ScatterPlot_wing_wize_normalized.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.chdir('/Users/jessesnaust/Documents/python scripts/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## load data"
   ]
  },
  {
   "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>variants</th>\n",
       "      <th>normalized_mean</th>\n",
       "      <th>stdev</th>\n",
       "      <th>num_replicates</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4A</td>\n",
       "      <td>3.739984</td>\n",
       "      <td>0.278252</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C124S</td>\n",
       "      <td>-0.440364</td>\n",
       "      <td>-0.034346</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>D22E</td>\n",
       "      <td>-0.804356</td>\n",
       "      <td>-0.073384</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>P38H</td>\n",
       "      <td>-0.199249</td>\n",
       "      <td>-0.013897</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>L70V</td>\n",
       "      <td>0.039773</td>\n",
       "      <td>0.002589</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants  normalized_mean     stdev  num_replicates\n",
       "0       4A         3.739984  0.278252              10\n",
       "1    C124S        -0.440364 -0.034346              10\n",
       "2     D22E        -0.804356 -0.073384               9\n",
       "3     P38H        -0.199249 -0.013897               9\n",
       "4     L70V         0.039773  0.002589               8"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_wing = pd.read_csv('Fly stuff/wing_size_normalized.csv', index_col=False)\n",
    "df_wing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>variants</th>\n",
       "      <th>ClinVAR</th>\n",
       "      <th>ClinVar significance</th>\n",
       "      <th>ClinVar Condition</th>\n",
       "      <th>Cosmic</th>\n",
       "      <th>Cosmic incidence</th>\n",
       "      <th>Cosmic - primary tissue type</th>\n",
       "      <th>Cosmic -histology</th>\n",
       "      <th>gnomAD</th>\n",
       "      <th>gnomAD - Allele frequency</th>\n",
       "      <th>...</th>\n",
       "      <th>same amino acid to disease info (from ClinVar/COSMIC/gnomAD)</th>\n",
       "      <th>Additional comments</th>\n",
       "      <th>Wing Assay function calls</th>\n",
       "      <th>Post_expt_function calls</th>\n",
       "      <th>HEK pAKT/AKT data (Post et al)</th>\n",
       "      <th>Yeast (Post et al)</th>\n",
       "      <th>Yeast (Mighell et al)</th>\n",
       "      <th>function_calls</th>\n",
       "      <th>Additional comments on function calls</th>\n",
       "      <th>Eclosion Assay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>D268E</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Likely Benign</td>\n",
       "      <td>Hereditary cancer-predisposing syndrome</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3*</td>\n",
       "      <td>Thyroid</td>\n",
       "      <td>Carcinoma</td>\n",
       "      <td>Yes</td>\n",
       "      <td>137.0</td>\n",
       "      <td>...</td>\n",
       "      <td>COSMIC - D268Y 2*</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Like WT</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT</td>\n",
       "      <td>WT</td>\n",
       "      <td>0.225343</td>\n",
       "      <td>Benign</td>\n",
       "      <td>Additionally ClinVar says it is likely benign....</td>\n",
       "      <td>Partial LOF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>I203V</td>\n",
       "      <td>Yes</td>\n",
       "      <td>VUS</td>\n",
       "      <td>Hereditary cancer-predisposing syndrome|PTEN h...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1*</td>\n",
       "      <td>Thyroid</td>\n",
       "      <td>Carcinoma</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Like WT</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT</td>\n",
       "      <td>WT</td>\n",
       "      <td>0.449992</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT in Drosophila eclosion as well</td>\n",
       "      <td>WT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>I400V</td>\n",
       "      <td>Yes</td>\n",
       "      <td>VUS</td>\n",
       "      <td>PTEN hamartoma tumor syndrome</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Like WT</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT</td>\n",
       "      <td>WT</td>\n",
       "      <td>-0.338630</td>\n",
       "      <td>Benign</td>\n",
       "      <td>pLOF in Drosophila eclosion</td>\n",
       "      <td>Partial LOF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>N117S</td>\n",
       "      <td>Yes</td>\n",
       "      <td>VUS</td>\n",
       "      <td>Hereditary cancer-predisposing syndrome|not pr...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1*</td>\n",
       "      <td>Lung</td>\n",
       "      <td>Carcinoma</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>ClinVar - N117H - VUS N117H in gnomAD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Like WT</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT</td>\n",
       "      <td>WT</td>\n",
       "      <td>0.240901</td>\n",
       "      <td>Benign</td>\n",
       "      <td>GOF in Drosophila eclosion</td>\n",
       "      <td>GOF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>N262S</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>No</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Yes</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>COSMIC - N262I 1*</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Like WT</td>\n",
       "      <td>Benign</td>\n",
       "      <td>WT</td>\n",
       "      <td>WT</td>\n",
       "      <td>-0.208005</td>\n",
       "      <td>Benign</td>\n",
       "      <td>pLOF in Drosophila eclosion</td>\n",
       "      <td>Partial LOF</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants ClinVAR ClinVar significance  \\\n",
       "0    D268E     Yes        Likely Benign   \n",
       "1    I203V     Yes                  VUS   \n",
       "2    I400V     Yes                  VUS   \n",
       "3    N117S     Yes                  VUS   \n",
       "4    N262S      No                  NaN   \n",
       "\n",
       "                                   ClinVar Condition Cosmic Cosmic incidence  \\\n",
       "0            Hereditary cancer-predisposing syndrome    Yes               3*   \n",
       "1  Hereditary cancer-predisposing syndrome|PTEN h...    Yes               1*   \n",
       "2                      PTEN hamartoma tumor syndrome     No              NaN   \n",
       "3  Hereditary cancer-predisposing syndrome|not pr...    Yes               1*   \n",
       "4                                                NaN     No              NaN   \n",
       "\n",
       "  Cosmic - primary tissue type Cosmic -histology gnomAD  \\\n",
       "0                      Thyroid         Carcinoma    Yes   \n",
       "1                      Thyroid         Carcinoma     No   \n",
       "2                          NaN               NaN     No   \n",
       "3                         Lung         Carcinoma    Yes   \n",
       "4                          NaN               NaN    Yes   \n",
       "\n",
       "   gnomAD - Allele frequency  ...  \\\n",
       "0                      137.0  ...   \n",
       "1                        NaN  ...   \n",
       "2                        NaN  ...   \n",
       "3                        2.0  ...   \n",
       "4                        1.0  ...   \n",
       "\n",
       "   same amino acid to disease info (from ClinVar/COSMIC/gnomAD)  \\\n",
       "0                                  COSMIC - D268Y 2*              \n",
       "1                                                NaN              \n",
       "2                                                NaN              \n",
       "3              ClinVar - N117H - VUS N117H in gnomAD              \n",
       "4                                  COSMIC - N262I 1*              \n",
       "\n",
       "  Additional comments Wing Assay function calls Post_expt_function calls  \\\n",
       "0                 NaN                   Like WT                   Benign   \n",
       "1                 NaN                   Like WT                   Benign   \n",
       "2                 NaN                   Like WT                   Benign   \n",
       "3                 NaN                   Like WT                   Benign   \n",
       "4                 NaN                   Like WT                   Benign   \n",
       "\n",
       "  HEK pAKT/AKT data (Post et al) Yeast (Post et al) Yeast (Mighell et al)  \\\n",
       "0                             WT                 WT              0.225343   \n",
       "1                             WT                 WT              0.449992   \n",
       "2                             WT                 WT             -0.338630   \n",
       "3                             WT                 WT              0.240901   \n",
       "4                             WT                 WT             -0.208005   \n",
       "\n",
       "  function_calls              Additional comments on function calls  \\\n",
       "0         Benign  Additionally ClinVar says it is likely benign....   \n",
       "1         Benign                  WT in Drosophila eclosion as well   \n",
       "2         Benign                        pLOF in Drosophila eclosion   \n",
       "3         Benign                         GOF in Drosophila eclosion   \n",
       "4         Benign                        pLOF in Drosophila eclosion   \n",
       "\n",
       "  Eclosion Assay  \n",
       "0    Partial LOF  \n",
       "1             WT  \n",
       "2    Partial LOF  \n",
       "3            GOF  \n",
       "4    Partial LOF  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_var = pd.read_csv('Fly stuff/PTEN_vars_summary_updated.csv', index_col=False)\n",
    "df_var.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## scatterplot"
   ]
  },
  {
   "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>variants</th>\n",
       "      <th>normalized_mean</th>\n",
       "      <th>stdev</th>\n",
       "      <th>num_replicates</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>D22E</td>\n",
       "      <td>-0.804356</td>\n",
       "      <td>-0.073384</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>G127R</td>\n",
       "      <td>-0.482834</td>\n",
       "      <td>-0.036107</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C124S</td>\n",
       "      <td>-0.440364</td>\n",
       "      <td>-0.034346</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>R130Q</td>\n",
       "      <td>-0.407882</td>\n",
       "      <td>-0.028380</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F241S</td>\n",
       "      <td>-0.392750</td>\n",
       "      <td>-0.028349</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>H93R</td>\n",
       "      <td>-0.359511</td>\n",
       "      <td>-0.027831</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Y68N</td>\n",
       "      <td>-0.320319</td>\n",
       "      <td>-0.031615</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>N276S</td>\n",
       "      <td>-0.314472</td>\n",
       "      <td>-0.022544</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>A34D</td>\n",
       "      <td>-0.307810</td>\n",
       "      <td>-0.020746</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>H123Y</td>\n",
       "      <td>-0.295581</td>\n",
       "      <td>-0.021910</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants  normalized_mean     stdev  num_replicates\n",
       "0     D22E        -0.804356 -0.073384               9\n",
       "1    G127R        -0.482834 -0.036107               8\n",
       "2    C124S        -0.440364 -0.034346              10\n",
       "3    R130Q        -0.407882 -0.028380               7\n",
       "4    F241S        -0.392750 -0.028349              10\n",
       "5     H93R        -0.359511 -0.027831               9\n",
       "6     Y68N        -0.320319 -0.031615              10\n",
       "7    N276S        -0.314472 -0.022544               8\n",
       "8     A34D        -0.307810 -0.020746               8\n",
       "9    H123Y        -0.295581 -0.021910              10"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sort dataframe according to fc_code and by wing_mean_normalized\n",
    "df_wing_sorted = df_wing.sort_values(by=['normalized_mean']).reset_index(drop=True)\n",
    "df_wing_sorted.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot using matplotlib.pyplot\n",
    "x = df_wing_sorted.index\n",
    "y = df_wing_sorted['normalized_mean']\n",
    "x_labels = df_wing_sorted['variants']\n",
    "err = df_wing_sorted['stdev']\n",
    "\n",
    "fig = plt.figure(figsize=(18, 6))\n",
    "plt.errorbar(x, y, yerr=err, xerr=None, fmt='ob',\n",
    "             ecolor='lightblue', elinewidth=2.5, capsize=3.0)\n",
    "xmin, xmax = plt.xlim()\n",
    "plt.hlines(0, xmin, xmax, colors='red', linestyles='dashed')\n",
    "plt.hlines(1, xmin, xmax, colors='green', linestyles='dashed')\n",
    "# formatting arguments\n",
    "plt.xticks(ticks=x, labels=x_labels, rotation=90)\n",
    "plt.ylabel('normalized wing size')\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot using matplotlib.axes.Axes\n",
    "x = df_wing_sorted.index\n",
    "y = df_wing_sorted['normalized_mean']\n",
    "err = df_wing_sorted['stdev']\n",
    "fig, ax = plt.subplots(1, 1, figsize=(18,6))\n",
    "ax.errorbar(x, y, xerr=None, yerr=err, fmt='ob', ecolor='lightblue', capsize=2.0)\n",
    "ax.axhline(y = 0, linestyle='--', c='r')\n",
    "ax.axhline(y = 1, linestyle='--', c='g')\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(df_wing_sorted['variants'].to_list(), rotation=90)\n",
    "ax.set_ylabel('normalized wing size')\n",
    "\n",
    "# save fig\n",
    "mpl.rcParams['pdf.fonttype'] = 42\n",
    "plt.savefig('Fly stuff/wing_scatterplot_normalized_sorted.pdf', transparent=True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## classified scatterplot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "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>variants</th>\n",
       "      <th>normalized_mean</th>\n",
       "      <th>stdev</th>\n",
       "      <th>num_replicates</th>\n",
       "      <th>function_calls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4A</td>\n",
       "      <td>3.739984</td>\n",
       "      <td>0.278252</td>\n",
       "      <td>10</td>\n",
       "      <td>biochemical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C124S</td>\n",
       "      <td>-0.440364</td>\n",
       "      <td>-0.034346</td>\n",
       "      <td>10</td>\n",
       "      <td>biochemical</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>D22E</td>\n",
       "      <td>-0.804356</td>\n",
       "      <td>-0.073384</td>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>P38H</td>\n",
       "      <td>-0.199249</td>\n",
       "      <td>-0.013897</td>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>L70V</td>\n",
       "      <td>0.039773</td>\n",
       "      <td>0.002589</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants  normalized_mean     stdev  num_replicates function_calls\n",
       "0       4A         3.739984  0.278252              10    biochemical\n",
       "1    C124S        -0.440364 -0.034346              10    biochemical\n",
       "2     D22E        -0.804356 -0.073384               9            NaN\n",
       "3     P38H        -0.199249 -0.013897               9            NaN\n",
       "4     L70V         0.039773  0.002589               8            NaN"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# merge in function calls\n",
    "df_class = df_wing.merge(df_var[['variants', 'function_calls']], how='left', on='variants')\n",
    "df_class.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Pathogenic': 0, 'Benign': 1, 'biochemical': 2, 'NaN': 3}\n"
     ]
    }
   ],
   "source": [
    "# make a dictionary to map categorical\n",
    "d = {'Pathogenic':0,'Benign':1,'biochemical':2,'NaN':3}\n",
    "print(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "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>variants</th>\n",
       "      <th>normalized_mean</th>\n",
       "      <th>stdev</th>\n",
       "      <th>num_replicates</th>\n",
       "      <th>function_calls</th>\n",
       "      <th>fc_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4A</td>\n",
       "      <td>3.739984</td>\n",
       "      <td>0.278252</td>\n",
       "      <td>10</td>\n",
       "      <td>biochemical</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C124S</td>\n",
       "      <td>-0.440364</td>\n",
       "      <td>-0.034346</td>\n",
       "      <td>10</td>\n",
       "      <td>biochemical</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>D22E</td>\n",
       "      <td>-0.804356</td>\n",
       "      <td>-0.073384</td>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>P38H</td>\n",
       "      <td>-0.199249</td>\n",
       "      <td>-0.013897</td>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>L70V</td>\n",
       "      <td>0.039773</td>\n",
       "      <td>0.002589</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants  normalized_mean     stdev  num_replicates function_calls fc_code\n",
       "0       4A         3.739984  0.278252              10    biochemical       2\n",
       "1    C124S        -0.440364 -0.034346              10    biochemical       2\n",
       "2     D22E        -0.804356 -0.073384               9            NaN       3\n",
       "3     P38H        -0.199249 -0.013897               9            NaN       3\n",
       "4     L70V         0.039773  0.002589               8            NaN       3"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# convert NaN in \"function_calls\" to string\n",
    "df_class['function_calls'].fillna('NaN', inplace=True)\n",
    "\n",
    "# convert \"function_calls\" to categorical to allow encoding\n",
    "df_class['function_calls'] = df_class['function_calls'].astype('category')\n",
    "\n",
    "# perform categorical encoding\n",
    "df_class['fc_code'] = df_class['function_calls'].map(d)\n",
    "\n",
    "df_class.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "variants             object\n",
       "normalized_mean     float64\n",
       "stdev               float64\n",
       "num_replicates        int64\n",
       "function_calls     category\n",
       "fc_code            category\n",
       "dtype: object"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check data types\n",
    "df_class.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "# change the dtype of fc_code from 'category' to 'int'\n",
    "# so it can be sorted\n",
    "df_class['fc_code'] = df_class['fc_code'].astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "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>variants</th>\n",
       "      <th>normalized_mean</th>\n",
       "      <th>stdev</th>\n",
       "      <th>num_replicates</th>\n",
       "      <th>function_calls</th>\n",
       "      <th>fc_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>R130Q</td>\n",
       "      <td>-0.407882</td>\n",
       "      <td>-0.028380</td>\n",
       "      <td>7</td>\n",
       "      <td>Pathogenic</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>H123Y</td>\n",
       "      <td>-0.295581</td>\n",
       "      <td>-0.021910</td>\n",
       "      <td>10</td>\n",
       "      <td>Pathogenic</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Y68H</td>\n",
       "      <td>-0.254862</td>\n",
       "      <td>-0.026123</td>\n",
       "      <td>10</td>\n",
       "      <td>Pathogenic</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>D107V</td>\n",
       "      <td>-0.209826</td>\n",
       "      <td>-0.015396</td>\n",
       "      <td>7</td>\n",
       "      <td>Pathogenic</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>T131I</td>\n",
       "      <td>-0.154321</td>\n",
       "      <td>-0.011719</td>\n",
       "      <td>10</td>\n",
       "      <td>Pathogenic</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  variants  normalized_mean     stdev  num_replicates function_calls  fc_code\n",
       "0    R130Q        -0.407882 -0.028380               7     Pathogenic        0\n",
       "1    H123Y        -0.295581 -0.021910              10     Pathogenic        0\n",
       "2     Y68H        -0.254862 -0.026123              10     Pathogenic        0\n",
       "3    D107V        -0.209826 -0.015396               7     Pathogenic        0\n",
       "4    T131I        -0.154321 -0.011719              10     Pathogenic        0"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sort dataframe according to fc_code then by mean\n",
    "df_class_sorted = df_class.sort_values(by=['fc_code', 'normalized_mean'])\n",
    "df_class_sorted = df_class_sorted.reset_index(drop=True)\n",
    "\n",
    "df_class_sorted.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   variants  normalized_mean     stdev  num_replicates function_calls  fc_code\n",
      "12    P246L         0.462501  0.054069               9     Pathogenic        0\n",
      "   variants  normalized_mean     stdev  num_replicates function_calls  fc_code\n",
      "13    R130L         0.714415  0.059732              10     Pathogenic        0\n",
      "   variants  normalized_mean     stdev  num_replicates function_calls  fc_code\n",
      "14    I400V         0.742121  0.080273              10         Benign        1\n"
     ]
    }
   ],
   "source": [
    "# inspect variants at thresholds\n",
    "vars = ['P246L', 'R130L', 'I400V']\n",
    "for v in vars:\n",
    "    print(df_class_sorted[df_class_sorted['variants']==v])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot using matplotlib.axes.Axes\n",
    "x = df_class_sorted.index\n",
    "y = df_class_sorted['normalized_mean']\n",
    "err = df_class_sorted['stdev']\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(18,6))\n",
    "ax.scatter(x, y, c='darkorange', marker='o')\n",
    "ax.errorbar(x, y, xerr=None, yerr=err, fmt='none', ecolor='grey', capsize=2.0)\n",
    "\n",
    "ax.axhline(y=0, linestyle='--', c='r')\n",
    "ax.axhline(y=1, linestyle='--', c='g')\n",
    "ax.axhline(y=0.462501, linestyle='--', c='deepskyblue')\n",
    "ax.axhline(y=0.742121, linestyle='--', c='deepskyblue')\n",
    "\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(df_class_sorted['variants'].to_list(), rotation=90)\n",
    "ax.set_ylabel('normalized wing size')\n",
    "\n",
    "# save fig\n",
    "mpl.rcParams['pdf.fonttype'] = 42\n",
    "plt.savefig('Fly stuff/wing_scatterplot_classified.pdf', transparent=True)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}