growth_curves/20240129_validations_test3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ec8ac722-12d4-423e-b173-5ba897b4182c",
   "metadata": {},
   "source": [
    "# Growth curve analysis of validation mutants"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11890dcb-5fe2-4621-b644-b216e282810f",
   "metadata": {},
   "source": [
    "## Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ec072e2b-9a8f-4d99-bcdf-8daf296b98c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.rcParams['svg.fonttype'] = 'none'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "edffbb3d-393f-4961-b010-5c09a4e44aa4",
   "metadata": {},
   "source": [
    "## Specify paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c22c9445-cb76-491b-b0b7-5bd4d68b15ec",
   "metadata": {},
   "outputs": [],
   "source": [
    "layout_path = '20240129_full_layout.tsv'\n",
    "source_data = ['20240128_validations_DMS_control.xlsx',\n",
    "               '20240128_validations_DMS_caspo.xlsx',\n",
    "               '20240131_validations_DMS_mica.xlsx',\n",
    "               '20240131_validations_DMS_ani.xlsx']\n",
    "DMS_data = '../aggregated_data/classified_variants/BY4741_FKS1-HS1_single/nt_refined_classification.csv'\n",
    "aggdata_outpath = '../aggregated_data/'\n",
    "graph_outpath = '../manuscript_figures/'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a78dcede-4e8b-461c-9744-bfafa9f469a2",
   "metadata": {},
   "source": [
    "## Get layout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "126d5095-80a7-48a0-8d50-4d9e0f7c98fd",
   "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>plate</th>\n",
       "      <th>well</th>\n",
       "      <th>compound</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>genotype</th>\n",
       "      <th>clone</th>\n",
       "      <th>Sanger_validated</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>A1</td>\n",
       "      <td>none</td>\n",
       "      <td>LVLSLRDP</td>\n",
       "      <td>F639delta</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>B1</td>\n",
       "      <td>none</td>\n",
       "      <td>LVLSLRDP</td>\n",
       "      <td>F639delta</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>C1</td>\n",
       "      <td>none</td>\n",
       "      <td>FVLSLRDP</td>\n",
       "      <td>L40delta</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>D1</td>\n",
       "      <td>none</td>\n",
       "      <td>FVLSLRDP</td>\n",
       "      <td>L40delta</td>\n",
       "      <td>4</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>E1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLLSLRDP</td>\n",
       "      <td>V641delta</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>315</th>\n",
       "      <td>4</td>\n",
       "      <td>D10</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>QLVLSLRDP</td>\n",
       "      <td>F639Q</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>316</th>\n",
       "      <td>4</td>\n",
       "      <td>E10</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>WT</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>317</th>\n",
       "      <td>4</td>\n",
       "      <td>F10</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>WT</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>318</th>\n",
       "      <td>4</td>\n",
       "      <td>G10</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>319</th>\n",
       "      <td>4</td>\n",
       "      <td>H10</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>320 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     plate well       compound     aa_seq   genotype  clone  Sanger_validated\n",
       "0        1   A1           none   LVLSLRDP  F639delta      2             False\n",
       "1        1   B1           none   LVLSLRDP  F639delta      3             False\n",
       "2        1   C1           none   FVLSLRDP   L40delta      3             False\n",
       "3        1   D1           none   FVLSLRDP   L40delta      4             False\n",
       "4        1   E1           none   FLLSLRDP  V641delta      3             False\n",
       "..     ...  ...            ...        ...        ...    ...               ...\n",
       "315      4  D10  anidulafungin  QLVLSLRDP      F639Q      3             False\n",
       "316      4  E10  anidulafungin  FLVLSLRDP         WT      1              True\n",
       "317      4  F10  anidulafungin  FLVLSLRDP         WT      2              True\n",
       "318      4  G10  anidulafungin  FLVLSLRDP         BY      1              True\n",
       "319      4  H10  anidulafungin  FLVLSLRDP         BY      2              True\n",
       "\n",
       "[320 rows x 7 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "layout = pd.read_csv(layout_path, sep='\\t', header=0, dtype={'Sanger_validated': 'boolean'})\n",
    "layout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8d9a5c9c-0a9e-4eb7-b96b-84cd79551ff6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique genotypes validated by Sanger:\n",
      " ['F639C' 'L640D' 'L642G' 'P647Q' 'P647N' 'F639V' 'F639S' 'L642Y' 'L642F'\n",
      " 'S643P' 'S643F' 'S643Y' 'S643C' 'L640K' 'V641S' 'L640G' 'D646W' 'F639D'\n",
      " 'F639E' 'V641F' 'V641H' 'L642K' 'S643L' 'R645D' 'R645S' 'V641W' 'R645Q'\n",
      " 'F639Q' 'WT' 'BY']\n"
     ]
    }
   ],
   "source": [
    "print('Unique genotypes validated by Sanger:\\n', layout.loc[layout.Sanger_validated, 'genotype'].unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28e177b7-f44b-475a-b21c-b7510bc06aeb",
   "metadata": {},
   "source": [
    "## Get plate reader data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f488d429-8b10-49a4-b79d-ca0486a11936",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data(fpath):\n",
    "    ### Import libraries\n",
    "    import numpy as np\n",
    "    from datetime import datetime, date, time\n",
    "    \n",
    "    ### Read excel file\n",
    "    source_df = pd.read_excel(fpath, index_col=0, header=0, skiprows=26, skipfooter=36)\n",
    "    \n",
    "    ### Parse timepoints\n",
    "    t0, t1 = source_df.iloc[[0,1],0] # Get the first and second timepoints\n",
    "\n",
    "    # Convert first timepoint from time (of the day) object to time duration (h)\n",
    "    t0_h = (datetime.combine(date.today(), t0) - datetime.combine(date.today(), time.min)).total_seconds() / 3600\n",
    "\n",
    "    # Convert difference between second and first timepoint into duration in h (measurement interval)\n",
    "    delta = (datetime.combine(date.today(), t1) - datetime.combine(date.today(), t0)).total_seconds() / 3600\n",
    "\n",
    "    # Rewrite column of timepoints using calculated values\n",
    "    source_df['Time'] = np.arange(t0_h, t0_h + len(source_df.Time)*delta, delta)\n",
    "\n",
    "    # Delete column with temperature\n",
    "    source_df.drop(columns='T° 600', inplace=True)\n",
    "\n",
    "    # Rename columns\n",
    "    source_df.columns = [x.split('=')[-1] if '=' in x else x for x in source_df.columns]\n",
    "    \n",
    "    # Reshape\n",
    "    longdf = source_df.melt(id_vars = 'Time', var_name = 'well', value_name = 'OD') # wide to long dataframe\n",
    "    return longdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c4c640b7-50e5-4ce7-90c9-96c8ac4ea849",
   "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>Time</th>\n",
       "      <th>well</th>\n",
       "      <th>OD</th>\n",
       "      <th>plate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00</td>\n",
       "      <td>A1</td>\n",
       "      <td>0.116</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.25</td>\n",
       "      <td>A1</td>\n",
       "      <td>0.116</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.50</td>\n",
       "      <td>A1</td>\n",
       "      <td>0.117</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.75</td>\n",
       "      <td>A1</td>\n",
       "      <td>0.118</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.00</td>\n",
       "      <td>A1</td>\n",
       "      <td>0.119</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110971</th>\n",
       "      <td>71.00</td>\n",
       "      <td>H12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110972</th>\n",
       "      <td>71.25</td>\n",
       "      <td>H12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110973</th>\n",
       "      <td>71.50</td>\n",
       "      <td>H12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110974</th>\n",
       "      <td>71.75</td>\n",
       "      <td>H12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110975</th>\n",
       "      <td>72.00</td>\n",
       "      <td>H12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>110976 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Time well     OD  plate\n",
       "0        0.00   A1  0.116      1\n",
       "1        0.25   A1  0.116      1\n",
       "2        0.50   A1  0.117      1\n",
       "3        0.75   A1  0.118      1\n",
       "4        1.00   A1  0.119      1\n",
       "...       ...  ...    ...    ...\n",
       "110971  71.00  H12    NaN      4\n",
       "110972  71.25  H12    NaN      4\n",
       "110973  71.50  H12    NaN      4\n",
       "110974  71.75  H12    NaN      4\n",
       "110975  72.00  H12    NaN      4\n",
       "\n",
       "[110976 rows x 4 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_list = []\n",
    "\n",
    "for i,f in enumerate(source_data):\n",
    "        df = get_data(f)\n",
    "        df['plate'] = i+1\n",
    "        df_list.append(df)\n",
    "\n",
    "fulldf = pd.concat(df_list, ignore_index=True) \n",
    "fulldf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95adf0bd-dc80-4fd2-9c5a-0ce2a3fb14a1",
   "metadata": {},
   "source": [
    "## Annotate data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7a21413b-04ef-40e0-bdea-0d4a2dab09ce",
   "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>Time</th>\n",
       "      <th>well</th>\n",
       "      <th>OD</th>\n",
       "      <th>plate</th>\n",
       "      <th>compound</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>genotype</th>\n",
       "      <th>clone</th>\n",
       "      <th>Sanger_validated</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00</td>\n",
       "      <td>A4</td>\n",
       "      <td>0.118</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>P647Q</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.25</td>\n",
       "      <td>A4</td>\n",
       "      <td>0.118</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>P647Q</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.50</td>\n",
       "      <td>A4</td>\n",
       "      <td>0.119</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>P647Q</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.75</td>\n",
       "      <td>A4</td>\n",
       "      <td>0.120</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>P647Q</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.00</td>\n",
       "      <td>A4</td>\n",
       "      <td>0.122</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>P647Q</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57795</th>\n",
       "      <td>71.00</td>\n",
       "      <td>H10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57796</th>\n",
       "      <td>71.25</td>\n",
       "      <td>H10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57797</th>\n",
       "      <td>71.50</td>\n",
       "      <td>H10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57798</th>\n",
       "      <td>71.75</td>\n",
       "      <td>H10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57799</th>\n",
       "      <td>72.00</td>\n",
       "      <td>H10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>BY</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>57800 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Time well     OD  plate       compound     aa_seq genotype  clone  \\\n",
       "0       0.00   A4  0.118      1           none  FLVLSLRDQ    P647Q      2   \n",
       "1       0.25   A4  0.118      1           none  FLVLSLRDQ    P647Q      2   \n",
       "2       0.50   A4  0.119      1           none  FLVLSLRDQ    P647Q      2   \n",
       "3       0.75   A4  0.120      1           none  FLVLSLRDQ    P647Q      2   \n",
       "4       1.00   A4  0.122      1           none  FLVLSLRDQ    P647Q      2   \n",
       "...      ...  ...    ...    ...            ...        ...      ...    ...   \n",
       "57795  71.00  H10    NaN      4  anidulafungin  FLVLSLRDP       BY      2   \n",
       "57796  71.25  H10    NaN      4  anidulafungin  FLVLSLRDP       BY      2   \n",
       "57797  71.50  H10    NaN      4  anidulafungin  FLVLSLRDP       BY      2   \n",
       "57798  71.75  H10    NaN      4  anidulafungin  FLVLSLRDP       BY      2   \n",
       "57799  72.00  H10    NaN      4  anidulafungin  FLVLSLRDP       BY      2   \n",
       "\n",
       "       Sanger_validated  \n",
       "0                  True  \n",
       "1                  True  \n",
       "2                  True  \n",
       "3                  True  \n",
       "4                  True  \n",
       "...                 ...  \n",
       "57795              True  \n",
       "57796              True  \n",
       "57797              True  \n",
       "57798              True  \n",
       "57799              True  \n",
       "\n",
       "[57800 rows x 9 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "annotdf = fulldf.merge(right=layout[layout.Sanger_validated], on=['plate','well'])\n",
    "annotdf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6a12666-026a-4df7-96a7-d825a8ade977",
   "metadata": {},
   "source": [
    "## Visualize growth curves"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "344b4205-5b50-4753-8592-f8a59ec46e7a",
   "metadata": {},
   "source": [
    "```\n",
    "custom_params = {\"axes.spines.right\": False, \"axes.spines.top\": False}\n",
    "sns.set_theme(style=\"ticks\", rc=custom_params)\n",
    "\n",
    "grid = sns.FacetGrid(data=annotdf, col = 'compound', hue='genotype', palette='hls')\n",
    "grid.map(sns.lineplot, 'Time', 'OD')\n",
    "\n",
    "grid.set_titles(row_template='{row_name}', col_template='{col_name}')\n",
    "grid.set_axis_labels('Time (h)', 'OD')\n",
    "grid.add_legend(title = 'Genotype')\n",
    "grid.fig.subplots_adjust(top=0.9)\n",
    "grid.tight_layout()\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98cb2b1d-9d78-49da-8d1c-21001a8c16f2",
   "metadata": {},
   "source": [
    "## Calculate normalized area under the curve (AUC)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3b459857-127a-4c6d-a4ef-9674c3ff757d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_auc(g):\n",
    "    import numpy as np\n",
    "    return np.trapz(g.OD)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "04f046e8-ee58-4353-8087-8b0b4ac86317",
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>clone</th>\n",
       "      <th>compound</th>\n",
       "      <th>well</th>\n",
       "      <th>auc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>1</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>G10</td>\n",
       "      <td>22.6925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>1</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>G10</td>\n",
       "      <td>21.9170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>1</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>G10</td>\n",
       "      <td>21.3405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>G10</td>\n",
       "      <td>147.2240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>2</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>H10</td>\n",
       "      <td>29.2160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>1</td>\n",
       "      <td>none</td>\n",
       "      <td>E10</td>\n",
       "      <td>144.9465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>2</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>F10</td>\n",
       "      <td>24.2550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>2</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>F10</td>\n",
       "      <td>20.9890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>2</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>F10</td>\n",
       "      <td>22.4110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>2</td>\n",
       "      <td>none</td>\n",
       "      <td>F10</td>\n",
       "      <td>147.2450</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>200 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    genotype     aa_seq  clone       compound well       auc\n",
       "0         BY  FLVLSLRDP      1  anidulafungin  G10   22.6925\n",
       "1         BY  FLVLSLRDP      1    caspofungin  G10   21.9170\n",
       "2         BY  FLVLSLRDP      1     micafungin  G10   21.3405\n",
       "3         BY  FLVLSLRDP      1           none  G10  147.2240\n",
       "4         BY  FLVLSLRDP      2  anidulafungin  H10   29.2160\n",
       "..       ...        ...    ...            ...  ...       ...\n",
       "195       WT  FLVLSLRDP      1           none  E10  144.9465\n",
       "196       WT  FLVLSLRDP      2  anidulafungin  F10   24.2550\n",
       "197       WT  FLVLSLRDP      2    caspofungin  F10   20.9890\n",
       "198       WT  FLVLSLRDP      2     micafungin  F10   22.4110\n",
       "199       WT  FLVLSLRDP      2           none  F10  147.2450\n",
       "\n",
       "[200 rows x 6 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aucdf = annotdf[annotdf['Time'] <= 40].groupby(['genotype','aa_seq','clone','compound','well'\n",
    "                                                   ])[['OD']].apply(func=get_auc).reset_index(name='auc')\n",
    "aucdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9985a933-d095-4547-a609-6e34ae4b4fac",
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>compound</th>\n",
       "      <th>auc</th>\n",
       "      <th>auc_min</th>\n",
       "      <th>auc_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>25.95425</td>\n",
       "      <td>22.6925</td>\n",
       "      <td>29.2160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>22.90100</td>\n",
       "      <td>21.9170</td>\n",
       "      <td>23.8850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>24.07575</td>\n",
       "      <td>21.3405</td>\n",
       "      <td>26.8110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>134.43275</td>\n",
       "      <td>121.6415</td>\n",
       "      <td>147.2240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>122.99300</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>122.9930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>126.28600</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>126.2860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>41.22150</td>\n",
       "      <td>41.2215</td>\n",
       "      <td>41.2215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>none</td>\n",
       "      <td>142.10700</td>\n",
       "      <td>142.1070</td>\n",
       "      <td>142.1070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>143.19975</td>\n",
       "      <td>139.3225</td>\n",
       "      <td>147.0770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>135.58750</td>\n",
       "      <td>132.0890</td>\n",
       "      <td>139.0860</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  genotype     aa_seq       compound        auc   auc_min   auc_max\n",
       "0       BY  FLVLSLRDP  anidulafungin   25.95425   22.6925   29.2160\n",
       "1       BY  FLVLSLRDP    caspofungin   22.90100   21.9170   23.8850\n",
       "2       BY  FLVLSLRDP     micafungin   24.07575   21.3405   26.8110\n",
       "3       BY  FLVLSLRDP           none  134.43275  121.6415  147.2240\n",
       "4    D646W  FLVLSLRWP  anidulafungin  122.99300  122.9930  122.9930\n",
       "5    D646W  FLVLSLRWP    caspofungin  126.28600  126.2860  126.2860\n",
       "6    D646W  FLVLSLRWP     micafungin   41.22150   41.2215   41.2215\n",
       "7    D646W  FLVLSLRWP           none  142.10700  142.1070  142.1070\n",
       "8    F639C  CLVLSLRDP  anidulafungin  143.19975  139.3225  147.0770\n",
       "9    F639C  CLVLSLRDP    caspofungin  135.58750  132.0890  139.0860"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggdf = aucdf.groupby(['genotype','aa_seq','compound']\n",
    "                     )[['auc']].agg(auc = ('auc', 'mean'),\n",
    "                                    auc_min = ('auc', 'min'),\n",
    "                                    auc_max = ('auc', 'max')\n",
    "                                   ).reset_index()\n",
    "aggdf.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "5c3162eb-4d9b-4fdf-be74-9bb6a8b538af",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_l2fc(val, minv, maxv, comp, df):    \n",
    "    import numpy as np\n",
    "    \n",
    "    # Retrieve corresponding WT value for the condition\n",
    "    wt = df.loc[(df.compound == comp)\n",
    "                & (df.genotype == 'BY'),\n",
    "                'auc'\n",
    "               ].values\n",
    "    \n",
    "    # Make sure a single value was extracted\n",
    "    if len(wt) == 1:\n",
    "        # Return log2 fold-change for mean, min and max\n",
    "        return np.log2(val/wt[0]), np.log2(minv/wt[0]), np.log2(maxv/wt[0])\n",
    "    else:\n",
    "        return 'error'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f91424e3-73fa-49c3-98ca-5b0fa2c04210",
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>compound</th>\n",
       "      <th>auc</th>\n",
       "      <th>auc_min</th>\n",
       "      <th>auc_max</th>\n",
       "      <th>L2FC</th>\n",
       "      <th>L2FC_min</th>\n",
       "      <th>L2FC_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>25.95425</td>\n",
       "      <td>22.6925</td>\n",
       "      <td>29.2160</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.193755</td>\n",
       "      <td>0.170788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>22.90100</td>\n",
       "      <td>21.9170</td>\n",
       "      <td>23.8850</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.063360</td>\n",
       "      <td>0.060694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>24.07575</td>\n",
       "      <td>21.3405</td>\n",
       "      <td>26.8110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.173987</td>\n",
       "      <td>0.155244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>134.43275</td>\n",
       "      <td>121.6415</td>\n",
       "      <td>147.2240</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.144249</td>\n",
       "      <td>0.131128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>122.99300</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>126.28600</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>41.22150</td>\n",
       "      <td>41.2215</td>\n",
       "      <td>41.2215</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.775816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>none</td>\n",
       "      <td>142.10700</td>\n",
       "      <td>142.1070</td>\n",
       "      <td>142.1070</td>\n",
       "      <td>0.080093</td>\n",
       "      <td>0.080093</td>\n",
       "      <td>0.080093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>143.19975</td>\n",
       "      <td>139.3225</td>\n",
       "      <td>147.0770</td>\n",
       "      <td>2.463986</td>\n",
       "      <td>2.424386</td>\n",
       "      <td>2.502529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>135.58750</td>\n",
       "      <td>132.0890</td>\n",
       "      <td>139.0860</td>\n",
       "      <td>2.565742</td>\n",
       "      <td>2.528028</td>\n",
       "      <td>2.602495</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  genotype     aa_seq       compound        auc   auc_min   auc_max      L2FC  \\\n",
       "0       BY  FLVLSLRDP  anidulafungin   25.95425   22.6925   29.2160  0.000000   \n",
       "1       BY  FLVLSLRDP    caspofungin   22.90100   21.9170   23.8850  0.000000   \n",
       "2       BY  FLVLSLRDP     micafungin   24.07575   21.3405   26.8110  0.000000   \n",
       "3       BY  FLVLSLRDP           none  134.43275  121.6415  147.2240  0.000000   \n",
       "4    D646W  FLVLSLRWP  anidulafungin  122.99300  122.9930  122.9930  2.244534   \n",
       "5    D646W  FLVLSLRWP    caspofungin  126.28600  126.2860  126.2860  2.463212   \n",
       "6    D646W  FLVLSLRWP     micafungin   41.22150   41.2215   41.2215  0.775816   \n",
       "7    D646W  FLVLSLRWP           none  142.10700  142.1070  142.1070  0.080093   \n",
       "8    F639C  CLVLSLRDP  anidulafungin  143.19975  139.3225  147.0770  2.463986   \n",
       "9    F639C  CLVLSLRDP    caspofungin  135.58750  132.0890  139.0860  2.565742   \n",
       "\n",
       "   L2FC_min  L2FC_max  \n",
       "0 -0.193755  0.170788  \n",
       "1 -0.063360  0.060694  \n",
       "2 -0.173987  0.155244  \n",
       "3 -0.144249  0.131128  \n",
       "4  2.244534  2.244534  \n",
       "5  2.463212  2.463212  \n",
       "6  0.775816  0.775816  \n",
       "7  0.080093  0.080093  \n",
       "8  2.424386  2.502529  \n",
       "9  2.528028  2.602495  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggdf['L2FC'], aggdf['L2FC_min'], aggdf['L2FC_max'] = zip(*aggdf.apply(lambda row: get_l2fc(row.auc, row.auc_min, row.auc_max, row.compound, aggdf), axis=1))\n",
    "aggdf.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "87d7b99b-709a-4891-8d62-147880380476",
   "metadata": {
    "tags": []
   },
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>compound</th>\n",
       "      <th>auc</th>\n",
       "      <th>auc_min</th>\n",
       "      <th>auc_max</th>\n",
       "      <th>L2FC</th>\n",
       "      <th>L2FC_min</th>\n",
       "      <th>L2FC_max</th>\n",
       "      <th>min_yerr</th>\n",
       "      <th>max_yerr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>25.95425</td>\n",
       "      <td>22.6925</td>\n",
       "      <td>29.2160</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.193755</td>\n",
       "      <td>0.170788</td>\n",
       "      <td>0.193755</td>\n",
       "      <td>0.170788</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>22.90100</td>\n",
       "      <td>21.9170</td>\n",
       "      <td>23.8850</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.063360</td>\n",
       "      <td>0.060694</td>\n",
       "      <td>0.063360</td>\n",
       "      <td>0.060694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>24.07575</td>\n",
       "      <td>21.3405</td>\n",
       "      <td>26.8110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.173987</td>\n",
       "      <td>0.155244</td>\n",
       "      <td>0.173987</td>\n",
       "      <td>0.155244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>134.43275</td>\n",
       "      <td>121.6415</td>\n",
       "      <td>147.2240</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.144249</td>\n",
       "      <td>0.131128</td>\n",
       "      <td>0.144249</td>\n",
       "      <td>0.131128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>122.99300</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>126.28600</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>41.22150</td>\n",
       "      <td>41.2215</td>\n",
       "      <td>41.2215</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>none</td>\n",
       "      <td>142.10700</td>\n",
       "      <td>142.1070</td>\n",
       "      <td>142.1070</td>\n",
       "      <td>0.080093</td>\n",
       "      <td>0.080093</td>\n",
       "      <td>0.080093</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>143.19975</td>\n",
       "      <td>139.3225</td>\n",
       "      <td>147.0770</td>\n",
       "      <td>2.463986</td>\n",
       "      <td>2.424386</td>\n",
       "      <td>2.502529</td>\n",
       "      <td>0.039601</td>\n",
       "      <td>0.038543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>135.58750</td>\n",
       "      <td>132.0890</td>\n",
       "      <td>139.0860</td>\n",
       "      <td>2.565742</td>\n",
       "      <td>2.528028</td>\n",
       "      <td>2.602495</td>\n",
       "      <td>0.037714</td>\n",
       "      <td>0.036753</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  genotype     aa_seq       compound        auc   auc_min   auc_max      L2FC  \\\n",
       "0       BY  FLVLSLRDP  anidulafungin   25.95425   22.6925   29.2160  0.000000   \n",
       "1       BY  FLVLSLRDP    caspofungin   22.90100   21.9170   23.8850  0.000000   \n",
       "2       BY  FLVLSLRDP     micafungin   24.07575   21.3405   26.8110  0.000000   \n",
       "3       BY  FLVLSLRDP           none  134.43275  121.6415  147.2240  0.000000   \n",
       "4    D646W  FLVLSLRWP  anidulafungin  122.99300  122.9930  122.9930  2.244534   \n",
       "5    D646W  FLVLSLRWP    caspofungin  126.28600  126.2860  126.2860  2.463212   \n",
       "6    D646W  FLVLSLRWP     micafungin   41.22150   41.2215   41.2215  0.775816   \n",
       "7    D646W  FLVLSLRWP           none  142.10700  142.1070  142.1070  0.080093   \n",
       "8    F639C  CLVLSLRDP  anidulafungin  143.19975  139.3225  147.0770  2.463986   \n",
       "9    F639C  CLVLSLRDP    caspofungin  135.58750  132.0890  139.0860  2.565742   \n",
       "\n",
       "   L2FC_min  L2FC_max  min_yerr  max_yerr  \n",
       "0 -0.193755  0.170788  0.193755  0.170788  \n",
       "1 -0.063360  0.060694  0.063360  0.060694  \n",
       "2 -0.173987  0.155244  0.173987  0.155244  \n",
       "3 -0.144249  0.131128  0.144249  0.131128  \n",
       "4  2.244534  2.244534  0.000000  0.000000  \n",
       "5  2.463212  2.463212  0.000000  0.000000  \n",
       "6  0.775816  0.775816  0.000000  0.000000  \n",
       "7  0.080093  0.080093  0.000000  0.000000  \n",
       "8  2.424386  2.502529  0.039601  0.038543  \n",
       "9  2.528028  2.602495  0.037714  0.036753  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggdf['min_yerr'] = aggdf['L2FC'] - aggdf['L2FC_min']\n",
    "aggdf['max_yerr'] = aggdf['L2FC_max'] - aggdf['L2FC']\n",
    "aggdf.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed06ef8c-3127-447f-8e59-096b953d7b83",
   "metadata": {},
   "source": [
    "## Visualize log2 fold changes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "063e30a3-e06d-49fc-8c17-79850df10987",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f400f57a050>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2101x500 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(aggdf, col='compound', y='L2FC', hue='genotype')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e63868e7-8847-43c7-b1e3-32976d1514a3",
   "metadata": {},
   "source": [
    "## Compare with DMS scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e3c26605-1b99-4fe2-829c-54ef6fe3079b",
   "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>compound</th>\n",
       "      <th>seq_type</th>\n",
       "      <th>Nham_aa</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>nt_seq</th>\n",
       "      <th>s</th>\n",
       "      <th>cscore</th>\n",
       "      <th>label</th>\n",
       "      <th>rclass</th>\n",
       "      <th>refined_class</th>\n",
       "      <th>aa_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ortho</td>\n",
       "      <td>1.0</td>\n",
       "      <td>FLILSLRDP</td>\n",
       "      <td>tttttgattttgtctttgagagatcca</td>\n",
       "      <td>-0.223979</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ortho</td>\n",
       "      <td>1.0</td>\n",
       "      <td>FLTLSLRDP</td>\n",
       "      <td>tttttgactttgtctttgagagatcca</td>\n",
       "      <td>0.127139</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ortho</td>\n",
       "      <td>2.0</td>\n",
       "      <td>FLALSFRDP</td>\n",
       "      <td>tttttggctttgtcttttagagatcca</td>\n",
       "      <td>0.370152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ortho</td>\n",
       "      <td>2.0</td>\n",
       "      <td>FLALSIRDP</td>\n",
       "      <td>tttttggctttgtctattagagatcca</td>\n",
       "      <td>-0.229898</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ortho</td>\n",
       "      <td>2.0</td>\n",
       "      <td>FLSLSFRDP</td>\n",
       "      <td>tttttgtctttgtcttttagagatcca</td>\n",
       "      <td>1.883312</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>intermediary</td>\n",
       "      <td>resistant</td>\n",
       "      <td>resistant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1251</th>\n",
       "      <td>none</td>\n",
       "      <td>single</td>\n",
       "      <td>1.0</td>\n",
       "      <td>TLVLSLRDP</td>\n",
       "      <td>acgttagttttatctttgagagatcca</td>\n",
       "      <td>-0.449800</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>slightly deleterious</td>\n",
       "      <td>slightly deleterious</td>\n",
       "      <td>slightly deleterious</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1252</th>\n",
       "      <td>none</td>\n",
       "      <td>single</td>\n",
       "      <td>1.0</td>\n",
       "      <td>TLVLSLRDP</td>\n",
       "      <td>actttagttttatctttgagagatcca</td>\n",
       "      <td>0.033715</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>slightly deleterious</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1253</th>\n",
       "      <td>none</td>\n",
       "      <td>single</td>\n",
       "      <td>1.0</td>\n",
       "      <td>VLVLSLRDP</td>\n",
       "      <td>gtgttagttttatctttgagagatcca</td>\n",
       "      <td>0.037623</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1254</th>\n",
       "      <td>none</td>\n",
       "      <td>single</td>\n",
       "      <td>1.0</td>\n",
       "      <td>WLVLSLRDP</td>\n",
       "      <td>tggttagttttatctttgagagatcca</td>\n",
       "      <td>0.070207</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1255</th>\n",
       "      <td>none</td>\n",
       "      <td>single</td>\n",
       "      <td>1.0</td>\n",
       "      <td>YLVLSLRDP</td>\n",
       "      <td>tatttagttttatctttgagagatcca</td>\n",
       "      <td>0.080017</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>WT-like</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1256 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           compound seq_type  Nham_aa     aa_seq                       nt_seq  \\\n",
       "0     anidulafungin    ortho      1.0  FLILSLRDP  tttttgattttgtctttgagagatcca   \n",
       "1     anidulafungin    ortho      1.0  FLTLSLRDP  tttttgactttgtctttgagagatcca   \n",
       "2     anidulafungin    ortho      2.0  FLALSFRDP  tttttggctttgtcttttagagatcca   \n",
       "3     anidulafungin    ortho      2.0  FLALSIRDP  tttttggctttgtctattagagatcca   \n",
       "4     anidulafungin    ortho      2.0  FLSLSFRDP  tttttgtctttgtcttttagagatcca   \n",
       "...             ...      ...      ...        ...                          ...   \n",
       "1251           none   single      1.0  TLVLSLRDP  acgttagttttatctttgagagatcca   \n",
       "1252           none   single      1.0  TLVLSLRDP  actttagttttatctttgagagatcca   \n",
       "1253           none   single      1.0  VLVLSLRDP  gtgttagttttatctttgagagatcca   \n",
       "1254           none   single      1.0  WLVLSLRDP  tggttagttttatctttgagagatcca   \n",
       "1255           none   single      1.0  YLVLSLRDP  tatttagttttatctttgagagatcca   \n",
       "\n",
       "             s  cscore  label                rclass         refined_class  \\\n",
       "0    -0.223979     1.0    1.0               WT-like               WT-like   \n",
       "1     0.127139     1.0    1.0               WT-like               WT-like   \n",
       "2     0.370152     1.0    1.0               WT-like               WT-like   \n",
       "3    -0.229898     1.0    1.0               WT-like               WT-like   \n",
       "4     1.883312     1.0    2.0          intermediary             resistant   \n",
       "...        ...     ...    ...                   ...                   ...   \n",
       "1251 -0.449800     1.0    1.0  slightly deleterious  slightly deleterious   \n",
       "1252  0.033715     1.0    0.0               WT-like               WT-like   \n",
       "1253  0.037623     1.0    0.0               WT-like               WT-like   \n",
       "1254  0.070207     1.0    0.0               WT-like               WT-like   \n",
       "1255  0.080017     1.0    0.0               WT-like               WT-like   \n",
       "\n",
       "                  aa_class  \n",
       "0                  WT-like  \n",
       "1                  WT-like  \n",
       "2                  WT-like  \n",
       "3                  WT-like  \n",
       "4                resistant  \n",
       "...                    ...  \n",
       "1251  slightly deleterious  \n",
       "1252  slightly deleterious  \n",
       "1253               WT-like  \n",
       "1254               WT-like  \n",
       "1255               WT-like  \n",
       "\n",
       "[1256 rows x 11 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "BY11nnkNT = pd.read_csv(DMS_data, index_col=0)\n",
    "BY11nnkNT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "0906b0e3-c59b-48f7-90b5-189ac1d6b87d",
   "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>compound</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>DMS_s</th>\n",
       "      <th>DMS_q1</th>\n",
       "      <th>DMS_q3</th>\n",
       "      <th>aa_class</th>\n",
       "      <th>min_xerr</th>\n",
       "      <th>max_xerr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>*LVLSLRDP</td>\n",
       "      <td>0.230635</td>\n",
       "      <td>0.230635</td>\n",
       "      <td>0.230635</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ALVLSLRDP</td>\n",
       "      <td>1.756300</td>\n",
       "      <td>1.620616</td>\n",
       "      <td>1.891984</td>\n",
       "      <td>intermediary</td>\n",
       "      <td>0.135684</td>\n",
       "      <td>0.135684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>DLVLSLRDP</td>\n",
       "      <td>0.047285</td>\n",
       "      <td>0.047285</td>\n",
       "      <td>0.047285</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>ELVLSLRDP</td>\n",
       "      <td>0.266907</td>\n",
       "      <td>0.266907</td>\n",
       "      <td>0.266907</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>F*VLSLRDP</td>\n",
       "      <td>0.165205</td>\n",
       "      <td>0.165205</td>\n",
       "      <td>0.165205</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>695</th>\n",
       "      <td>none</td>\n",
       "      <td>SLVLSLRDP</td>\n",
       "      <td>0.015051</td>\n",
       "      <td>-0.116203</td>\n",
       "      <td>0.035546</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.131254</td>\n",
       "      <td>0.020496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>696</th>\n",
       "      <td>none</td>\n",
       "      <td>TLVLSLRDP</td>\n",
       "      <td>-0.208043</td>\n",
       "      <td>-0.328922</td>\n",
       "      <td>-0.087164</td>\n",
       "      <td>slightly deleterious</td>\n",
       "      <td>0.120879</td>\n",
       "      <td>0.120879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>697</th>\n",
       "      <td>none</td>\n",
       "      <td>VLVLSLRDP</td>\n",
       "      <td>0.037623</td>\n",
       "      <td>0.037623</td>\n",
       "      <td>0.037623</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>698</th>\n",
       "      <td>none</td>\n",
       "      <td>WLVLSLRDP</td>\n",
       "      <td>0.070207</td>\n",
       "      <td>0.070207</td>\n",
       "      <td>0.070207</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>699</th>\n",
       "      <td>none</td>\n",
       "      <td>YLVLSLRDP</td>\n",
       "      <td>0.080017</td>\n",
       "      <td>0.080017</td>\n",
       "      <td>0.080017</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>700 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          compound     aa_seq     DMS_s    DMS_q1    DMS_q3  \\\n",
       "0    anidulafungin  *LVLSLRDP  0.230635  0.230635  0.230635   \n",
       "1    anidulafungin  ALVLSLRDP  1.756300  1.620616  1.891984   \n",
       "2    anidulafungin  DLVLSLRDP  0.047285  0.047285  0.047285   \n",
       "3    anidulafungin  ELVLSLRDP  0.266907  0.266907  0.266907   \n",
       "4    anidulafungin  F*VLSLRDP  0.165205  0.165205  0.165205   \n",
       "..             ...        ...       ...       ...       ...   \n",
       "695           none  SLVLSLRDP  0.015051 -0.116203  0.035546   \n",
       "696           none  TLVLSLRDP -0.208043 -0.328922 -0.087164   \n",
       "697           none  VLVLSLRDP  0.037623  0.037623  0.037623   \n",
       "698           none  WLVLSLRDP  0.070207  0.070207  0.070207   \n",
       "699           none  YLVLSLRDP  0.080017  0.080017  0.080017   \n",
       "\n",
       "                 aa_class  min_xerr  max_xerr  \n",
       "0                 WT-like  0.000000  0.000000  \n",
       "1            intermediary  0.135684  0.135684  \n",
       "2                 WT-like  0.000000  0.000000  \n",
       "3                 WT-like  0.000000  0.000000  \n",
       "4                 WT-like  0.000000  0.000000  \n",
       "..                    ...       ...       ...  \n",
       "695               WT-like  0.131254  0.020496  \n",
       "696  slightly deleterious  0.120879  0.120879  \n",
       "697               WT-like  0.000000  0.000000  \n",
       "698               WT-like  0.000000  0.000000  \n",
       "699               WT-like  0.000000  0.000000  \n",
       "\n",
       "[700 rows x 8 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DMS_aa = BY11nnkNT[(BY11nnkNT.Nham_aa <= 1)\n",
    "                   & (BY11nnkNT.cscore < 3)\n",
    "                  ].groupby(['compound','aa_seq'])[['s','aa_class']\n",
    "                                                  ].agg(DMS_s = ('s','median'),\n",
    "                                                        DMS_q1 = ('s', lambda x: np.quantile(x, 0.25)),\n",
    "                                                        DMS_q3 = ('s', lambda x: np.quantile(x, 0.75)),\n",
    "                                                        aa_class = ('aa_class','first')\n",
    "                                                       ).reset_index()\n",
    "DMS_aa['min_xerr'] = DMS_aa['DMS_s'] - DMS_aa['DMS_q1']\n",
    "DMS_aa['max_xerr'] = DMS_aa['DMS_q3'] - DMS_aa['DMS_s']\n",
    "DMS_aa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "de0960aa-128f-4ab1-b38e-cb02afd186df",
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>compound</th>\n",
       "      <th>auc</th>\n",
       "      <th>auc_min</th>\n",
       "      <th>auc_max</th>\n",
       "      <th>L2FC</th>\n",
       "      <th>L2FC_min</th>\n",
       "      <th>L2FC_max</th>\n",
       "      <th>min_yerr</th>\n",
       "      <th>max_yerr</th>\n",
       "      <th>DMS_s</th>\n",
       "      <th>DMS_q1</th>\n",
       "      <th>DMS_q3</th>\n",
       "      <th>aa_class</th>\n",
       "      <th>min_xerr</th>\n",
       "      <th>max_xerr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>25.95425</td>\n",
       "      <td>22.6925</td>\n",
       "      <td>29.2160</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.193755</td>\n",
       "      <td>0.170788</td>\n",
       "      <td>0.193755</td>\n",
       "      <td>0.170788</td>\n",
       "      <td>-0.062763</td>\n",
       "      <td>-0.133165</td>\n",
       "      <td>0.063575</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.070402</td>\n",
       "      <td>0.126338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>23.75450</td>\n",
       "      <td>23.2540</td>\n",
       "      <td>24.2550</td>\n",
       "      <td>-0.127770</td>\n",
       "      <td>-0.158492</td>\n",
       "      <td>-0.097689</td>\n",
       "      <td>0.030722</td>\n",
       "      <td>0.030081</td>\n",
       "      <td>-0.062763</td>\n",
       "      <td>-0.133165</td>\n",
       "      <td>0.063575</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.070402</td>\n",
       "      <td>0.126338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>22.90100</td>\n",
       "      <td>21.9170</td>\n",
       "      <td>23.8850</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.063360</td>\n",
       "      <td>0.060694</td>\n",
       "      <td>0.063360</td>\n",
       "      <td>0.060694</td>\n",
       "      <td>0.021054</td>\n",
       "      <td>-0.138335</td>\n",
       "      <td>0.073044</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.159389</td>\n",
       "      <td>0.051989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>21.04525</td>\n",
       "      <td>20.9890</td>\n",
       "      <td>21.1015</td>\n",
       "      <td>-0.121916</td>\n",
       "      <td>-0.125777</td>\n",
       "      <td>-0.118065</td>\n",
       "      <td>0.003861</td>\n",
       "      <td>0.003851</td>\n",
       "      <td>0.021054</td>\n",
       "      <td>-0.138335</td>\n",
       "      <td>0.073044</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.159389</td>\n",
       "      <td>0.051989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>24.07575</td>\n",
       "      <td>21.3405</td>\n",
       "      <td>26.8110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.173987</td>\n",
       "      <td>0.155244</td>\n",
       "      <td>0.173987</td>\n",
       "      <td>0.155244</td>\n",
       "      <td>0.005049</td>\n",
       "      <td>-0.134869</td>\n",
       "      <td>0.104351</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.139918</td>\n",
       "      <td>0.099302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>22.51675</td>\n",
       "      <td>22.4110</td>\n",
       "      <td>22.6225</td>\n",
       "      <td>-0.096582</td>\n",
       "      <td>-0.103374</td>\n",
       "      <td>-0.089822</td>\n",
       "      <td>0.006792</td>\n",
       "      <td>0.006760</td>\n",
       "      <td>0.005049</td>\n",
       "      <td>-0.134869</td>\n",
       "      <td>0.104351</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.139918</td>\n",
       "      <td>0.099302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>134.43275</td>\n",
       "      <td>121.6415</td>\n",
       "      <td>147.2240</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.144249</td>\n",
       "      <td>0.131128</td>\n",
       "      <td>0.144249</td>\n",
       "      <td>0.131128</td>\n",
       "      <td>0.010453</td>\n",
       "      <td>-0.194294</td>\n",
       "      <td>0.014599</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.204747</td>\n",
       "      <td>0.004146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>146.09575</td>\n",
       "      <td>144.9465</td>\n",
       "      <td>147.2450</td>\n",
       "      <td>0.120030</td>\n",
       "      <td>0.108636</td>\n",
       "      <td>0.131334</td>\n",
       "      <td>0.011394</td>\n",
       "      <td>0.011304</td>\n",
       "      <td>0.010453</td>\n",
       "      <td>-0.194294</td>\n",
       "      <td>0.014599</td>\n",
       "      <td>WT-like</td>\n",
       "      <td>0.204747</td>\n",
       "      <td>0.004146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>122.99300</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>122.9930</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.808480</td>\n",
       "      <td>1.808480</td>\n",
       "      <td>1.808480</td>\n",
       "      <td>intermediary</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>126.28600</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>126.2860</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.865695</td>\n",
       "      <td>1.865695</td>\n",
       "      <td>1.865695</td>\n",
       "      <td>resistant</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  genotype     aa_seq       compound        auc   auc_min   auc_max      L2FC  \\\n",
       "0       BY  FLVLSLRDP  anidulafungin   25.95425   22.6925   29.2160  0.000000   \n",
       "1       WT  FLVLSLRDP  anidulafungin   23.75450   23.2540   24.2550 -0.127770   \n",
       "2       BY  FLVLSLRDP    caspofungin   22.90100   21.9170   23.8850  0.000000   \n",
       "3       WT  FLVLSLRDP    caspofungin   21.04525   20.9890   21.1015 -0.121916   \n",
       "4       BY  FLVLSLRDP     micafungin   24.07575   21.3405   26.8110  0.000000   \n",
       "5       WT  FLVLSLRDP     micafungin   22.51675   22.4110   22.6225 -0.096582   \n",
       "6       BY  FLVLSLRDP           none  134.43275  121.6415  147.2240  0.000000   \n",
       "7       WT  FLVLSLRDP           none  146.09575  144.9465  147.2450  0.120030   \n",
       "8    D646W  FLVLSLRWP  anidulafungin  122.99300  122.9930  122.9930  2.244534   \n",
       "9    D646W  FLVLSLRWP    caspofungin  126.28600  126.2860  126.2860  2.463212   \n",
       "\n",
       "   L2FC_min  L2FC_max  min_yerr  max_yerr     DMS_s    DMS_q1    DMS_q3  \\\n",
       "0 -0.193755  0.170788  0.193755  0.170788 -0.062763 -0.133165  0.063575   \n",
       "1 -0.158492 -0.097689  0.030722  0.030081 -0.062763 -0.133165  0.063575   \n",
       "2 -0.063360  0.060694  0.063360  0.060694  0.021054 -0.138335  0.073044   \n",
       "3 -0.125777 -0.118065  0.003861  0.003851  0.021054 -0.138335  0.073044   \n",
       "4 -0.173987  0.155244  0.173987  0.155244  0.005049 -0.134869  0.104351   \n",
       "5 -0.103374 -0.089822  0.006792  0.006760  0.005049 -0.134869  0.104351   \n",
       "6 -0.144249  0.131128  0.144249  0.131128  0.010453 -0.194294  0.014599   \n",
       "7  0.108636  0.131334  0.011394  0.011304  0.010453 -0.194294  0.014599   \n",
       "8  2.244534  2.244534  0.000000  0.000000  1.808480  1.808480  1.808480   \n",
       "9  2.463212  2.463212  0.000000  0.000000  1.865695  1.865695  1.865695   \n",
       "\n",
       "       aa_class  min_xerr  max_xerr  \n",
       "0       WT-like  0.070402  0.126338  \n",
       "1       WT-like  0.070402  0.126338  \n",
       "2       WT-like  0.159389  0.051989  \n",
       "3       WT-like  0.159389  0.051989  \n",
       "4       WT-like  0.139918  0.099302  \n",
       "5       WT-like  0.139918  0.099302  \n",
       "6       WT-like  0.204747  0.004146  \n",
       "7       WT-like  0.204747  0.004146  \n",
       "8  intermediary  0.000000  0.000000  \n",
       "9     resistant  0.000000  0.000000  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corrdf = aggdf.merge(right=DMS_aa, on=['compound', 'aa_seq'])\n",
    "corrdf.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "42df8035-f4ba-45b7-b987-d89872bb4394",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(rc = {\n",
    "              'font.family':'Arial',\n",
    "              'font.size':8,\n",
    "              'legend.title_fontsize':8, 'legend.fontsize':8,\n",
    "              'axes.labelsize':8,'axes.titlesize':8,\n",
    "              'xtick.labelsize':8, 'ytick.labelsize':8,\n",
    "              'xtick.major.pad':2, 'ytick.major.pad':2,\n",
    "              'xtick.bottom': True, 'ytick.left': True,\n",
    "              'xtick.major.size':2, 'ytick.major.size':2,\n",
    "             },\n",
    "       style='ticks')\n",
    "\n",
    "lcomp = ['none','anidulafungin','caspofungin','micafungin']\n",
    "comp_dict = {'caspofungin':'Caspofungin',\n",
    "             'micafungin':'Micafungin',\n",
    "             'anidulafungin':'Anidulafungin',\n",
    "             'none':'Control'\n",
    "            }\n",
    "classes = ['resistant','intermediary','WT-like','slightly deleterious','deleterious']\n",
    "class_palette = ['#C75DAB','#D691C1','#F1F1F1','#42B7B9','#009B9E']\n",
    "class_cmap = dict(zip(classes, class_palette))\n",
    "\n",
    "# Initialize list to save linear regression parameters for each condition\n",
    "reglist=[]\n",
    "\n",
    "grid = sns.lmplot(corrdf, x='DMS_s', y='L2FC', col='compound', col_order=lcomp,\n",
    "                  markers='none', # dots are redrawn with color = resistance class\n",
    "                  # note: i've tried coloring the markers using scatter_kws, doesn't work\n",
    "                  facet_kws={'despine':False},\n",
    "                  line_kws={'color': 'lightgrey'},\n",
    "                  height=2, aspect=.9\n",
    "                 )\n",
    "\n",
    "for i,c in enumerate(lcomp):\n",
    "    graphdf = corrdf[corrdf.compound == c]\n",
    "    grid.axes[0][i].title.set_text(comp_dict[c])\n",
    "    \n",
    "    # Drawing error bars\n",
    "    grid.axes[0][i].errorbar(x=graphdf.DMS_s, y=graphdf.L2FC,\n",
    "                   xerr=[graphdf.min_xerr, graphdf.max_xerr],\n",
    "                   yerr=[graphdf.min_yerr, graphdf.max_yerr],\n",
    "                   fmt='none', ecolor='grey' # only error bars, no dots\n",
    "                  )\n",
    "    \n",
    "    # Drawing dots, colored by their classification (resistant, WT-like, etc)\n",
    "    grid.axes[0][i].scatter(x=graphdf.DMS_s, y=graphdf.L2FC,\n",
    "                            c=graphdf.aa_class.map(class_cmap),\n",
    "                            ec='grey',\n",
    "                            zorder=100\n",
    "                           )\n",
    "    \n",
    "    # Calculate and display Spearman correlation coefficient\n",
    "    from scipy import stats\n",
    "    sr, sp = stats.spearmanr(graphdf.DMS_s, graphdf.L2FC)\n",
    "    grid.axes[0][i].text(.5, -0.5, fr'$\\rho$ = {sr:.2f}'+\n",
    "                         '\\n$\\it{p}$-val = '+f'{sp:.1e}', ha='left')\n",
    "    \n",
    "    # I am leaving some unused code below to compare linear regression with seaborn (statsmodels OLS) and sklearn\n",
    "    # but ultimately I use the regression done by seaborn (statsmodels OLS)\n",
    "    # then I fetch the slope and intercept from seaborn drawn lines and append them to a dataframe\n",
    "    \n",
    "    # Perform linear regression using sklearn package\n",
    "    from sklearn.model_selection import train_test_split\n",
    "    from sklearn.linear_model import LinearRegression\n",
    "    X = graphdf.DMS_s.values.reshape(-1,1)\n",
    "    y = graphdf.L2FC.values.reshape(-1,1)\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=6) # train on 20% of dataset\n",
    "    reg = LinearRegression().fit(X_train, y_train)\n",
    "    \n",
    "    # Extract slope and intercept from sklearn Linear Regression model\n",
    "    slopeSK = reg.coef_.flatten()[0]\n",
    "    interceptSK = reg.intercept_[0]\n",
    "    \n",
    "    # Extract slope and intercept from seaborn lmplot (statsmodels OLS)\n",
    "    slope, intercept, r_value, p_value, std_err = stats.linregress(x=grid.axes[0][i].get_lines()[0].get_xdata(),\n",
    "                                                                     y=grid.axes[0][i].get_lines()[0].get_ydata())\n",
    "    \n",
    "    # Save slope and intercept for each condition in a list -> append to master list\n",
    "    reglist.append([c, slope, intercept])\n",
    "    \n",
    "    # Add predicted data on plot and display equation\n",
    "    #y_pred = reg.predict(X_test)\n",
    "    #grid.axes[0][i].plot(X_test, y_pred, color='blue', lw=2) # This highlights the train/test split\n",
    "    #grid.axes[0][i].axline((0, interceptSK), slope=slopeSK, color='blue', lw=2)\n",
    "    #grid.axes[0][i].text(-1, -1, f'y = {slope:.2f}x {intercept:+.2f}', ha='left', color='grey')\n",
    "    #grid.axes[0][i].text(-1, 3, f'y = {slopeSK:.2f}x {interceptSK:+.2f}', ha='left', color='blue')\n",
    "\n",
    "# Highlighting interesting cases\n",
    "int_mut = ['V641W','L642K']\n",
    "deviating_df = corrdf[(corrdf.compound == 'micafungin')\n",
    "                       & (corrdf.genotype.isin(int_mut))]\n",
    "grid.axes[0][3].scatter(x=deviating_df.DMS_s, y=deviating_df.L2FC,\n",
    "                        c=deviating_df.aa_class.map(class_cmap),\n",
    "                        ec='k', lw=2,\n",
    "                        zorder=200\n",
    "                       )\n",
    "# .. and label them\n",
    "for i, m in enumerate(int_mut):\n",
    "    grid.axes[0][3].annotate(m, (deviating_df.DMS_s.values[i], # x\n",
    "                                 deviating_df.L2FC.values[i] # y\n",
    "                                ),\n",
    "                             xytext = (-4,0), # distance of text label from xy coords\n",
    "                             textcoords = 'offset fontsize' # xytext coords given in fontsize\n",
    "                            )\n",
    "\n",
    "grid.set_axis_labels('DMS selection coefficient', 'Log2FC(AUC)')\n",
    "\n",
    "# Convert list of list to dataframe\n",
    "regdf = pd.DataFrame(reglist, columns=['compound', 'slope', 'intercept'])\n",
    "\n",
    "plt.savefig(f'{graph_outpath}/validations.svg', format='svg', dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8979637a-292f-4174-90d5-ce6c2841361f",
   "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>compound</th>\n",
       "      <th>slope</th>\n",
       "      <th>intercept</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>none</td>\n",
       "      <td>0.392665</td>\n",
       "      <td>0.092369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>1.107270</td>\n",
       "      <td>0.239401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>caspofungin</td>\n",
       "      <td>1.209763</td>\n",
       "      <td>0.245074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>micafungin</td>\n",
       "      <td>0.959684</td>\n",
       "      <td>0.698749</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        compound     slope  intercept\n",
       "0           none  0.392665   0.092369\n",
       "1  anidulafungin  1.107270   0.239401\n",
       "2    caspofungin  1.209763   0.245074\n",
       "3     micafungin  0.959684   0.698749"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2eaa5ad4-2920-4fe0-8cd2-048474b233c5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "def get_estimate_DMS_score(v, r, df):\n",
    "    # Merge row and dataframe containing linear regression parameters\n",
    "    merg = pd.merge(r.to_frame().T, df, how='left', on='compound')\n",
    "    \n",
    "    # Get slope and intercept for this condition\n",
    "    s = merg.slope[0]\n",
    "    i = merg.intercept[0]\n",
    "    \n",
    "    # Return corrected value\n",
    "    return (v-i)/s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "f1a7f945-6968-4db3-8dbc-7ff50c26a51d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['F639C', 'L640D', 'L642G', 'P647N', 'P647Q']\n"
     ]
    }
   ],
   "source": [
    "# Mutants for which we need to estimate the DMS score\n",
    "mut_no_DMS = [x for x in aggdf.genotype.unique() if x not in corrdf.genotype.unique()]\n",
    "print(mut_no_DMS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "fd4fbd1f-4499-4284-a594-f07a071bb8ba",
   "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>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>compound</th>\n",
       "      <th>auc</th>\n",
       "      <th>auc_min</th>\n",
       "      <th>auc_max</th>\n",
       "      <th>L2FC</th>\n",
       "      <th>L2FC_min</th>\n",
       "      <th>L2FC_max</th>\n",
       "      <th>min_yerr</th>\n",
       "      <th>max_yerr</th>\n",
       "      <th>estimated_s</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>143.19975</td>\n",
       "      <td>139.3225</td>\n",
       "      <td>147.0770</td>\n",
       "      <td>2.463986</td>\n",
       "      <td>2.424386</td>\n",
       "      <td>2.502529</td>\n",
       "      <td>0.039601</td>\n",
       "      <td>0.038543</td>\n",
       "      <td>2.009071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>135.58750</td>\n",
       "      <td>132.0890</td>\n",
       "      <td>139.0860</td>\n",
       "      <td>2.565742</td>\n",
       "      <td>2.528028</td>\n",
       "      <td>2.602495</td>\n",
       "      <td>0.037714</td>\n",
       "      <td>0.036753</td>\n",
       "      <td>1.918283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>142.16350</td>\n",
       "      <td>135.8590</td>\n",
       "      <td>148.4680</td>\n",
       "      <td>2.561898</td>\n",
       "      <td>2.496457</td>\n",
       "      <td>2.624499</td>\n",
       "      <td>0.065441</td>\n",
       "      <td>0.062601</td>\n",
       "      <td>1.941420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>F639C</td>\n",
       "      <td>CLVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>148.17825</td>\n",
       "      <td>148.1085</td>\n",
       "      <td>148.2480</td>\n",
       "      <td>0.140449</td>\n",
       "      <td>0.139770</td>\n",
       "      <td>0.141128</td>\n",
       "      <td>0.000679</td>\n",
       "      <td>0.000679</td>\n",
       "      <td>0.122445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>L640D</td>\n",
       "      <td>FDVLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>30.37775</td>\n",
       "      <td>30.0705</td>\n",
       "      <td>30.6850</td>\n",
       "      <td>0.227044</td>\n",
       "      <td>0.212378</td>\n",
       "      <td>0.241563</td>\n",
       "      <td>0.014666</td>\n",
       "      <td>0.014519</td>\n",
       "      <td>-0.011160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>L640D</td>\n",
       "      <td>FDVLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>25.44800</td>\n",
       "      <td>22.6865</td>\n",
       "      <td>28.2095</td>\n",
       "      <td>0.152142</td>\n",
       "      <td>-0.013577</td>\n",
       "      <td>0.300770</td>\n",
       "      <td>0.165718</td>\n",
       "      <td>0.148629</td>\n",
       "      <td>-0.076818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>L640D</td>\n",
       "      <td>FDVLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>62.52275</td>\n",
       "      <td>61.7745</td>\n",
       "      <td>63.2710</td>\n",
       "      <td>1.376800</td>\n",
       "      <td>1.359431</td>\n",
       "      <td>1.393964</td>\n",
       "      <td>0.017370</td>\n",
       "      <td>0.017163</td>\n",
       "      <td>0.706536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>L640D</td>\n",
       "      <td>FDVLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>145.69875</td>\n",
       "      <td>144.7380</td>\n",
       "      <td>146.6595</td>\n",
       "      <td>0.116104</td>\n",
       "      <td>0.106559</td>\n",
       "      <td>0.125586</td>\n",
       "      <td>0.009545</td>\n",
       "      <td>0.009482</td>\n",
       "      <td>0.060445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>L642G</td>\n",
       "      <td>FLVGSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>24.61900</td>\n",
       "      <td>24.6190</td>\n",
       "      <td>24.6190</td>\n",
       "      <td>-0.076199</td>\n",
       "      <td>-0.076199</td>\n",
       "      <td>-0.076199</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.285025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>L642G</td>\n",
       "      <td>FLVGSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>26.25650</td>\n",
       "      <td>26.2565</td>\n",
       "      <td>26.2565</td>\n",
       "      <td>0.197264</td>\n",
       "      <td>0.197264</td>\n",
       "      <td>0.197264</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.039520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>L642G</td>\n",
       "      <td>FLVGSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>24.19700</td>\n",
       "      <td>24.1970</td>\n",
       "      <td>24.1970</td>\n",
       "      <td>0.007247</td>\n",
       "      <td>0.007247</td>\n",
       "      <td>0.007247</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.720551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>L642G</td>\n",
       "      <td>FLVGSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>143.37700</td>\n",
       "      <td>143.3770</td>\n",
       "      <td>143.3770</td>\n",
       "      <td>0.092929</td>\n",
       "      <td>0.092929</td>\n",
       "      <td>0.092929</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>L642K</td>\n",
       "      <td>FLVKSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>66.36800</td>\n",
       "      <td>66.3680</td>\n",
       "      <td>66.3680</td>\n",
       "      <td>1.354517</td>\n",
       "      <td>1.354517</td>\n",
       "      <td>1.354517</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.007085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>L642K</td>\n",
       "      <td>FLVKSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>129.87600</td>\n",
       "      <td>129.8760</td>\n",
       "      <td>129.8760</td>\n",
       "      <td>2.503652</td>\n",
       "      <td>2.503652</td>\n",
       "      <td>2.503652</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.866959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>L642K</td>\n",
       "      <td>FLVKSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>107.28350</td>\n",
       "      <td>107.2835</td>\n",
       "      <td>107.2835</td>\n",
       "      <td>2.155776</td>\n",
       "      <td>2.155776</td>\n",
       "      <td>2.155776</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.518236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>L642K</td>\n",
       "      <td>FLVKSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>143.80500</td>\n",
       "      <td>143.8050</td>\n",
       "      <td>143.8050</td>\n",
       "      <td>0.097229</td>\n",
       "      <td>0.097229</td>\n",
       "      <td>0.097229</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.012377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>P647N</td>\n",
       "      <td>FLVLSLRDN</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>133.79375</td>\n",
       "      <td>132.6275</td>\n",
       "      <td>134.9600</td>\n",
       "      <td>2.365968</td>\n",
       "      <td>2.353337</td>\n",
       "      <td>2.378489</td>\n",
       "      <td>0.012631</td>\n",
       "      <td>0.012521</td>\n",
       "      <td>1.920549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>P647N</td>\n",
       "      <td>FLVLSLRDN</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>128.85275</td>\n",
       "      <td>128.7080</td>\n",
       "      <td>128.9975</td>\n",
       "      <td>2.492241</td>\n",
       "      <td>2.490619</td>\n",
       "      <td>2.493861</td>\n",
       "      <td>0.001622</td>\n",
       "      <td>0.001620</td>\n",
       "      <td>1.857526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62</th>\n",
       "      <td>P647N</td>\n",
       "      <td>FLVLSLRDN</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>133.65450</td>\n",
       "      <td>128.1990</td>\n",
       "      <td>139.1100</td>\n",
       "      <td>2.472856</td>\n",
       "      <td>2.412732</td>\n",
       "      <td>2.530573</td>\n",
       "      <td>0.060123</td>\n",
       "      <td>0.057718</td>\n",
       "      <td>1.848636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>P647N</td>\n",
       "      <td>FLVLSLRDN</td>\n",
       "      <td>none</td>\n",
       "      <td>143.87500</td>\n",
       "      <td>142.4025</td>\n",
       "      <td>145.3475</td>\n",
       "      <td>0.097931</td>\n",
       "      <td>0.083090</td>\n",
       "      <td>0.112622</td>\n",
       "      <td>0.014841</td>\n",
       "      <td>0.014690</td>\n",
       "      <td>0.014165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>P647Q</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>145.50350</td>\n",
       "      <td>145.5035</td>\n",
       "      <td>145.5035</td>\n",
       "      <td>2.487011</td>\n",
       "      <td>2.487011</td>\n",
       "      <td>2.487011</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.029866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>65</th>\n",
       "      <td>P647Q</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>137.14600</td>\n",
       "      <td>137.1460</td>\n",
       "      <td>137.1460</td>\n",
       "      <td>2.582230</td>\n",
       "      <td>2.582230</td>\n",
       "      <td>2.582230</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.931912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>P647Q</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>145.07750</td>\n",
       "      <td>145.0775</td>\n",
       "      <td>145.0775</td>\n",
       "      <td>2.591171</td>\n",
       "      <td>2.591171</td>\n",
       "      <td>2.591171</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.971922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>P647Q</td>\n",
       "      <td>FLVLSLRDQ</td>\n",
       "      <td>none</td>\n",
       "      <td>143.59500</td>\n",
       "      <td>143.5950</td>\n",
       "      <td>143.5950</td>\n",
       "      <td>0.095121</td>\n",
       "      <td>0.095121</td>\n",
       "      <td>0.095121</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>V641W</td>\n",
       "      <td>FLWLSLRDP</td>\n",
       "      <td>anidulafungin</td>\n",
       "      <td>92.92925</td>\n",
       "      <td>87.8415</td>\n",
       "      <td>98.0170</td>\n",
       "      <td>1.840162</td>\n",
       "      <td>1.758932</td>\n",
       "      <td>1.917061</td>\n",
       "      <td>0.081230</td>\n",
       "      <td>0.076899</td>\n",
       "      <td>1.445682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>V641W</td>\n",
       "      <td>FLWLSLRDP</td>\n",
       "      <td>caspofungin</td>\n",
       "      <td>64.96775</td>\n",
       "      <td>61.4600</td>\n",
       "      <td>68.4755</td>\n",
       "      <td>1.504313</td>\n",
       "      <td>1.424237</td>\n",
       "      <td>1.580177</td>\n",
       "      <td>0.080076</td>\n",
       "      <td>0.075864</td>\n",
       "      <td>1.040897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>V641W</td>\n",
       "      <td>FLWLSLRDP</td>\n",
       "      <td>micafungin</td>\n",
       "      <td>71.27425</td>\n",
       "      <td>69.7260</td>\n",
       "      <td>72.8225</td>\n",
       "      <td>1.565800</td>\n",
       "      <td>1.534116</td>\n",
       "      <td>1.596804</td>\n",
       "      <td>0.031684</td>\n",
       "      <td>0.031003</td>\n",
       "      <td>0.903476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>V641W</td>\n",
       "      <td>FLWLSLRDP</td>\n",
       "      <td>none</td>\n",
       "      <td>147.61350</td>\n",
       "      <td>144.7250</td>\n",
       "      <td>150.5020</td>\n",
       "      <td>0.134940</td>\n",
       "      <td>0.106430</td>\n",
       "      <td>0.162898</td>\n",
       "      <td>0.028511</td>\n",
       "      <td>0.027958</td>\n",
       "      <td>0.108415</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    genotype     aa_seq       compound        auc   auc_min   auc_max  \\\n",
       "8      F639C  CLVLSLRDP  anidulafungin  143.19975  139.3225  147.0770   \n",
       "9      F639C  CLVLSLRDP    caspofungin  135.58750  132.0890  139.0860   \n",
       "10     F639C  CLVLSLRDP     micafungin  142.16350  135.8590  148.4680   \n",
       "11     F639C  CLVLSLRDP           none  148.17825  148.1085  148.2480   \n",
       "32     L640D  FDVLSLRDP  anidulafungin   30.37775   30.0705   30.6850   \n",
       "33     L640D  FDVLSLRDP    caspofungin   25.44800   22.6865   28.2095   \n",
       "34     L640D  FDVLSLRDP     micafungin   62.52275   61.7745   63.2710   \n",
       "35     L640D  FDVLSLRDP           none  145.69875  144.7380  146.6595   \n",
       "48     L642G  FLVGSLRDP  anidulafungin   24.61900   24.6190   24.6190   \n",
       "49     L642G  FLVGSLRDP    caspofungin   26.25650   26.2565   26.2565   \n",
       "50     L642G  FLVGSLRDP     micafungin   24.19700   24.1970   24.1970   \n",
       "51     L642G  FLVGSLRDP           none  143.37700  143.3770  143.3770   \n",
       "52     L642K  FLVKSLRDP  anidulafungin   66.36800   66.3680   66.3680   \n",
       "53     L642K  FLVKSLRDP    caspofungin  129.87600  129.8760  129.8760   \n",
       "54     L642K  FLVKSLRDP     micafungin  107.28350  107.2835  107.2835   \n",
       "55     L642K  FLVKSLRDP           none  143.80500  143.8050  143.8050   \n",
       "60     P647N  FLVLSLRDN  anidulafungin  133.79375  132.6275  134.9600   \n",
       "61     P647N  FLVLSLRDN    caspofungin  128.85275  128.7080  128.9975   \n",
       "62     P647N  FLVLSLRDN     micafungin  133.65450  128.1990  139.1100   \n",
       "63     P647N  FLVLSLRDN           none  143.87500  142.4025  145.3475   \n",
       "64     P647Q  FLVLSLRDQ  anidulafungin  145.50350  145.5035  145.5035   \n",
       "65     P647Q  FLVLSLRDQ    caspofungin  137.14600  137.1460  137.1460   \n",
       "66     P647Q  FLVLSLRDQ     micafungin  145.07750  145.0775  145.0775   \n",
       "67     P647Q  FLVLSLRDQ           none  143.59500  143.5950  143.5950   \n",
       "112    V641W  FLWLSLRDP  anidulafungin   92.92925   87.8415   98.0170   \n",
       "113    V641W  FLWLSLRDP    caspofungin   64.96775   61.4600   68.4755   \n",
       "114    V641W  FLWLSLRDP     micafungin   71.27425   69.7260   72.8225   \n",
       "115    V641W  FLWLSLRDP           none  147.61350  144.7250  150.5020   \n",
       "\n",
       "         L2FC  L2FC_min  L2FC_max  min_yerr  max_yerr  estimated_s  \n",
       "8    2.463986  2.424386  2.502529  0.039601  0.038543     2.009071  \n",
       "9    2.565742  2.528028  2.602495  0.037714  0.036753     1.918283  \n",
       "10   2.561898  2.496457  2.624499  0.065441  0.062601     1.941420  \n",
       "11   0.140449  0.139770  0.141128  0.000679  0.000679     0.122445  \n",
       "32   0.227044  0.212378  0.241563  0.014666  0.014519    -0.011160  \n",
       "33   0.152142 -0.013577  0.300770  0.165718  0.148629    -0.076818  \n",
       "34   1.376800  1.359431  1.393964  0.017370  0.017163     0.706536  \n",
       "35   0.116104  0.106559  0.125586  0.009545  0.009482     0.060445  \n",
       "48  -0.076199 -0.076199 -0.076199  0.000000  0.000000    -0.285025  \n",
       "49   0.197264  0.197264  0.197264  0.000000  0.000000    -0.039520  \n",
       "50   0.007247  0.007247  0.007247  0.000000  0.000000    -0.720551  \n",
       "51   0.092929  0.092929  0.092929  0.000000  0.000000     0.001425  \n",
       "52   1.354517  1.354517  1.354517  0.000000  0.000000     1.007085  \n",
       "53   2.503652  2.503652  2.503652  0.000000  0.000000     1.866959  \n",
       "54   2.155776  2.155776  2.155776  0.000000  0.000000     1.518236  \n",
       "55   0.097229  0.097229  0.097229  0.000000  0.000000     0.012377  \n",
       "60   2.365968  2.353337  2.378489  0.012631  0.012521     1.920549  \n",
       "61   2.492241  2.490619  2.493861  0.001622  0.001620     1.857526  \n",
       "62   2.472856  2.412732  2.530573  0.060123  0.057718     1.848636  \n",
       "63   0.097931  0.083090  0.112622  0.014841  0.014690     0.014165  \n",
       "64   2.487011  2.487011  2.487011  0.000000  0.000000     2.029866  \n",
       "65   2.582230  2.582230  2.582230  0.000000  0.000000     1.931912  \n",
       "66   2.591171  2.591171  2.591171  0.000000  0.000000     1.971922  \n",
       "67   0.095121  0.095121  0.095121  0.000000  0.000000     0.007008  \n",
       "112  1.840162  1.758932  1.917061  0.081230  0.076899     1.445682  \n",
       "113  1.504313  1.424237  1.580177  0.080076  0.075864     1.040897  \n",
       "114  1.565800  1.534116  1.596804  0.031684  0.031003     0.903476  \n",
       "115  0.134940  0.106430  0.162898  0.028511  0.027958     0.108415  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggdf['estimated_s'] = aggdf.apply(lambda row: get_estimate_DMS_score(row.L2FC, row, regdf), axis=1)\n",
    "rescued_df = aggdf[aggdf.genotype.isin(mut_no_DMS+['V641W','L642K'])]\n",
    "rescued_df.to_csv(f'{aggdata_outpath}/validation_DMS_missing_estimates.csv')\n",
    "rescued_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "67cc7129-bca8-40b4-a0b9-caf3d2f74aa8",
   "metadata": {
    "tags": []
   },
   "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>compound</th>\n",
       "      <th>genotype</th>\n",
       "      <th>aa_seq</th>\n",
       "      <th>anidulafungin</th>\n",
       "      <th>caspofungin</th>\n",
       "      <th>micafungin</th>\n",
       "      <th>none</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BY</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>D646W</td>\n",
       "      <td>FLVLSLRWP</td>\n",
       "      <td>2.244534</td>\n",
       "      <td>2.463212</td>\n",
       "      <td>0.775816</td>\n",
       "      <td>0.080093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>F639D</td>\n",
       "      <td>DLVLSLRDP</td>\n",
       "      <td>0.324003</td>\n",
       "      <td>2.420326</td>\n",
       "      <td>2.429784</td>\n",
       "      <td>0.051189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F639E</td>\n",
       "      <td>ELVLSLRDP</td>\n",
       "      <td>-0.009970</td>\n",
       "      <td>2.632952</td>\n",
       "      <td>2.166640</td>\n",
       "      <td>0.172062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>F639Q</td>\n",
       "      <td>QLVLSLRDP</td>\n",
       "      <td>-0.257678</td>\n",
       "      <td>2.486612</td>\n",
       "      <td>-0.023516</td>\n",
       "      <td>0.117442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>F639S</td>\n",
       "      <td>SLVLSLRDP</td>\n",
       "      <td>2.439779</td>\n",
       "      <td>2.538386</td>\n",
       "      <td>2.586103</td>\n",
       "      <td>0.107717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>F639V</td>\n",
       "      <td>VLVLSLRDP</td>\n",
       "      <td>2.366154</td>\n",
       "      <td>2.502741</td>\n",
       "      <td>2.528672</td>\n",
       "      <td>0.057305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>L640G</td>\n",
       "      <td>FGVLSLRDP</td>\n",
       "      <td>1.827176</td>\n",
       "      <td>0.694793</td>\n",
       "      <td>1.621450</td>\n",
       "      <td>0.109009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>L640K</td>\n",
       "      <td>FKVLSLRDP</td>\n",
       "      <td>0.325101</td>\n",
       "      <td>0.121691</td>\n",
       "      <td>0.585766</td>\n",
       "      <td>-0.380418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>L642F</td>\n",
       "      <td>FLVFSLRDP</td>\n",
       "      <td>2.446581</td>\n",
       "      <td>2.581567</td>\n",
       "      <td>2.532537</td>\n",
       "      <td>0.121088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>L642K</td>\n",
       "      <td>FLVKSLRDP</td>\n",
       "      <td>1.354517</td>\n",
       "      <td>2.503652</td>\n",
       "      <td>2.155776</td>\n",
       "      <td>0.097229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>L642Y</td>\n",
       "      <td>FLVYSLRDP</td>\n",
       "      <td>2.517802</td>\n",
       "      <td>2.533451</td>\n",
       "      <td>2.650863</td>\n",
       "      <td>0.148094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>R645D</td>\n",
       "      <td>FLVLSLDDP</td>\n",
       "      <td>0.745492</td>\n",
       "      <td>0.868791</td>\n",
       "      <td>2.563458</td>\n",
       "      <td>0.089108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>R645Q</td>\n",
       "      <td>FLVLSLQDP</td>\n",
       "      <td>0.358760</td>\n",
       "      <td>0.268029</td>\n",
       "      <td>2.688388</td>\n",
       "      <td>0.123950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>R645S</td>\n",
       "      <td>FLVLSLSDP</td>\n",
       "      <td>1.946704</td>\n",
       "      <td>1.688577</td>\n",
       "      <td>2.586405</td>\n",
       "      <td>0.074924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>S643C</td>\n",
       "      <td>FLVLCLRDP</td>\n",
       "      <td>2.277147</td>\n",
       "      <td>1.987862</td>\n",
       "      <td>2.195967</td>\n",
       "      <td>0.032369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>S643F</td>\n",
       "      <td>FLVLFLRDP</td>\n",
       "      <td>2.394790</td>\n",
       "      <td>2.541795</td>\n",
       "      <td>2.487122</td>\n",
       "      <td>0.118157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>S643L</td>\n",
       "      <td>FLVLLLRDP</td>\n",
       "      <td>2.290042</td>\n",
       "      <td>2.018100</td>\n",
       "      <td>2.100405</td>\n",
       "      <td>0.188320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>S643P</td>\n",
       "      <td>FLVLPLRDP</td>\n",
       "      <td>2.406915</td>\n",
       "      <td>2.666173</td>\n",
       "      <td>2.623090</td>\n",
       "      <td>0.130599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>S643Y</td>\n",
       "      <td>FLVLYLRDP</td>\n",
       "      <td>2.517416</td>\n",
       "      <td>2.531333</td>\n",
       "      <td>2.631319</td>\n",
       "      <td>0.019684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>V641F</td>\n",
       "      <td>FLFLSLRDP</td>\n",
       "      <td>1.102037</td>\n",
       "      <td>0.173414</td>\n",
       "      <td>-0.118891</td>\n",
       "      <td>0.110138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>V641H</td>\n",
       "      <td>FLHLSLRDP</td>\n",
       "      <td>-0.025714</td>\n",
       "      <td>2.223484</td>\n",
       "      <td>2.398363</td>\n",
       "      <td>0.083596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>V641S</td>\n",
       "      <td>FLSLSLRDP</td>\n",
       "      <td>2.033074</td>\n",
       "      <td>0.885494</td>\n",
       "      <td>0.592389</td>\n",
       "      <td>0.098891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>V641W</td>\n",
       "      <td>FLWLSLRDP</td>\n",
       "      <td>1.840162</td>\n",
       "      <td>1.504313</td>\n",
       "      <td>1.565800</td>\n",
       "      <td>0.134940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>WT</td>\n",
       "      <td>FLVLSLRDP</td>\n",
       "      <td>-0.127770</td>\n",
       "      <td>-0.121916</td>\n",
       "      <td>-0.096582</td>\n",
       "      <td>0.120030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "compound genotype     aa_seq  anidulafungin  caspofungin  micafungin      none\n",
       "0              BY  FLVLSLRDP       0.000000     0.000000    0.000000  0.000000\n",
       "1           D646W  FLVLSLRWP       2.244534     2.463212    0.775816  0.080093\n",
       "2           F639D  DLVLSLRDP       0.324003     2.420326    2.429784  0.051189\n",
       "3           F639E  ELVLSLRDP      -0.009970     2.632952    2.166640  0.172062\n",
       "4           F639Q  QLVLSLRDP      -0.257678     2.486612   -0.023516  0.117442\n",
       "5           F639S  SLVLSLRDP       2.439779     2.538386    2.586103  0.107717\n",
       "6           F639V  VLVLSLRDP       2.366154     2.502741    2.528672  0.057305\n",
       "7           L640G  FGVLSLRDP       1.827176     0.694793    1.621450  0.109009\n",
       "8           L640K  FKVLSLRDP       0.325101     0.121691    0.585766 -0.380418\n",
       "9           L642F  FLVFSLRDP       2.446581     2.581567    2.532537  0.121088\n",
       "10          L642K  FLVKSLRDP       1.354517     2.503652    2.155776  0.097229\n",
       "11          L642Y  FLVYSLRDP       2.517802     2.533451    2.650863  0.148094\n",
       "12          R645D  FLVLSLDDP       0.745492     0.868791    2.563458  0.089108\n",
       "13          R645Q  FLVLSLQDP       0.358760     0.268029    2.688388  0.123950\n",
       "14          R645S  FLVLSLSDP       1.946704     1.688577    2.586405  0.074924\n",
       "15          S643C  FLVLCLRDP       2.277147     1.987862    2.195967  0.032369\n",
       "16          S643F  FLVLFLRDP       2.394790     2.541795    2.487122  0.118157\n",
       "17          S643L  FLVLLLRDP       2.290042     2.018100    2.100405  0.188320\n",
       "18          S643P  FLVLPLRDP       2.406915     2.666173    2.623090  0.130599\n",
       "19          S643Y  FLVLYLRDP       2.517416     2.531333    2.631319  0.019684\n",
       "20          V641F  FLFLSLRDP       1.102037     0.173414   -0.118891  0.110138\n",
       "21          V641H  FLHLSLRDP      -0.025714     2.223484    2.398363  0.083596\n",
       "22          V641S  FLSLSLRDP       2.033074     0.885494    0.592389  0.098891\n",
       "23          V641W  FLWLSLRDP       1.840162     1.504313    1.565800  0.134940\n",
       "24             WT  FLVLSLRDP      -0.127770    -0.121916   -0.096582  0.120030"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "widedf = corrdf.pivot_table(index=['genotype','aa_seq'], columns='compound', values='L2FC').reset_index()\n",
    "widedf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "013a0466-497e-4be1-a026-f63f52f15dd2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7f4009e92990>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 450x450 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid = sns.pairplot(widedf,\n",
    "                    x_vars=[x for x in lcomp[:-1]],\n",
    "                    y_vars=[x for x in lcomp[:-1]],\n",
    "                    diag_kind='kde', corner=True,\n",
    "                    height=1.5, aspect=1\n",
    "                   )\n",
    "\n",
    "grid.set(xlim=(-1,3), ylim=(-1,3))\n",
    "grid.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9002c4ac-6499-495b-97c4-98ff4bee3553",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dms-env",
   "language": "python",
   "name": "dms-env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}