{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Anomalous Difference Map  \n",
    "\n",
    "In [Basics](1_basics.ipynb), we loaded a room-temperature dataset that was collected at ~6550 eV of tetragonal HEWL. We then computed the $CC_{1/2}$ and $CC_{anom}$ for this data in [Merging Statistics](2_mergingstats.ipynb) and observed that there was significant anomalous signal. Let's now use that data to generate an anomalous difference map based on the anomalous scattering from the native sulfur atoms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import reciprocalspaceship as rs\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set_context(\"notebook\", font_scale=1.3)\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.10.3\n"
     ]
    }
   ],
   "source": [
    "print(rs.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>I(+)</th>\n",
       "      <th>SIGI(+)</th>\n",
       "      <th>I(-)</th>\n",
       "      <th>SIGI(-)</th>\n",
       "      <th>N(+)</th>\n",
       "      <th>N(-)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th>4</th>\n",
       "      <td>661.29987</td>\n",
       "      <td>21.953098</td>\n",
       "      <td>661.29987</td>\n",
       "      <td>21.953098</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3229.649</td>\n",
       "      <td>105.980934</td>\n",
       "      <td>3229.649</td>\n",
       "      <td>105.980934</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1361.8672</td>\n",
       "      <td>43.06085</td>\n",
       "      <td>1361.8672</td>\n",
       "      <td>43.06085</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>4124.393</td>\n",
       "      <td>196.89108</td>\n",
       "      <td>4124.393</td>\n",
       "      <td>196.89108</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <td>559.33685</td>\n",
       "      <td>8.6263</td>\n",
       "      <td>559.33685</td>\n",
       "      <td>8.6263</td>\n",
       "      <td>64</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            I(+)    SIGI(+)      I(-)    SIGI(-)  N(+)  N(-)\n",
       "H K L                                                       \n",
       "0 0 4  661.29987  21.953098 661.29987  21.953098    16    16\n",
       "    8   3229.649 105.980934  3229.649 105.980934    16    16\n",
       "    12 1361.8672   43.06085 1361.8672   43.06085    16    16\n",
       "    16  4124.393  196.89108  4124.393  196.89108     8     8\n",
       "1 0 1  559.33685     8.6263 559.33685     8.6263    64    64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load data and extract relevant columns\n",
    "refltable = rs.read_mtz(\"data/HEWL_SSAD_24IDC.mtz\")\n",
    "refltable = refltable[[\"I(+)\", \"SIGI(+)\", \"I(-)\", \"SIGI(-)\", \"N(+)\", \"N(-)\"]]\n",
    "refltable.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of reflections: 12542\n"
     ]
    }
   ],
   "source": [
    "print(f\"Number of reflections: {len(refltable)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "### Background\n",
    "\n",
    "Since this dataset was collected at a single wavelength, we can compute an anomalous difference map from the anomalous structure factor amplitudes, $|F_{A}|$, and their phase shifts, $\\alpha$, relative to the computed phases from an isomorphous structure of tetragonal HEWL. We will compute $|F_{A}|$ based on the following:\n",
    "\n",
    "\\begin{equation*}\n",
    "|F_{A}| \\propto \\Delta_{\\mathrm{ano}} = |F_{HKL}| - |F_{\\overline{HKL}}|\n",
    "\\end{equation*}\n",
    "\n",
    "We will then use a model refined from this data to obtain the phases, $\\phi_c$, which can be used to determine the phases for the anomalous contribution, $\\phi_A$, using the phase shift, $\\alpha$:\n",
    "\n",
    "\\begin{equation*}\n",
    "\\phi_A = \\phi_c - \\alpha\n",
    "\\end{equation*}\n",
    "\n",
    "Since this is a SAD experiment, we can assume $\\alpha$ is 90˚ when $\\Delta_{\\mathrm{ano}}$ is negative and 270˚ when it is positive. This formalism is based on [Thorn and Sheldrick, J Appl. Cryst. (2011)](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3246834/pdf/j-44-01285.pdf)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "### Computing Structure Factor Amplitudes\n",
    "\n",
    "The dataset being used is the direct output from running scaling and merging in [AIMLESS](http://www.ccp4.ac.uk/html/aimless.html). As a first processing step, we need to convert the observed merged intensities, $I(+)$ and $I(-)$, into observed structure factor amplitudes, $F(+)$ and $F(-)$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>I</th>\n",
       "      <th>SIGI</th>\n",
       "      <th>N</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">-45</th>\n",
       "      <th>-10</th>\n",
       "      <th>-1</th>\n",
       "      <td>6.185645</td>\n",
       "      <td>2.932488</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">-9</th>\n",
       "      <th>-2</th>\n",
       "      <td>27.028767</td>\n",
       "      <td>3.8457258</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-1</th>\n",
       "      <td>3.0018542</td>\n",
       "      <td>2.6649861</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">-8</th>\n",
       "      <th>-3</th>\n",
       "      <td>-0.9806365</td>\n",
       "      <td>2.7741797</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-2</th>\n",
       "      <td>12.085027</td>\n",
       "      <td>3.0270035</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    I      SIGI  N\n",
       "H   K   L                         \n",
       "-45 -10 -1   6.185645  2.932488  4\n",
       "    -9  -2  27.028767 3.8457258  4\n",
       "        -1  3.0018542 2.6649861  4\n",
       "    -8  -3 -0.9806365 2.7741797  4\n",
       "        -2  12.085027 3.0270035  4"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Stack intensities from 2-column anomalous to 1-column format\n",
    "stacked = refltable.stack_anomalous()\n",
    "stacked = stacked.loc[stacked[\"N\"] > 0]\n",
    "stacked.sort_index(inplace=True)\n",
    "stacked.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to get structure factor amplitudes, we must first account for mean intensities that are negative due to background subtraction. We will use a method based on the Bayesian approach first proposed by [French and Wilson](https://scripts.iucr.org/cgi-bin/paper?a15572) to ensure that all intensities are strictly positive. This method is implemented in `rs.algorithms.scale_merged_intensities()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaled = rs.algorithms.scale_merged_intensities(stacked, \"I\", \"SIGI\", \n",
    "                                                mean_intensity_method=\"anisotropic\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>I</th>\n",
       "      <th>SIGI</th>\n",
       "      <th>N</th>\n",
       "      <th>dHKL</th>\n",
       "      <th>CENTRIC</th>\n",
       "      <th>FW-I</th>\n",
       "      <th>FW-SIGI</th>\n",
       "      <th>FW-F</th>\n",
       "      <th>FW-SIGF</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">-45</th>\n",
       "      <th>-10</th>\n",
       "      <th>-1</th>\n",
       "      <td>6.185645</td>\n",
       "      <td>2.932488</td>\n",
       "      <td>4</td>\n",
       "      <td>1.7194302</td>\n",
       "      <td>False</td>\n",
       "      <td>6.1471047</td>\n",
       "      <td>2.7739754</td>\n",
       "      <td>2.4032679</td>\n",
       "      <td>0.6094333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">-9</th>\n",
       "      <th>-2</th>\n",
       "      <td>27.028767</td>\n",
       "      <td>3.8457258</td>\n",
       "      <td>4</td>\n",
       "      <td>1.7217722</td>\n",
       "      <td>False</td>\n",
       "      <td>26.716537</td>\n",
       "      <td>3.8457258</td>\n",
       "      <td>5.1551414</td>\n",
       "      <td>0.37557602</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-1</th>\n",
       "      <td>3.0018542</td>\n",
       "      <td>2.6649861</td>\n",
       "      <td>4</td>\n",
       "      <td>1.7271528</td>\n",
       "      <td>False</td>\n",
       "      <td>3.5474997</td>\n",
       "      <td>2.1485696</td>\n",
       "      <td>1.7800175</td>\n",
       "      <td>0.6156602</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">-8</th>\n",
       "      <th>-3</th>\n",
       "      <td>-0.9806365</td>\n",
       "      <td>2.7741797</td>\n",
       "      <td>4</td>\n",
       "      <td>1.7197413</td>\n",
       "      <td>False</td>\n",
       "      <td>1.8468517</td>\n",
       "      <td>1.4775624</td>\n",
       "      <td>1.2411455</td>\n",
       "      <td>0.5535428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-2</th>\n",
       "      <td>12.085027</td>\n",
       "      <td>3.0270035</td>\n",
       "      <td>4</td>\n",
       "      <td>1.7287054</td>\n",
       "      <td>False</td>\n",
       "      <td>11.893286</td>\n",
       "      <td>3.02595</td>\n",
       "      <td>3.4186602</td>\n",
       "      <td>0.45392564</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    I      SIGI  N      dHKL  CENTRIC      FW-I   FW-SIGI  \\\n",
       "H   K   L                                                                   \n",
       "-45 -10 -1   6.185645  2.932488  4 1.7194302    False 6.1471047 2.7739754   \n",
       "    -9  -2  27.028767 3.8457258  4 1.7217722    False 26.716537 3.8457258   \n",
       "        -1  3.0018542 2.6649861  4 1.7271528    False 3.5474997 2.1485696   \n",
       "    -8  -3 -0.9806365 2.7741797  4 1.7197413    False 1.8468517 1.4775624   \n",
       "        -2  12.085027 3.0270035  4 1.7287054    False 11.893286   3.02595   \n",
       "\n",
       "                FW-F    FW-SIGF  \n",
       "H   K   L                        \n",
       "-45 -10 -1 2.4032679  0.6094333  \n",
       "    -9  -2 5.1551414 0.37557602  \n",
       "        -1 1.7800175  0.6156602  \n",
       "    -8  -3 1.2411455  0.5535428  \n",
       "        -2 3.4186602 0.45392564  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "French-Wilson scaling leaves large intensities relatively unchanged, but rescales negative and small intensities to be positive by imposing Wilson's distribution as a prior:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = np.linspace(-3, 10000, 2)\n",
    "plt.figure(figsize=(6, 6))\n",
    "ax = plt.gca()\n",
    "plt.plot(x, x, '--r', scalex=False, scaley=True)\n",
    "plt.plot(scaled[\"I\"].to_numpy(), scaled[\"FW-I\"].to_numpy(), \"k.\", alpha=0.25)\n",
    "plt.plot(plt.xlim(), plt.xlim(), '--r', scalex=False, scaley=True)\n",
    "plt.xlabel(\"Merged Intensity (AIMLESS)\")\n",
    "plt.ylabel(\"Posterior Intensity\")\n",
    "plt.xscale(\"log\")\n",
    "plt.yscale(\"log\")\n",
    "plt.xlim(right=1e4)\n",
    "plt.ylim(top=1e4)\n",
    "\n",
    "# Inset\n",
    "axins = ax.inset_axes([0.1, 0.5, 0.43, 0.43])\n",
    "axins.plot(scaled[\"I\"].to_numpy(), scaled[\"FW-I\"].to_numpy(), \"k.\", alpha=0.25)\n",
    "axins.plot(x, x, '--r')\n",
    "axins.set_xlim(-3, 10)\n",
    "axins.set_ylim(-3, 10)\n",
    "axins.grid(linestyle=\"--\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove extra columns\n",
    "scaled = scaled[[\"FW-F\", \"FW-SIGF\", \"N\"]]\n",
    "\n",
    "# \"Unstack\" anomalous data from one-column to two-column format\n",
    "anom = scaled.unstack_anomalous([\"FW-F\", \"FW-SIGF\", \"N\"]).dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>FW-F(+)</th>\n",
       "      <th>FW-SIGF(+)</th>\n",
       "      <th>N(+)</th>\n",
       "      <th>FW-F(-)</th>\n",
       "      <th>FW-SIGF(-)</th>\n",
       "      <th>N(-)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th>4</th>\n",
       "      <td>25.7036</td>\n",
       "      <td>0.4273095</td>\n",
       "      <td>16</td>\n",
       "      <td>25.7036</td>\n",
       "      <td>0.4273095</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>56.794975</td>\n",
       "      <td>0.93358195</td>\n",
       "      <td>16</td>\n",
       "      <td>56.794975</td>\n",
       "      <td>0.93358195</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>36.885887</td>\n",
       "      <td>0.5840336</td>\n",
       "      <td>16</td>\n",
       "      <td>36.885887</td>\n",
       "      <td>0.5840336</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>64.02966</td>\n",
       "      <td>1.5395098</td>\n",
       "      <td>8</td>\n",
       "      <td>64.02966</td>\n",
       "      <td>1.5395098</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <td>23.647089</td>\n",
       "      <td>0.18242109</td>\n",
       "      <td>64</td>\n",
       "      <td>23.647089</td>\n",
       "      <td>0.18242109</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         FW-F(+)  FW-SIGF(+)  N(+)   FW-F(-)  FW-SIGF(-)  N(-)\n",
       "H K L                                                         \n",
       "0 0 4    25.7036   0.4273095    16   25.7036   0.4273095    16\n",
       "    8  56.794975  0.93358195    16 56.794975  0.93358195    16\n",
       "    12 36.885887   0.5840336    16 36.885887   0.5840336    16\n",
       "    16  64.02966   1.5395098     8  64.02966   1.5395098     8\n",
       "1 0 1  23.647089  0.18242109    64 23.647089  0.18242109    64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anom.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute differences\n",
    "dF    = np.abs(anom[\"FW-F(+)\"] - anom[\"FW-F(-)\"])\n",
    "sigDF = np.sqrt(anom[\"FW-SIGF(+)\"]**2 + anom[\"FW-SIGF(-)\"]**2)\n",
    "anom[\"ANOM\"] = rs.DataSeries(dF, dtype=\"SFAmplitude\")\n",
    "anom[\"SigANOM\"] = rs.DataSeries(sigDF, dtype=\"Stddev\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>FW-F(+)</th>\n",
       "      <th>FW-SIGF(+)</th>\n",
       "      <th>N(+)</th>\n",
       "      <th>FW-F(-)</th>\n",
       "      <th>FW-SIGF(-)</th>\n",
       "      <th>N(-)</th>\n",
       "      <th>ANOM</th>\n",
       "      <th>SigANOM</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>H</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th rowspan=\"4\" valign=\"top\">0</th>\n",
       "      <th>4</th>\n",
       "      <td>25.7036</td>\n",
       "      <td>0.4273095</td>\n",
       "      <td>16</td>\n",
       "      <td>25.7036</td>\n",
       "      <td>0.4273095</td>\n",
       "      <td>16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.60430694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>56.794975</td>\n",
       "      <td>0.93358195</td>\n",
       "      <td>16</td>\n",
       "      <td>56.794975</td>\n",
       "      <td>0.93358195</td>\n",
       "      <td>16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.3202842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>36.885887</td>\n",
       "      <td>0.5840336</td>\n",
       "      <td>16</td>\n",
       "      <td>36.885887</td>\n",
       "      <td>0.5840336</td>\n",
       "      <td>16</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.82594824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>64.02966</td>\n",
       "      <td>1.5395098</td>\n",
       "      <td>8</td>\n",
       "      <td>64.02966</td>\n",
       "      <td>1.5395098</td>\n",
       "      <td>8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.1771955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <td>23.647089</td>\n",
       "      <td>0.18242109</td>\n",
       "      <td>64</td>\n",
       "      <td>23.647089</td>\n",
       "      <td>0.18242109</td>\n",
       "      <td>64</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.25798237</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         FW-F(+)  FW-SIGF(+)  N(+)   FW-F(-)  FW-SIGF(-)  N(-)  ANOM  \\\n",
       "H K L                                                                  \n",
       "0 0 4    25.7036   0.4273095    16   25.7036   0.4273095    16   0.0   \n",
       "    8  56.794975  0.93358195    16 56.794975  0.93358195    16   0.0   \n",
       "    12 36.885887   0.5840336    16 36.885887   0.5840336    16   0.0   \n",
       "    16  64.02966   1.5395098     8  64.02966   1.5395098     8   0.0   \n",
       "1 0 1  23.647089  0.18242109    64 23.647089  0.18242109    64   0.0   \n",
       "\n",
       "          SigANOM  \n",
       "H K L              \n",
       "0 0 4  0.60430694  \n",
       "    8   1.3202842  \n",
       "    12 0.82594824  \n",
       "    16  2.1771955  \n",
       "1 0 1  0.25798237  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anom.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "### Phasing the Anomalous Difference Map\n",
    "\n",
    "Below, we will compute the necessary phase shifts to go from the phases of a $2 F_o - F_c$ map to the phases associated with the anomalous difference structure factors. Although this model was refined to this data in PHENIX, the phases from any isomorphous structure could have been used to obtain a reasonable map. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "ref = rs.read_mtz(\"data/HEWL_refined.mtz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find common HKL indices\n",
    "hkls = anom.index.intersection(ref.index).sort_values()\n",
    "anom = anom.loc[hkls]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As mentioned in [Background](4_anomalousmap.ipynb#Background), we can compute the anomalous phases as follows:\n",
    "\n",
    "\\begin{equation*}\n",
    "\\phi_A = \\phi_c - \\alpha\n",
    "\\end{equation*}\n",
    "\n",
    "where $\\alpha$ is 90˚ when $\\Delta_{\\mathrm{ano}}$ is negative and 270˚ when it is positive."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "alpha = 90 + 180*(anom[\"FW-F(+)\"] >= anom[\"FW-F(-)\"])\n",
    "anom[\"PHANOM\"] = ref.loc[hkls, \"PH2FOFCWT\"] + alpha\n",
    "anom[\"PHANOM\"] = rs.utils.canonicalize_phases(anom[\"PHANOM\"]).astype(\"Phase\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Viewing the Map  \n",
    "\n",
    "Since we have structure factor amplitudes, `anom[\"ANOM\"]`, and phases, `anom[\"PHANOM\"]`, for the anomalous structure factors, we can now view the anomalous difference map. This can be done by writing out an MTZ file, and loading it into COOT, PyMOL, or any other molecular graphics package. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "FW-F(+)       FriedelSFAmplitude\n",
       "FW-SIGF(+)       StddevFriedelSF\n",
       "N(+)                      MTZInt\n",
       "FW-F(-)       FriedelSFAmplitude\n",
       "FW-SIGF(-)       StddevFriedelSF\n",
       "N(-)                      MTZInt\n",
       "ANOM                 SFAmplitude\n",
       "SigANOM                   Stddev\n",
       "PHANOM                     Phase\n",
       "dtype: object"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "anom.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "anom.write_mtz(\"data/anomdiff.mtz\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the anomalous difference map on the refined structure, we can see positive difference density on all of the sulfurs in HEWL (shown below in purple):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<center>\n",
       "    <iframe src=\"https://rs-station.github.io/reciprocalspaceship/data/anomdiff/anomdiff.html#zoom=25\", width=600, height=600></iframe>\n",
       "    <br>Anomalous difference map from HEWL crystal overlayed with refined model (PDB: 7L84). Map is contoured at +5&sigma;.\n",
       "</center>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%html\n",
    "<center>\n",
    "    <iframe src=\"https://rs-station.github.io/reciprocalspaceship/data/anomdiff/anomdiff.html#zoom=25\", width=600, height=600></iframe>\n",
    "    <br>Anomalous difference map from HEWL crystal overlayed with refined model (PDB: 7L84). Map is contoured at +5&sigma;.\n",
    "</center>"
   ]
  }
 ],
 "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}