all_, there exists also the opposite extreme where (some) calibration parameters must be estimated alongside the process model.\n",
    "\n",
    "For simplicity let's assume the following scenario:\n",
    "\n",
    "* A linear relationship between independent/dependent variable in the measurement system.\n",
    "* A batch-effect on the intercept parameter of the calibration.\n",
    "* A \"production process\" that is a 1st order chemical rection. This can be described by $P(t) = S_0 \\cdot (1 - e^{-t \\cdot k})$.\n",
    "* One \"production process\" experiment was performed, with samples taken at 5 time points.\n",
    "* The product concentrations were measured in three batches, each having a different calibration intercept."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "41474238",
   "metadata": {},
   "outputs": [],
   "source": [
    "import arviz\n",
    "import numpy\n",
    "import calibr8\n",
    "import pymc as pm\n",
    "from matplotlib import pyplot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23f5463e",
   "metadata": {},
   "source": [
    "## Create the models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "adfea32b",
   "metadata": {},
   "outputs": [],
   "source": [
    "class LinearCalibration(calibr8.ContinuousUnivariateModel, calibr8.NormalNoise):\n",
    "    def __init__(self):\n",
    "        super().__init__(\n",
    "            independent_key=\"P\",\n",
    "            dependent_key=\"readout\",\n",
    "            theta_names=[\"intercept\", \"slope\", \"sigma\"],\n",
    "        )\n",
    "\n",
    "    def predict_dependent(self, x, *, theta=None):\n",
    "        if theta is None:\n",
    "            theta = self.theta_fitted\n",
    "        intercept, slope, sd = theta[0], theta[1], theta[2]\n",
    "        return calibr8.polynomial(x, [intercept, slope]), sd\n",
    "\n",
    "\n",
    "def predict_P(S0, k, time):\n",
    "    return S0 * (1 - numpy.exp(-time * k))\n",
    "\n",
    "\n",
    "# Create the raw calibration model. We will use it to generate data.\n",
    "cmodel = LinearCalibration()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8c3ef52",
   "metadata": {},
   "source": [
    "## Simulate a dataset\n",
    "For the purpose of this example, we're generating "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0ebd71fc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'batch #1': array([0.57288426, 1.009389  , 1.22126874, 1.5490391 ]),\n",
       " 'batch #2': array([0.85039   , 1.21933846, 1.73710129, 1.79041372]),\n",
       " 'batch #3': array([0.64525864, 1.04893182, 1.43234941, 1.67851469])}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "time = numpy.array([5, 15, 30, 60])  # seconds\n",
    "true_S0 = 5.0  # mM\n",
    "true_k = 0.05  # 1/s\n",
    "true_intercepts = [0.15, 0.45, 0.28]  # arbitrary units [a.u.]\n",
    "true_slope = 0.3  # a.u./mM\n",
    "true_sigma = 0.08  # a.u.\n",
    "rng = numpy.random.RandomState(20211118)\n",
    "\n",
    "dataset = {}\n",
    "for ib, intercept in enumerate(true_intercepts):\n",
    "    batch_id = f\"batch #{ib + 1}\"\n",
    "    # Simulate product concentrations\n",
    "    P = predict_P(true_S0, true_k, time)\n",
    "\n",
    "    # Simulate measurement readouts\n",
    "    mu, sd = cmodel.predict_dependent(P, theta=[intercept, true_slope, true_sigma])\n",
    "    dataset[batch_id] = cmodel.scipy_dist.rvs(\n",
    "        **cmodel.to_scipy(mu, sd),\n",
    "        random_state=rng,\n",
    "    )\n",
    "\n",
    "dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e8187c8",
   "metadata": {},
   "source": [
    "When visualizing the data, we can see a clear difference between the three batches."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9f2b37f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = pyplot.subplots(dpi=100)\n",
    "\n",
    "for batch_id, y_obs in dataset.items():\n",
    "    ax.scatter(time, y_obs, label=batch_id)\n",
    "ax.legend(loc=\"lower right\")\n",
    "ax.set(\n",
    "    xlim=(0, None),\n",
    "    ylim=(0, None),\n",
    "    xlabel=\"time / s\",\n",
    "    ylabel=\"readout / a.u.\",\n",
    "    title=r\"experimental data\\n$P_0=0, S_0=5\\ \\mathrm{mM}$\",\n",
    ")\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b513352",
   "metadata": {},
   "source": [
    "## Building the Bayesian model\n",
    "For analyzing our data, we should think about how it was generated.\n",
    "For every bit of information about the experimental setup we have to specify a _prior probability_ distribution to be used in the model.\n",
    "\n",
    "We started with an initial product concentration of exactly zero ($P_0=0\\ mM$).\n",
    "\n",
    "The reaction was started with 5 mM of substrate, but it was hard to handle, so we might have gotten the concentration wrong by ~5 %.\n",
    "\n",
    "$$ S_0 \\sim \\mathrm{LogNormal}(log(5), 0.05) $$\n",
    "\n",
    "The chemical reaction is a 1st order reaction.\n",
    "\n",
    "$$ P(t) = S_0 \\cdot (1 - e^{-t \\cdot k}) $$\n",
    "\n",
    "From literature we expect a rate coefficient of around 0.1/s. It can only be positive.\n",
    "\n",
    "$$ k \\sim \\mathrm{HalfNormal}(0.1) $$\n",
    "\n",
    "In this example we don't have a separate calibration dataset, or separately fitted calibration model.\n",
    "But from previous studies we have some prior information to fill the gaps in the final model:\n",
    "\n",
    "We expect an intercept around 0.3 ± 0.2 a.u.\n",
    "\n",
    "$$\\mathrm{intercept} \\sim \\mathcal{N}(0.3, 0.2)$$\n",
    "\n",
    "But there's a batch effect, so the actual intercepts may vary 0.1 a.u., but they're centered around the intercept we would expect on average.\n",
    "With this, our \"intercept\" variable is a so-called _hyperprior_ or _group mean prior_ for the batch-wise intercept.\n",
    "This makes the model a _hierarchical_ Bayesian model.\n",
    "\n",
    "$$\\mathrm{intercept_{batch}} \\sim \\mathcal{N}(\\mathrm{intercept}, 0.1)$$\n",
    "\n",
    "The slope should be around 0.5 a.u./mM.\n",
    "\n",
    "$$\\mathrm{slope} \\sim \\mathcal{N}(0.5, 0.5)$$\n",
    "\n",
    "The standard deviation of the observations should be around 0.1 a.u.\n",
    "\n",
    "$$ \\mathrm{sigma} \\sim \\mathrm{LogNormal}(log(0.1), 0.1) $$\n",
    "\n",
    "Note that we could also have obtained a separate calibration dataset and left only the intercept parameter to be estimated from our dataset of interest.\n",
    "\n",
    "What follows now is the PyMC model definition.\n",
    "It reads very much like the mathematical notation from above and can be visualized as a so-called _plate model_ showing how the variables are connected and which shapes they have."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4c14db96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 14.1.2 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"516pt\" height=\"475pt\"\n",
       " viewBox=\"0.00 0.00 516.00 475.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 470.54)\">\n",
       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-470.54 512,-470.54 512,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\">\n",
       "<title>clustertimepoint (4)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M20,-243.23C20,-243.23 102,-243.23 102,-243.23 108,-243.23 114,-249.23 114,-255.23 114,-255.23 114,-446.54 114,-446.54 114,-452.54 108,-458.54 102,-458.54 102,-458.54 20,-458.54 20,-458.54 14,-458.54 8,-452.54 8,-446.54 8,-446.54 8,-255.23 8,-255.23 8,-249.23 14,-243.23 20,-243.23\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"70\" y=\"-250.43\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">timepoint (4)</text>\n",
       "</g>\n",
       "<g id=\"clust2\" class=\"cluster\">\n",
       "<title>clusterbatch (3) x timepoint (4)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M172,-8C172,-8 299,-8 299,-8 305,-8 311,-14 311,-20 311,-20 311,-211.32 311,-211.32 311,-217.32 305,-223.32 299,-223.32 299,-223.32 172,-223.32 172,-223.32 166,-223.32 160,-217.32 160,-211.32 160,-211.32 160,-20 160,-20 160,-14 166,-8 172,-8\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235.5\" y=\"-15.2\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">batch (3) x timepoint (4)</text>\n",
       "</g>\n",
       "<g id=\"clust3\" class=\"cluster\">\n",
       "<title>clusterbatch (3)</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M356,-231.32C356,-231.32 488,-231.32 488,-231.32 494,-231.32 500,-237.32 500,-243.32 500,-243.32 500,-341.13 500,-341.13 500,-347.13 494,-353.13 488,-353.13 488,-353.13 356,-353.13 356,-353.13 350,-353.13 344,-347.13 344,-341.13 344,-341.13 344,-243.32 344,-243.32 344,-237.32 350,-231.32 356,-231.32\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"467.62\" y=\"-238.52\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">batch (3)</text>\n",
       "</g>\n",
       "<!-- time -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>time</title>\n",
       "<path fill=\"lightgrey\" stroke=\"black\" d=\"M76,-450.54C76,-450.54 46,-450.54 46,-450.54 40,-450.54 34,-444.54 34,-438.54 34,-438.54 34,-405.04 34,-405.04 34,-399.04 40,-393.04 46,-393.04 46,-393.04 76,-393.04 76,-393.04 82,-393.04 88,-399.04 88,-405.04 88,-405.04 88,-438.54 88,-438.54 88,-444.54 82,-450.54 76,-450.54\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-433.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">time</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-416.74\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-400.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Data</text>\n",
       "</g>\n",
       "<!-- P -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>P</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"106.12,-333.23 15.88,-333.23 15.88,-275.73 106.12,-275.73 106.12,-333.23\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-315.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">P</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-299.43\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"61\" y=\"-282.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Deterministic</text>\n",
       "</g>\n",
       "<!-- time&#45;&gt;P -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>time&#45;&gt;P</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M61,-392.58C61,-378.24 61,-360.57 61,-344.9\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"64.5,-344.91 61,-334.91 57.5,-344.91 64.5,-344.91\"/>\n",
       "</g>\n",
       "<!-- likelihood -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>likelihood</title>\n",
       "<ellipse fill=\"lightgrey\" stroke=\"black\" cx=\"235\" cy=\"-174.66\" rx=\"50.03\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-186.11\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">likelihood</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-169.61\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-153.11\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- P&#45;&gt;likelihood -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>P&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M77.66,-275.48C87.41,-260.89 100.79,-243.58 116,-231.32 134.46,-216.44 157.54,-204.47 178.46,-195.49\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"179.53,-198.84 187.44,-191.78 176.86,-192.37 179.53,-198.84\"/>\n",
       "</g>\n",
       "<!-- y_obs -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>y_obs</title>\n",
       "<path fill=\"lightgrey\" stroke=\"black\" d=\"M250,-98C250,-98 220,-98 220,-98 214,-98 208,-92 208,-86 208,-86 208,-52.5 208,-52.5 208,-46.5 214,-40.5 220,-40.5 220,-40.5 250,-40.5 250,-40.5 256,-40.5 262,-46.5 262,-52.5 262,-52.5 262,-86 262,-86 262,-92 256,-98 250,-98\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-80.7\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">y_obs</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-64.2\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"235\" y=\"-47.7\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Data</text>\n",
       "</g>\n",
       "<!-- likelihood&#45;&gt;y_obs -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>likelihood&#45;&gt;y_obs</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M235,-133.68C235,-125.83 235,-117.6 235,-109.76\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"238.5,-109.79 235,-99.79 231.5,-109.79 238.5,-109.79\"/>\n",
       "</g>\n",
       "<!-- sigma -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>sigma</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"179\" cy=\"-304.48\" rx=\"54.27\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"179\" y=\"-315.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">sigma</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"179\" y=\"-299.43\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"179\" y=\"-282.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Lognormal</text>\n",
       "</g>\n",
       "<!-- sigma&#45;&gt;likelihood -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>sigma&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M195.65,-265.47C201.41,-252.34 207.93,-237.44 213.99,-223.62\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"217.1,-225.25 217.9,-214.68 210.68,-222.44 217.1,-225.25\"/>\n",
       "</g>\n",
       "<!-- intercept -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>intercept</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"423\" cy=\"-421.79\" rx=\"44.72\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"423\" y=\"-433.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">intercept</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"423\" y=\"-416.74\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"423\" y=\"-400.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- intercept_batch -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>intercept_batch</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"422\" cy=\"-304.48\" rx=\"70.18\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"422\" y=\"-315.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">intercept_batch</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"422\" y=\"-299.43\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"422\" y=\"-282.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- intercept&#45;&gt;intercept_batch -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>intercept&#45;&gt;intercept_batch</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M422.65,-380.78C422.59,-373 422.51,-364.75 422.44,-356.66\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"425.95,-356.9 422.36,-346.93 418.95,-356.96 425.95,-356.9\"/>\n",
       "</g>\n",
       "<!-- slope -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>slope</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"292\" cy=\"-304.48\" rx=\"41.01\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"292\" y=\"-315.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">slope</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"292\" y=\"-299.43\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"292\" y=\"-282.93\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Normal</text>\n",
       "</g>\n",
       "<!-- slope&#45;&gt;likelihood -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>slope&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.67,-266.85C269.63,-253.31 262.7,-237.77 256.29,-223.39\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"259.62,-222.27 252.35,-214.56 253.23,-225.12 259.62,-222.27\"/>\n",
       "</g>\n",
       "<!-- S0 -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>S0</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"176\" cy=\"-421.79\" rx=\"54.27\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"176\" y=\"-433.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">S0</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"176\" y=\"-416.74\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"176\" y=\"-400.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Lognormal</text>\n",
       "</g>\n",
       "<!-- S0&#45;&gt;P -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>S0&#45;&gt;P</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M143.97,-388.67C129.39,-374.05 112.07,-356.69 97.04,-341.61\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"99.56,-339.19 90.02,-334.58 94.6,-344.13 99.56,-339.19\"/>\n",
       "</g>\n",
       "<!-- k -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>k</title>\n",
       "<ellipse fill=\"none\" stroke=\"black\" cx=\"304\" cy=\"-421.79\" rx=\"55.86\" ry=\"40.66\"/>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"304\" y=\"-433.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">k</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"304\" y=\"-416.74\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">~</text>\n",
       "<text xml:space=\"preserve\" text-anchor=\"middle\" x=\"304\" y=\"-400.24\" font-family=\"Times New Roman,serif\" font-size=\"14.00\">Halfnormal</text>\n",
       "</g>\n",
       "<!-- k&#45;&gt;P -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>k&#45;&gt;P</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M263.44,-393.31C255.57,-388.77 247.21,-384.47 239,-381.13 187.06,-360.04 166.01,-378.47 116,-353.13 109.34,-349.76 102.82,-345.4 96.73,-340.69\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"99.27,-338.25 89.32,-334.58 94.81,-343.65 99.27,-338.25\"/>\n",
       "</g>\n",
       "<!-- intercept_batch&#45;&gt;likelihood -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>intercept_batch&#45;&gt;likelihood</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M387.39,-268.73C373.87,-256.13 357.83,-242.3 342,-231.32 325.46,-219.84 306.26,-209.13 288.72,-200.24\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"290.49,-197.21 279.97,-195.89 287.37,-203.48 290.49,-197.21\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x234c9f64440>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This model code uses the \"coords\" feature of PyMC models to track, for example, the names of batches.\n",
    "with pm.Model(\n",
    "    coords={\n",
    "        \"batch\": list(dataset.keys()),\n",
    "        \"timepoint\": list(numpy.arange(len(time))),\n",
    "    }\n",
    ") as pmodel:\n",
    "    t = pm.Data(\"time\", time, dims=\"timepoint\")\n",
    "    y_obs = pm.Data(\n",
    "        \"y_obs\", numpy.array(list(dataset.values())), dims=(\"batch\", \"timepoint\")\n",
    "    )\n",
    "\n",
    "    # Process model\n",
    "    k = pm.HalfNormal(\"k\", 0.1)\n",
    "    S0 = pm.Lognormal(\"S0\", mu=numpy.log(5.0), sigma=0.05)\n",
    "    P = pm.Deterministic(\"P\", predict_P(S0, k, t), dims=\"timepoint\")\n",
    "\n",
    "    # Calibration model\n",
    "    intercept = pm.Normal(\"intercept\", mu=0.3, sigma=0.2)\n",
    "    intercept_batch = pm.Normal(\n",
    "        \"intercept_batch\", mu=intercept, sigma=0.1, dims=\"batch\"\n",
    "    )\n",
    "    slope = pm.Normal(\"slope\", mu=0.5, sigma=0.5)\n",
    "    sigma = pm.Lognormal(\"sigma\", mu=numpy.log(0.1), sigma=0.1)\n",
    "\n",
    "    # Observing the product concentrations with the calibration model\n",
    "    cmodel.loglikelihood(\n",
    "        name=\"likelihood\",\n",
    "        y=y_obs,\n",
    "        # This 👇 broadcasts the product concentration over all batches.\n",
    "        x=P[None, :],\n",
    "        dims=(\"batch\", \"timepoint\"),\n",
    "        # Pass priors for the calibration model parameters:\n",
    "        theta=[intercept_batch[:, None], slope, sigma],\n",
    "        # This slicing here 👆 makes the intercept into a column-vector,\n",
    "        # such that it's shape broadcasting aligns with P[None, :] from above.\n",
    "    )\n",
    "\n",
    "pm.model_to_graphviz(pmodel)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8943759",
   "metadata": {},
   "source": [
    "We can now run MCMC, specifically PyMC's implementation of HMC-NUTS, to obtain samples from the posterior probability distribution of our model variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "744ad05a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\osthege\\AppData\\Local\\miniforge3\\envs\\dibecs_9.2.1\\Lib\\site-packages\\threadpoolctl.py:1226: RuntimeWarning: \n",
      "Found Intel OpenMP ('libiomp') and LLVM OpenMP ('libomp') loaded at\n",
      "the same time. Both libraries are known to be incompatible and this\n",
      "can cause random crashes or deadlocks on Linux when loaded in the\n",
      "same Python program.\n",
      "Using threadpoolctl may cause crashes or deadlocks. For more\n",
      "information and possible workarounds, please see\n",
      "    https://github.com/joblib/threadpoolctl/blob/master/multiple_openmp.md\n",
      "\n",
      "  warnings.warn(msg, RuntimeWarning)\n",
      "Initializing NUTS using jitter+adapt_diag...\n",
      "Multiprocess sampling (4 chains in 4 jobs)\n",
      "NUTS: [k, S0, intercept, intercept_batch, slope, sigma]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8f4a7718f99b45f19e960883eaeafcb0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Output()"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"></pre>\n"
      ],
      "text/plain": []
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sampling 4 chains for 1_000 tune and 1_000 draw iterations (4_000 + 4_000 draws total) took 26 seconds.\n"
     ]
    }
   ],
   "source": [
    "with pmodel:\n",
    "    idata = pm.sample()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "029254f4",
   "metadata": {},
   "source": [
    "## Visualization of the results\n",
    "Using the [ArviZ](https://arviz-devs.github.io/arviz/examples/index.html) library we can quickly create pretty plots of our model posteriors.\n",
    "The following one shows the highest density intervals (HDI) where we believe our model variables lie with a 99 % probability."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "72c4884b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x200 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "var_names = [\"intercept_batch\", \"intercept\", \"slope\", \"sigma\", \"k\", \"S0\"]\n",
    "units = [\"a.u\", \"a.u.\", \"a.u./mM\", \"a.u.\", \"mM/s\", \"mM\"]\n",
    "fig, axs = pyplot.subplots(\n",
    "    ncols=len(var_names), figsize=(12, 12 / len(var_names)), dpi=100\n",
    ")\n",
    "for ax, name, unit in zip(axs, var_names, units):\n",
    "    arviz.plot_forest(\n",
    "        idata, ax=ax, var_names=[name], kind=\"ridgeplot\", combined=True, hdi_prob=0.99\n",
    "    )\n",
    "    # Remove redundant texts from the automatic yticklabels\n",
    "    ax.set_yticklabels(\n",
    "        [\n",
    "            ytext._text.replace(name, \"\").replace(\":\", \"\")\n",
    "            for ytext in ax.get_yticklabels()\n",
    "        ]\n",
    "    )\n",
    "    # instead put the name and axis\n",
    "    ax.set_xlabel(f\"{name} / {unit}\")\n",
    "fig.tight_layout()\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c91a031",
   "metadata": {},
   "source": [
    "We can also plot the product concentration over time.\n",
    "The posterior samples in our model are only at 5 time points, so to get a smoother plot let's re-predict product concentrations from the $S_0$ and $k$ posterior samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f35d46e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Combine posterior samples from all MCMC chains\n",
    "pst = idata.posterior.stack(sample=(\"chain\", \"draw\"))\n",
    "\n",
    "# Predict product concentrations with higher time resolution\n",
    "time_dense = numpy.linspace(0, 60, 100)\n",
    "P_dense = predict_P(S0=pst.S0.values[:, None], k=pst.k.values[:, None], time=time_dense)\n",
    "\n",
    "fig, ax = pyplot.subplots(dpi=100)\n",
    "\n",
    "pm.gp.util.plot_gp_dist(\n",
    "    ax=ax,\n",
    "    x=time_dense,\n",
    "    samples=P_dense,\n",
    "    palette=pyplot.cm.Blues,\n",
    ")\n",
    "ax.set(\n",
    "    title=\"Posterior probability of product concentration over time\\n(lines are individual samples)\",\n",
    "    ylabel=\"product concentration / mM\",\n",
    "    xlabel=\"time / s\",\n",
    ")\n",
    "\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab28d3b8",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "In this example we have seen how to analyze experimental data containing batch-effects.\n",
    "We built a Bayesian model with PyMC, levaraging the tensors-compatibility of the `calibr8.CalibrationModel.loglikelihood` to construct a hierarchical calibration model that captures batch-effects in product concentration measurement.\n",
    "\n",
    "As a result, we have not only quantified uncertainty in the rate coefficient of our chemical reaction, but also the uncertainties in product concentration over time, initial substrate concentration and all parameters of the calibration model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2aa29323",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Last updated: Wed, 15 Apr 2026\n",
      "\n",
      "Python implementation: CPython\n",
      "Python version       : 3.13.12\n",
      "IPython version      : 9.11.0\n",
      "\n",
      "arviz     : 0.23.4\n",
      "calibr8   : 7.2.0\n",
      "matplotlib: 3.10.8\n",
      "numpy     : 2.4.2\n",
      "platform  : 1.0.8\n",
      "pymc      : 5.28.1\n",
      "\n",
      "Watermark: 2.6.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -n -u -v -iv -w"
   ]
  }
 ],
 "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": "3.13.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}