kinney modeling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a09fd221-9aae-462c-9ecf-38179b6d7f7d",
   "metadata": {},
   "source": [
    "# modeling testing - Kinney"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "782365af-a96a-487d-ae03-9d391417b35c",
   "metadata": {},
   "source": [
    "## define imports and get data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ec5e1391-8e79-4e49-82d0-cbd940eea1e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "#standard\n",
    "import pandas as pd\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "#custom\n",
    "from titeseq_functions import *\n",
    "from titeseq_modeling import *\n",
    "from protein_onehotencoder import *\n",
    "#preprocessing\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import linear_model\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "#models and metrics\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn import metrics\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor\n",
    "from sklearn import svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c85ecba7-623a-4175-aeaa-d11e77d623ab",
   "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>CDR1H</th>\n",
       "      <th>CDR1H_AA</th>\n",
       "      <th>CDR3H</th>\n",
       "      <th>CDR3H_AA</th>\n",
       "      <th>fluorescein10^-8bin0_A</th>\n",
       "      <th>fluorescein10^-8bin3_A</th>\n",
       "      <th>fluorescein10^-8bin0_B</th>\n",
       "      <th>fluorescein10^-8bin3_B</th>\n",
       "      <th>fluorescein10^-8bin0_C</th>\n",
       "      <th>fluorescein10^-8bin3_C</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AAATATAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KYSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>0.0</td>\n",
       "      <td>511.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>128.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1143.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AAATGCAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KCSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>80.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>70.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AAATTTAGTGACTACGACATGAACTGGGTC</td>\n",
       "      <td>KFSDYDMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>268.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>752.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AAATTTAGTGACTACTCAATGAACTGGGTC</td>\n",
       "      <td>KFSDYSMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>110.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2052.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>122.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AAATTTAGTGACTACTGGAAGAACTGGGTC</td>\n",
       "      <td>KFSDYWKNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>262.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>99.0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3677</th>\n",
       "      <td>TTTCGGAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FRSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>126.0</td>\n",
       "      <td>213.0</td>\n",
       "      <td>800.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3678</th>\n",
       "      <td>TTTTTTACTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFTDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>170.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3679</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACCACGTC</td>\n",
       "      <td>FFSDYWMNHV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>126.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3680</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>71.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>479.0</td>\n",
       "      <td>256.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>395.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3681</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGTGTTGGGTC</td>\n",
       "      <td>FFSDYWMCWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>84.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>507.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3682 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               CDR1H    CDR1H_AA  \\\n",
       "0     AAATATAGTGACTACTGGATGAACTGGGTC  KYSDYWMNWV   \n",
       "1     AAATGCAGTGACTACTGGATGAACTGGGTC  KCSDYWMNWV   \n",
       "2     AAATTTAGTGACTACGACATGAACTGGGTC  KFSDYDMNWV   \n",
       "3     AAATTTAGTGACTACTCAATGAACTGGGTC  KFSDYSMNWV   \n",
       "4     AAATTTAGTGACTACTGGAAGAACTGGGTC  KFSDYWKNWV   \n",
       "...                              ...         ...   \n",
       "3677  TTTCGGAGTGACTACTGGATGAACTGGGTC  FRSDYWMNWV   \n",
       "3678  TTTTTTACTGACTACTGGATGAACTGGGTC  FFTDYWMNWV   \n",
       "3679  TTTTTTAGTGACTACTGGATGAACCACGTC  FFSDYWMNHV   \n",
       "3680  TTTTTTAGTGACTACTGGATGAACTGGGTC  FFSDYWMNWV   \n",
       "3681  TTTTTTAGTGACTACTGGATGTGTTGGGTC  FFSDYWMCWV   \n",
       "\n",
       "                               CDR3H    CDR3H_AA  fluorescein10^-8bin0_A  \\\n",
       "0     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                     0.0   \n",
       "1     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                    80.0   \n",
       "2     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                   268.0   \n",
       "3     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                   110.0   \n",
       "4     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                     0.0   \n",
       "...                              ...         ...                     ...   \n",
       "3677  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                   126.0   \n",
       "3678  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                     0.0   \n",
       "3679  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                   126.0   \n",
       "3680  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                    71.0   \n",
       "3681  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG                    84.0   \n",
       "\n",
       "      fluorescein10^-8bin3_A  fluorescein10^-8bin0_B  fluorescein10^-8bin3_B  \\\n",
       "0                      511.0                   120.0                   128.0   \n",
       "1                        1.0                   250.0                     0.0   \n",
       "2                        1.0                   752.0                     1.0   \n",
       "3                        0.0                  2052.0                     0.0   \n",
       "4                        1.0                   262.0                     3.0   \n",
       "...                      ...                     ...                     ...   \n",
       "3677                   213.0                   800.0                     0.0   \n",
       "3678                    79.0                     0.0                   177.0   \n",
       "3679                     0.0                     0.0                     0.0   \n",
       "3680                   170.0                   479.0                   256.0   \n",
       "3681                    42.0                   507.0                     0.0   \n",
       "\n",
       "      fluorescein10^-8bin0_C  fluorescein10^-8bin3_C  \n",
       "0                        1.0                  1143.0  \n",
       "1                        3.0                    70.0  \n",
       "2                      104.0                   141.0  \n",
       "3                      122.0                    13.0  \n",
       "4                        0.0                    99.0  \n",
       "...                      ...                     ...  \n",
       "3677                    66.0                     2.0  \n",
       "3678                    34.0                   170.0  \n",
       "3679                     0.0                     0.0  \n",
       "3680                    11.0                   395.0  \n",
       "3681                    81.0                     2.0  \n",
       "\n",
       "[3682 rows x 10 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fit_data_save_location = '/scratch/gthurber_root/gthurber0/marcase/titeseq/fit_data/'\n",
    "binary_data = pd.read_csv(fit_data_save_location+'binary_jbkinney2016.csv')\n",
    "continuous_data = pd.read_csv(fit_data_save_location+'jbkinney_fit_params.csv')\n",
    "binary_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c6a83858-8b0f-406d-96e6-d03e2e692e6c",
   "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>CDR1H</th>\n",
       "      <th>CDR1H_AA</th>\n",
       "      <th>CDR3H</th>\n",
       "      <th>CDR3H_AA</th>\n",
       "      <th>fit_KD_1</th>\n",
       "      <th>fit_KD_sigma_1</th>\n",
       "      <th>bottom_1</th>\n",
       "      <th>top_1</th>\n",
       "      <th>Kd_1</th>\n",
       "      <th>fit_KD_2</th>\n",
       "      <th>fit_KD_sigma_2</th>\n",
       "      <th>bottom_2</th>\n",
       "      <th>top_2</th>\n",
       "      <th>Kd_2</th>\n",
       "      <th>fit_KD_3</th>\n",
       "      <th>fit_KD_sigma_3</th>\n",
       "      <th>bottom_3</th>\n",
       "      <th>top_3</th>\n",
       "      <th>Kd_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AAATATAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KYSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>2.150305e-09</td>\n",
       "      <td>2.877315e-02</td>\n",
       "      <td>1.739289</td>\n",
       "      <td>3.575290</td>\n",
       "      <td>2.213354e-09</td>\n",
       "      <td>4.097321e-10</td>\n",
       "      <td>1.233956e-01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>5.040806e-10</td>\n",
       "      <td>2.304446e-02</td>\n",
       "      <td>1.624784</td>\n",
       "      <td>3.303060</td>\n",
       "      <td>1.527599e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AAATGCAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KCSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>3.715352e-08</td>\n",
       "      <td>7.545361e-04</td>\n",
       "      <td>1.342356</td>\n",
       "      <td>2.735451</td>\n",
       "      <td>1.772702e-08</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>5.741405e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.172759e-09</td>\n",
       "      <td>1.394896e-02</td>\n",
       "      <td>1.434525</td>\n",
       "      <td>2.826896</td>\n",
       "      <td>1.353816e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AAATTTAGTGACTACGACATGAACTGGGTC</td>\n",
       "      <td>KFSDYDMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.462227e-78</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>6.284835e-34</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>8.370637e-76</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AAATTTAGTGACTACTCAATGAACTGGGTC</td>\n",
       "      <td>KFSDYSMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3.489924e-47</td>\n",
       "      <td>1.423728</td>\n",
       "      <td>1.974225</td>\n",
       "      <td>2.905324e-09</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>7.116351e-32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.321824e-42</td>\n",
       "      <td>1.638154</td>\n",
       "      <td>1.922830</td>\n",
       "      <td>7.099199e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>AAATTTAGTGACTACTGGAAGAACTGGGTC</td>\n",
       "      <td>KFSDYWKNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>4.245572e-07</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.958250e-06</td>\n",
       "      <td>5.081689e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.597270e-06</td>\n",
       "      <td>4.514893e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</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",
       "      <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>3677</th>\n",
       "      <td>TTTCGGAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FRSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>7.750417e-29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>9.425554e-18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>9.619143e-22</td>\n",
       "      <td>1.588565</td>\n",
       "      <td>2.067151</td>\n",
       "      <td>3.594640e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3678</th>\n",
       "      <td>TTTTTTACTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFTDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>3.801894e-09</td>\n",
       "      <td>2.193324e-02</td>\n",
       "      <td>1.874532</td>\n",
       "      <td>2.814975</td>\n",
       "      <td>1.257048e-10</td>\n",
       "      <td>4.677351e-09</td>\n",
       "      <td>3.144558e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.463865e-09</td>\n",
       "      <td>1.606536e-02</td>\n",
       "      <td>1.825989</td>\n",
       "      <td>3.028533</td>\n",
       "      <td>3.167904e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3679</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACCACGTC</td>\n",
       "      <td>FFSDYWMNHV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.061264e-08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>5.057135e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>2.068179e-03</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3680</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>2.511886e-09</td>\n",
       "      <td>1.953354e-03</td>\n",
       "      <td>2.349266</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.243920e-06</td>\n",
       "      <td>7.852356e-09</td>\n",
       "      <td>2.917752e-02</td>\n",
       "      <td>2.420853</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>1.280855e-09</td>\n",
       "      <td>2.391568e-04</td>\n",
       "      <td>1.738760</td>\n",
       "      <td>3.102289</td>\n",
       "      <td>3.096425e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3681</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGTGTTGGGTC</td>\n",
       "      <td>FFSDYWMCWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3.511612e-27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.793979e-36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.220181e-34</td>\n",
       "      <td>1.429000</td>\n",
       "      <td>1.794544</td>\n",
       "      <td>1.496713e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3682 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               CDR1H    CDR1H_AA  \\\n",
       "0     AAATATAGTGACTACTGGATGAACTGGGTC  KYSDYWMNWV   \n",
       "1     AAATGCAGTGACTACTGGATGAACTGGGTC  KCSDYWMNWV   \n",
       "2     AAATTTAGTGACTACGACATGAACTGGGTC  KFSDYDMNWV   \n",
       "3     AAATTTAGTGACTACTCAATGAACTGGGTC  KFSDYSMNWV   \n",
       "4     AAATTTAGTGACTACTGGAAGAACTGGGTC  KFSDYWKNWV   \n",
       "...                              ...         ...   \n",
       "3677  TTTCGGAGTGACTACTGGATGAACTGGGTC  FRSDYWMNWV   \n",
       "3678  TTTTTTACTGACTACTGGATGAACTGGGTC  FFTDYWMNWV   \n",
       "3679  TTTTTTAGTGACTACTGGATGAACCACGTC  FFSDYWMNHV   \n",
       "3680  TTTTTTAGTGACTACTGGATGAACTGGGTC  FFSDYWMNWV   \n",
       "3681  TTTTTTAGTGACTACTGGATGTGTTGGGTC  FFSDYWMCWV   \n",
       "\n",
       "                               CDR3H    CDR3H_AA      fit_KD_1  \\\n",
       "0     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  2.150305e-09   \n",
       "1     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  3.715352e-08   \n",
       "2     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "3     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "4     ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "...                              ...         ...           ...   \n",
       "3677  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "3678  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  3.801894e-09   \n",
       "3679  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "3680  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  2.511886e-09   \n",
       "3681  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG  1.000000e-05   \n",
       "\n",
       "      fit_KD_sigma_1  bottom_1     top_1          Kd_1      fit_KD_2  \\\n",
       "0       2.877315e-02  1.739289  3.575290  2.213354e-09  4.097321e-10   \n",
       "1       7.545361e-04  1.342356  2.735451  1.772702e-08  1.000000e-05   \n",
       "2       1.462227e-78       NaN       NaN           NaN  1.000000e-05   \n",
       "3       3.489924e-47  1.423728  1.974225  2.905324e-09  1.000000e-05   \n",
       "4       4.245572e-07       NaN       NaN           NaN  6.958250e-06   \n",
       "...              ...       ...       ...           ...           ...   \n",
       "3677    7.750417e-29       NaN       NaN           NaN  1.000000e-05   \n",
       "3678    2.193324e-02  1.874532  2.814975  1.257048e-10  4.677351e-09   \n",
       "3679    1.061264e-08       NaN       NaN           NaN  1.000000e-05   \n",
       "3680    1.953354e-03  2.349266  4.000000  7.243920e-06  7.852356e-09   \n",
       "3681    3.511612e-27       NaN       NaN           NaN  1.000000e-05   \n",
       "\n",
       "      fit_KD_sigma_2  bottom_2  top_2     Kd_2      fit_KD_3  fit_KD_sigma_3  \\\n",
       "0       1.233956e-01  0.000000    0.0  0.00000  5.040806e-10    2.304446e-02   \n",
       "1       5.741405e-04       NaN    NaN      NaN  9.172759e-09    1.394896e-02   \n",
       "2       6.284835e-34       NaN    NaN      NaN  1.000000e-05    8.370637e-76   \n",
       "3       7.116351e-32       NaN    NaN      NaN  1.000000e-05    1.321824e-42   \n",
       "4       5.081689e-02       NaN    NaN      NaN  4.597270e-06    4.514893e-02   \n",
       "...              ...       ...    ...      ...           ...             ...   \n",
       "3677    9.425554e-18       NaN    NaN      NaN  1.000000e-05    9.619143e-22   \n",
       "3678    3.144558e-02       NaN    NaN      NaN  5.463865e-09    1.606536e-02   \n",
       "3679    5.057135e-02       NaN    NaN      NaN  1.000000e-05    2.068179e-03   \n",
       "3680    2.917752e-02  2.420853    4.0  0.00001  1.280855e-09    2.391568e-04   \n",
       "3681    1.793979e-36       NaN    NaN      NaN  1.000000e-05    1.220181e-34   \n",
       "\n",
       "      bottom_3     top_3          Kd_3  \n",
       "0     1.624784  3.303060  1.527599e-10  \n",
       "1     1.434525  2.826896  1.353816e-09  \n",
       "2     0.000000  0.000000  0.000000e+00  \n",
       "3     1.638154  1.922830  7.099199e-08  \n",
       "4          NaN       NaN           NaN  \n",
       "...        ...       ...           ...  \n",
       "3677  1.588565  2.067151  3.594640e-08  \n",
       "3678  1.825989  3.028533  3.167904e-09  \n",
       "3679       NaN       NaN           NaN  \n",
       "3680  1.738760  3.102289  3.096425e-10  \n",
       "3681  1.429000  1.794544  1.496713e-08  \n",
       "\n",
       "[3682 rows x 19 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "continuous_data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "032f8456-38a3-4bf4-b46f-070767d6033f",
   "metadata": {
    "tags": []
   },
   "source": [
    "## binary data preprocesing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5bdce36e-44af-49ee-917f-b1593b8c8133",
   "metadata": {},
   "outputs": [],
   "source": [
    "binary_data = binary_data[['CDR1H_AA', 'CDR3H_AA', 'fluorescein10^-8bin0_A',\n",
    "       'fluorescein10^-8bin3_A', 'fluorescein10^-8bin0_B',\n",
    "       'fluorescein10^-8bin3_B', 'fluorescein10^-8bin0_C',\n",
    "       'fluorescein10^-8bin3_C']]\n",
    "binary_data.index = binary_data['CDR1H_AA'] + '_' + binary_data['CDR3H_AA']\n",
    "binary_data = binary_data.drop(columns=['CDR1H_AA','CDR3H_AA'])\n",
    "binary_data = binary_data.rename(columns={'fluorescein10^-8bin0_A':'low_A','fluorescein10^-8bin3_A':'high_A','fluorescein10^-8bin0_B':'low_B','fluorescein10^-8bin3_B':'high_B','fluorescein10^-8bin0_C':'low_C','fluorescein10^-8bin3_C':'high_C'})\n",
    "binary_data.index.name = 'sequence'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "752ab0b2-7550-4df8-abbb-720d7da6ae94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get datasize based on various cutoffs\n",
    "percentile_cutoffs = [0,0.01,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.75,0.8,0.825,0.85,0.875,0.9,0.925,0.95,0.99,1]\n",
    "replicate_cutoffs = [1,2,3,4,5,6,7,8,9,10,15,20,25,30,35,40]\n",
    "zero_tolerance = True\n",
    "plotDatasetCutoff(input_data=binary_data,data_type='duplicated',replicate_cutoffs=replicate_cutoffs,percentile_cutoffs=percentile_cutoffs,zero_tolerance=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "2e61f0c0-84ce-426b-81c0-ffc9754b6199",
   "metadata": {},
   "outputs": [],
   "source": [
    "positive,negative = generateBinaryDataset(binary_data,data_type='duplicated',replicate_cutoff=(2,2),percentile_cutoff=(0.8,0.2),zero_tolerance=True)\n",
    "positive['label'] = 1\n",
    "negative['label'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "aecb53e1-4b7c-4d97-a8f0-65a825a334cf",
   "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>mean</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sequence</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>KYSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>137.101548</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWKNWV_GSYYGMDYWG</th>\n",
       "      <td>7.651430</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWMNRV_GSYYGMDYWG</th>\n",
       "      <td>9.107354</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>107.827854</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFIDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>45.526543</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LFSDYWMNWQ_GSYYGMDYWG</th>\n",
       "      <td>13.972124</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LFSGYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>43.109552</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>2.397841</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMNWC_GSYYGMDYWG</th>\n",
       "      <td>105.884432</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFTDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>207.386804</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>513 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             mean  label\n",
       "sequence                                \n",
       "KYSDYWMNWV_GSYYGMDYWG  137.101548      1\n",
       "KFSDYWKNWV_GSYYGMDYWG    7.651430      1\n",
       "KFSDYWMNRV_GSYYGMDYWG    9.107354      1\n",
       "KFSDYWMNWV_GSYYGMDYWG  107.827854      1\n",
       "KFIDYWMNWV_GSYYGMDYWG   45.526543      1\n",
       "...                           ...    ...\n",
       "LFSDYWMNWQ_GSYYGMDYWG   13.972124      1\n",
       "LFSGYWMNWV_GSYYGMDYWG   43.109552      1\n",
       "FFSDYWMNWV_GSYYGMDYWG    2.397841      1\n",
       "FFSDYWMNWC_GSYYGMDYWG  105.884432      1\n",
       "FFTDYWMNWV_GSYYGMDYWG  207.386804      1\n",
       "\n",
       "[513 rows x 2 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "positive"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8b6fd628-ad6c-4cdf-8f1f-37ad3c891574",
   "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>mean</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sequence</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>NFSDYWENWL_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TTSDYWMNDV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TISDGWMNWV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TISDYWMRWV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TQSDYWMNTV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GRSDKWMNWV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GFSDYWPTWV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GFSDYWMNNV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YFSDYWMNQV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YWSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>127 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       mean  label\n",
       "sequence                          \n",
       "NFSDYWENWL_GSYYGMDYWG   0.0      0\n",
       "TTSDYWMNDV_GSYYGMDYWG   0.0      0\n",
       "TISDGWMNWV_GSYYGMDYWG   0.0      0\n",
       "TISDYWMRWV_GSYYGMDYWG   0.0      0\n",
       "TQSDYWMNTV_GSYYGMDYWG   0.0      0\n",
       "...                     ...    ...\n",
       "GRSDKWMNWV_GSYYGMDYWG   0.0      0\n",
       "GFSDYWPTWV_GSYYGMDYWG   0.0      0\n",
       "GFSDYWMNNV_GSYYGMDYWG   0.0      0\n",
       "YFSDYWMNQV_GSYYGMDYWG   0.0      0\n",
       "YWSDYWMNWV_GSYYGMDYWG   0.0      0\n",
       "\n",
       "[127 rows x 2 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "0fc19608-3a6d-4395-a085-7730cd8612c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of shared sequences between positive and datasets is 0\n"
     ]
    },
    {
     "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>mean</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sequence</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>KYSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>137.101548</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWKNWV_GSYYGMDYWG</th>\n",
       "      <td>7.651430</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWMNRV_GSYYGMDYWG</th>\n",
       "      <td>9.107354</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>107.827854</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFIDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>45.526543</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GRSDKWMNWV_GSYYGMDYWG</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GFSDYWPTWV_GSYYGMDYWG</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GFSDYWMNNV_GSYYGMDYWG</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YFSDYWMNQV_GSYYGMDYWG</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YWSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>640 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             mean  label\n",
       "sequence                                \n",
       "KYSDYWMNWV_GSYYGMDYWG  137.101548      1\n",
       "KFSDYWKNWV_GSYYGMDYWG    7.651430      1\n",
       "KFSDYWMNRV_GSYYGMDYWG    9.107354      1\n",
       "KFSDYWMNWV_GSYYGMDYWG  107.827854      1\n",
       "KFIDYWMNWV_GSYYGMDYWG   45.526543      1\n",
       "...                           ...    ...\n",
       "GRSDKWMNWV_GSYYGMDYWG    0.000000      0\n",
       "GFSDYWPTWV_GSYYGMDYWG    0.000000      0\n",
       "GFSDYWMNNV_GSYYGMDYWG    0.000000      0\n",
       "YFSDYWMNQV_GSYYGMDYWG    0.000000      0\n",
       "YWSDYWMNWV_GSYYGMDYWG    0.000000      0\n",
       "\n",
       "[640 rows x 2 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(f'number of shared sequences between positive and datasets is {len(positive.index.intersection(negative.index))}') #should be zero length\n",
    "binary_data = pd.concat([positive,negative],join='outer',axis=0)\n",
    "binary_data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d23d182f-3db6-41b6-a93a-66443a13ed5b",
   "metadata": {
    "tags": []
   },
   "source": [
    "## continuous data preprocesing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2450c239-8b5c-4a7e-ab3e-6136d3c64c43",
   "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>CDR1H</th>\n",
       "      <th>CDR1H_AA</th>\n",
       "      <th>CDR3H</th>\n",
       "      <th>CDR3H_AA</th>\n",
       "      <th>fit_KD_1</th>\n",
       "      <th>fit_KD_sigma_1</th>\n",
       "      <th>bottom_1</th>\n",
       "      <th>top_1</th>\n",
       "      <th>Kd_1</th>\n",
       "      <th>fit_KD_2</th>\n",
       "      <th>fit_KD_sigma_2</th>\n",
       "      <th>bottom_2</th>\n",
       "      <th>top_2</th>\n",
       "      <th>Kd_2</th>\n",
       "      <th>fit_KD_3</th>\n",
       "      <th>fit_KD_sigma_3</th>\n",
       "      <th>bottom_3</th>\n",
       "      <th>top_3</th>\n",
       "      <th>Kd_3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sequence</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>KYSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>AAATATAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KYSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>2.150305e-09</td>\n",
       "      <td>2.877315e-02</td>\n",
       "      <td>1.739289</td>\n",
       "      <td>3.575290</td>\n",
       "      <td>2.213354e-09</td>\n",
       "      <td>4.097321e-10</td>\n",
       "      <td>1.233956e-01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>5.040806e-10</td>\n",
       "      <td>2.304446e-02</td>\n",
       "      <td>1.624784</td>\n",
       "      <td>3.303060</td>\n",
       "      <td>1.527599e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KCSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>AAATGCAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>KCSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>3.715352e-08</td>\n",
       "      <td>7.545361e-04</td>\n",
       "      <td>1.342356</td>\n",
       "      <td>2.735451</td>\n",
       "      <td>1.772702e-08</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>5.741405e-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.172759e-09</td>\n",
       "      <td>1.394896e-02</td>\n",
       "      <td>1.434525</td>\n",
       "      <td>2.826896</td>\n",
       "      <td>1.353816e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYDMNWV_GSYYGMDYWG</th>\n",
       "      <td>AAATTTAGTGACTACGACATGAACTGGGTC</td>\n",
       "      <td>KFSDYDMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.462227e-78</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>6.284835e-34</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>8.370637e-76</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYSMNWV_GSYYGMDYWG</th>\n",
       "      <td>AAATTTAGTGACTACTCAATGAACTGGGTC</td>\n",
       "      <td>KFSDYSMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3.489924e-47</td>\n",
       "      <td>1.423728</td>\n",
       "      <td>1.974225</td>\n",
       "      <td>2.905324e-09</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>7.116351e-32</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.321824e-42</td>\n",
       "      <td>1.638154</td>\n",
       "      <td>1.922830</td>\n",
       "      <td>7.099199e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWKNWV_GSYYGMDYWG</th>\n",
       "      <td>AAATTTAGTGACTACTGGAAGAACTGGGTC</td>\n",
       "      <td>KFSDYWKNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>4.245572e-07</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.958250e-06</td>\n",
       "      <td>5.081689e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.597270e-06</td>\n",
       "      <td>4.514893e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</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",
       "      <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>FRSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>TTTCGGAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FRSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>7.750417e-29</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>9.425554e-18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>9.619143e-22</td>\n",
       "      <td>1.588565</td>\n",
       "      <td>2.067151</td>\n",
       "      <td>3.594640e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFTDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>TTTTTTACTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFTDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>3.801894e-09</td>\n",
       "      <td>2.193324e-02</td>\n",
       "      <td>1.874532</td>\n",
       "      <td>2.814975</td>\n",
       "      <td>1.257048e-10</td>\n",
       "      <td>4.677351e-09</td>\n",
       "      <td>3.144558e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.463865e-09</td>\n",
       "      <td>1.606536e-02</td>\n",
       "      <td>1.825989</td>\n",
       "      <td>3.028533</td>\n",
       "      <td>3.167904e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMNHV_GSYYGMDYWG</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACCACGTC</td>\n",
       "      <td>FFSDYWMNHV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.061264e-08</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>5.057135e-02</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>2.068179e-03</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGAACTGGGTC</td>\n",
       "      <td>FFSDYWMNWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>2.511886e-09</td>\n",
       "      <td>1.953354e-03</td>\n",
       "      <td>2.349266</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>7.243920e-06</td>\n",
       "      <td>7.852356e-09</td>\n",
       "      <td>2.917752e-02</td>\n",
       "      <td>2.420853</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>1.280855e-09</td>\n",
       "      <td>2.391568e-04</td>\n",
       "      <td>1.738760</td>\n",
       "      <td>3.102289</td>\n",
       "      <td>3.096425e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMCWV_GSYYGMDYWG</th>\n",
       "      <td>TTTTTTAGTGACTACTGGATGTGTTGGGTC</td>\n",
       "      <td>FFSDYWMCWV</td>\n",
       "      <td>ACCCCAGTAGTCCATACCATAGTAAGAACC</td>\n",
       "      <td>GSYYGMDYWG</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3.511612e-27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.793979e-36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>1.220181e-34</td>\n",
       "      <td>1.429000</td>\n",
       "      <td>1.794544</td>\n",
       "      <td>1.496713e-08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3682 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                CDR1H    CDR1H_AA  \\\n",
       "sequence                                                            \n",
       "KYSDYWMNWV_GSYYGMDYWG  AAATATAGTGACTACTGGATGAACTGGGTC  KYSDYWMNWV   \n",
       "KCSDYWMNWV_GSYYGMDYWG  AAATGCAGTGACTACTGGATGAACTGGGTC  KCSDYWMNWV   \n",
       "KFSDYDMNWV_GSYYGMDYWG  AAATTTAGTGACTACGACATGAACTGGGTC  KFSDYDMNWV   \n",
       "KFSDYSMNWV_GSYYGMDYWG  AAATTTAGTGACTACTCAATGAACTGGGTC  KFSDYSMNWV   \n",
       "KFSDYWKNWV_GSYYGMDYWG  AAATTTAGTGACTACTGGAAGAACTGGGTC  KFSDYWKNWV   \n",
       "...                                               ...         ...   \n",
       "FRSDYWMNWV_GSYYGMDYWG  TTTCGGAGTGACTACTGGATGAACTGGGTC  FRSDYWMNWV   \n",
       "FFTDYWMNWV_GSYYGMDYWG  TTTTTTACTGACTACTGGATGAACTGGGTC  FFTDYWMNWV   \n",
       "FFSDYWMNHV_GSYYGMDYWG  TTTTTTAGTGACTACTGGATGAACCACGTC  FFSDYWMNHV   \n",
       "FFSDYWMNWV_GSYYGMDYWG  TTTTTTAGTGACTACTGGATGAACTGGGTC  FFSDYWMNWV   \n",
       "FFSDYWMCWV_GSYYGMDYWG  TTTTTTAGTGACTACTGGATGTGTTGGGTC  FFSDYWMCWV   \n",
       "\n",
       "                                                CDR3H    CDR3H_AA  \\\n",
       "sequence                                                            \n",
       "KYSDYWMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "KCSDYWMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "KFSDYDMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "KFSDYSMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "KFSDYWKNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "...                                               ...         ...   \n",
       "FRSDYWMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "FFTDYWMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "FFSDYWMNHV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "FFSDYWMNWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "FFSDYWMCWV_GSYYGMDYWG  ACCCCAGTAGTCCATACCATAGTAAGAACC  GSYYGMDYWG   \n",
       "\n",
       "                           fit_KD_1  fit_KD_sigma_1  bottom_1     top_1  \\\n",
       "sequence                                                                  \n",
       "KYSDYWMNWV_GSYYGMDYWG  2.150305e-09    2.877315e-02  1.739289  3.575290   \n",
       "KCSDYWMNWV_GSYYGMDYWG  3.715352e-08    7.545361e-04  1.342356  2.735451   \n",
       "KFSDYDMNWV_GSYYGMDYWG  1.000000e-05    1.462227e-78       NaN       NaN   \n",
       "KFSDYSMNWV_GSYYGMDYWG  1.000000e-05    3.489924e-47  1.423728  1.974225   \n",
       "KFSDYWKNWV_GSYYGMDYWG  1.000000e-05    4.245572e-07       NaN       NaN   \n",
       "...                             ...             ...       ...       ...   \n",
       "FRSDYWMNWV_GSYYGMDYWG  1.000000e-05    7.750417e-29       NaN       NaN   \n",
       "FFTDYWMNWV_GSYYGMDYWG  3.801894e-09    2.193324e-02  1.874532  2.814975   \n",
       "FFSDYWMNHV_GSYYGMDYWG  1.000000e-05    1.061264e-08       NaN       NaN   \n",
       "FFSDYWMNWV_GSYYGMDYWG  2.511886e-09    1.953354e-03  2.349266  4.000000   \n",
       "FFSDYWMCWV_GSYYGMDYWG  1.000000e-05    3.511612e-27       NaN       NaN   \n",
       "\n",
       "                               Kd_1      fit_KD_2  fit_KD_sigma_2  bottom_2  \\\n",
       "sequence                                                                      \n",
       "KYSDYWMNWV_GSYYGMDYWG  2.213354e-09  4.097321e-10    1.233956e-01  0.000000   \n",
       "KCSDYWMNWV_GSYYGMDYWG  1.772702e-08  1.000000e-05    5.741405e-04       NaN   \n",
       "KFSDYDMNWV_GSYYGMDYWG           NaN  1.000000e-05    6.284835e-34       NaN   \n",
       "KFSDYSMNWV_GSYYGMDYWG  2.905324e-09  1.000000e-05    7.116351e-32       NaN   \n",
       "KFSDYWKNWV_GSYYGMDYWG           NaN  6.958250e-06    5.081689e-02       NaN   \n",
       "...                             ...           ...             ...       ...   \n",
       "FRSDYWMNWV_GSYYGMDYWG           NaN  1.000000e-05    9.425554e-18       NaN   \n",
       "FFTDYWMNWV_GSYYGMDYWG  1.257048e-10  4.677351e-09    3.144558e-02       NaN   \n",
       "FFSDYWMNHV_GSYYGMDYWG           NaN  1.000000e-05    5.057135e-02       NaN   \n",
       "FFSDYWMNWV_GSYYGMDYWG  7.243920e-06  7.852356e-09    2.917752e-02  2.420853   \n",
       "FFSDYWMCWV_GSYYGMDYWG           NaN  1.000000e-05    1.793979e-36       NaN   \n",
       "\n",
       "                       top_2     Kd_2      fit_KD_3  fit_KD_sigma_3  bottom_3  \\\n",
       "sequence                                                                        \n",
       "KYSDYWMNWV_GSYYGMDYWG    0.0  0.00000  5.040806e-10    2.304446e-02  1.624784   \n",
       "KCSDYWMNWV_GSYYGMDYWG    NaN      NaN  9.172759e-09    1.394896e-02  1.434525   \n",
       "KFSDYDMNWV_GSYYGMDYWG    NaN      NaN  1.000000e-05    8.370637e-76  0.000000   \n",
       "KFSDYSMNWV_GSYYGMDYWG    NaN      NaN  1.000000e-05    1.321824e-42  1.638154   \n",
       "KFSDYWKNWV_GSYYGMDYWG    NaN      NaN  4.597270e-06    4.514893e-02       NaN   \n",
       "...                      ...      ...           ...             ...       ...   \n",
       "FRSDYWMNWV_GSYYGMDYWG    NaN      NaN  1.000000e-05    9.619143e-22  1.588565   \n",
       "FFTDYWMNWV_GSYYGMDYWG    NaN      NaN  5.463865e-09    1.606536e-02  1.825989   \n",
       "FFSDYWMNHV_GSYYGMDYWG    NaN      NaN  1.000000e-05    2.068179e-03       NaN   \n",
       "FFSDYWMNWV_GSYYGMDYWG    4.0  0.00001  1.280855e-09    2.391568e-04  1.738760   \n",
       "FFSDYWMCWV_GSYYGMDYWG    NaN      NaN  1.000000e-05    1.220181e-34  1.429000   \n",
       "\n",
       "                          top_3          Kd_3  \n",
       "sequence                                       \n",
       "KYSDYWMNWV_GSYYGMDYWG  3.303060  1.527599e-10  \n",
       "KCSDYWMNWV_GSYYGMDYWG  2.826896  1.353816e-09  \n",
       "KFSDYDMNWV_GSYYGMDYWG  0.000000  0.000000e+00  \n",
       "KFSDYSMNWV_GSYYGMDYWG  1.922830  7.099199e-08  \n",
       "KFSDYWKNWV_GSYYGMDYWG       NaN           NaN  \n",
       "...                         ...           ...  \n",
       "FRSDYWMNWV_GSYYGMDYWG  2.067151  3.594640e-08  \n",
       "FFTDYWMNWV_GSYYGMDYWG  3.028533  3.167904e-09  \n",
       "FFSDYWMNHV_GSYYGMDYWG       NaN           NaN  \n",
       "FFSDYWMNWV_GSYYGMDYWG  3.102289  3.096425e-10  \n",
       "FFSDYWMCWV_GSYYGMDYWG  1.794544  1.496713e-08  \n",
       "\n",
       "[3682 rows x 19 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "continuous_data.index = continuous_data['CDR1H_AA'] + '_' + continuous_data['CDR3H_AA']\n",
    "continuous_data.index.name = 'sequence'\n",
    "continuous_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "4352b8c0-5b63-48af-a86b-58e0e4c5aa23",
   "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>Kds</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sequence</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TFSDYWMNWV_GSYYGMDYWG</th>\n",
       "      <td>[2.38506379547e-09, 9.88553094657e-10, 8.46252...</td>\n",
       "      <td>1.516271e-09</td>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFSDSWMNWV_GSYYGMDYWG</th>\n",
       "      <td>[2.16271852373e-07, 2.86747737558e-08, 4.24130...</td>\n",
       "      <td>1.081602e-06</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFSDYWMNWS_GSYYGMDYWG</th>\n",
       "      <td>[5.33949273574e-08, 4.78630092323e-10, 1.21618...</td>\n",
       "      <td>1.793770e-07</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFSDYWMNRV_GSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05, 2.93426950396e-09, 1.659...</td>\n",
       "      <td>5.381697e-06</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFSDYWMNSV_GSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05, 1e-05, 1e-05, 1e-05, 3.1...</td>\n",
       "      <td>7.240362e-06</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFGDYWRNWV_GSYYGMDYWG</th>\n",
       "      <td>[3.16227766017e-10, 1.25169927303e-08]</td>\n",
       "      <td>6.416610e-09</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFFDYWENWV_GSYYGMDYWG</th>\n",
       "      <td>[3.16227766017e-10, 3.16227766017e-10]</td>\n",
       "      <td>3.162278e-10</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>THSDYWMNWR_GSYYGMDYWG</th>\n",
       "      <td>[3.16227766017e-10, 1.58489319246e-07]</td>\n",
       "      <td>7.940277e-08</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFSDYWRNWW_GSYYGMDYWG</th>\n",
       "      <td>[3.52776786619e-08]</td>\n",
       "      <td>3.527768e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TFTDYWMNWP_GSYYGMDYWG</th>\n",
       "      <td>[5.92243086876e-08]</td>\n",
       "      <td>5.922431e-08</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2807 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                     Kds  \\\n",
       "sequence                                                                   \n",
       "TFSDYWMNWV_GSYYGMDYWG  [2.38506379547e-09, 9.88553094657e-10, 8.46252...   \n",
       "TFSDSWMNWV_GSYYGMDYWG  [2.16271852373e-07, 2.86747737558e-08, 4.24130...   \n",
       "TFSDYWMNWS_GSYYGMDYWG  [5.33949273574e-08, 4.78630092323e-10, 1.21618...   \n",
       "TFSDYWMNRV_GSYYGMDYWG  [1e-05, 1e-05, 1e-05, 2.93426950396e-09, 1.659...   \n",
       "TFSDYWMNSV_GSYYGMDYWG  [1e-05, 1e-05, 1e-05, 1e-05, 1e-05, 1e-05, 3.1...   \n",
       "...                                                                  ...   \n",
       "TFGDYWRNWV_GSYYGMDYWG             [3.16227766017e-10, 1.25169927303e-08]   \n",
       "TFFDYWENWV_GSYYGMDYWG             [3.16227766017e-10, 3.16227766017e-10]   \n",
       "THSDYWMNWR_GSYYGMDYWG             [3.16227766017e-10, 1.58489319246e-07]   \n",
       "TFSDYWRNWW_GSYYGMDYWG                                [3.52776786619e-08]   \n",
       "TFTDYWMNWP_GSYYGMDYWG                                [5.92243086876e-08]   \n",
       "\n",
       "                               mean  count  \n",
       "sequence                                    \n",
       "TFSDYWMNWV_GSYYGMDYWG  1.516271e-09    102  \n",
       "TFSDSWMNWV_GSYYGMDYWG  1.081602e-06     18  \n",
       "TFSDYWMNWS_GSYYGMDYWG  1.793770e-07     18  \n",
       "TFSDYWMNRV_GSYYGMDYWG  5.381697e-06     18  \n",
       "TFSDYWMNSV_GSYYGMDYWG  7.240362e-06     18  \n",
       "...                             ...    ...  \n",
       "TFGDYWRNWV_GSYYGMDYWG  6.416610e-09      2  \n",
       "TFFDYWENWV_GSYYGMDYWG  3.162278e-10      2  \n",
       "THSDYWMNWR_GSYYGMDYWG  7.940277e-08      2  \n",
       "TFSDYWRNWW_GSYYGMDYWG  3.527768e-08      1  \n",
       "TFTDYWMNWP_GSYYGMDYWG  5.922431e-08      1  \n",
       "\n",
       "[2807 rows x 3 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#aggregate and average replicate Kd values\n",
    "replicates = [1,2,3]\n",
    "df1 = pd.DataFrame(index=continuous_data.index.unique(),columns=continuous_data.columns)\n",
    "continuous_data['Kds'] = continuous_data[['fit_KD_' + str(r) for r in replicates]].values.tolist()\n",
    "df1['Kds'] = continuous_data.groupby('sequence')['Kds'].apply(lambda x: list([item for sublist in x for item in sublist]))\n",
    "df1 = df1.drop(columns=df1.columns[:-1])\n",
    "df1['Kds'] = df1['Kds'].apply(lambda x: [i for i in x if not np.isnan(i)])\n",
    "df1['mean'] = df1['Kds'].apply(lambda x: np.nanmean(x))\n",
    "df1['count'] = df1['Kds'].apply(lambda x: len(x))\n",
    "continuous_data = df1\n",
    "continuous_data.sort_values('count',ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08f00b73-11e4-4ee7-b421-875f23675f13",
   "metadata": {
    "tags": []
   },
   "source": [
    "## automatic one hot encoding via ProteinOneHotEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "0839be6e-28d4-41ea-bddd-e8ff83ae2fdb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "varying sites are [ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]\n"
     ]
    },
    {
     "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>Kds</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>KYSDYWMNWVGSYYGMDYWG</th>\n",
       "      <td>[2.15030467934e-09, 4.09732109814e-10, 5.04080...</td>\n",
       "      <td>1.021372e-09</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KCSDYWMNWVGSYYGMDYWG</th>\n",
       "      <td>[3.71535229097e-08, 1e-05, 9.1727593539e-09]</td>\n",
       "      <td>3.348775e-06</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYDMNWVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05]</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYSMNWVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05]</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KFSDYWKNWVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 6.95824954701e-06, 4.59726988531e-06]</td>\n",
       "      <td>7.185173e-06</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSAYWMNWVGSYYGMDYWG</th>\n",
       "      <td>[5.06990708275e-08, 5.3703179637e-06, 1.708047...</td>\n",
       "      <td>1.812699e-06</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LFSDYWDNWVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05]</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFTDYWMNWVGSYYGMDYWG</th>\n",
       "      <td>[3.80189396321e-09, 4.67735141287e-09, 5.46386...</td>\n",
       "      <td>4.647704e-09</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMNHVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05]</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FFSDYWMCWVGSYYGMDYWG</th>\n",
       "      <td>[1e-05, 1e-05, 1e-05]</td>\n",
       "      <td>1.000000e-05</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2807 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                    Kds  \\\n",
       "KYSDYWMNWVGSYYGMDYWG  [2.15030467934e-09, 4.09732109814e-10, 5.04080...   \n",
       "KCSDYWMNWVGSYYGMDYWG       [3.71535229097e-08, 1e-05, 9.1727593539e-09]   \n",
       "KFSDYDMNWVGSYYGMDYWG                              [1e-05, 1e-05, 1e-05]   \n",
       "KFSDYSMNWVGSYYGMDYWG                              [1e-05, 1e-05, 1e-05]   \n",
       "KFSDYWKNWVGSYYGMDYWG      [1e-05, 6.95824954701e-06, 4.59726988531e-06]   \n",
       "...                                                                 ...   \n",
       "FFSAYWMNWVGSYYGMDYWG  [5.06990708275e-08, 5.3703179637e-06, 1.708047...   \n",
       "LFSDYWDNWVGSYYGMDYWG                              [1e-05, 1e-05, 1e-05]   \n",
       "FFTDYWMNWVGSYYGMDYWG  [3.80189396321e-09, 4.67735141287e-09, 5.46386...   \n",
       "FFSDYWMNHVGSYYGMDYWG                              [1e-05, 1e-05, 1e-05]   \n",
       "FFSDYWMCWVGSYYGMDYWG                              [1e-05, 1e-05, 1e-05]   \n",
       "\n",
       "                              mean  count  \n",
       "KYSDYWMNWVGSYYGMDYWG  1.021372e-09      3  \n",
       "KCSDYWMNWVGSYYGMDYWG  3.348775e-06      3  \n",
       "KFSDYDMNWVGSYYGMDYWG  1.000000e-05      3  \n",
       "KFSDYSMNWVGSYYGMDYWG  1.000000e-05      3  \n",
       "KFSDYWKNWVGSYYGMDYWG  7.185173e-06      3  \n",
       "...                            ...    ...  \n",
       "FFSAYWMNWVGSYYGMDYWG  1.812699e-06      3  \n",
       "LFSDYWDNWVGSYYGMDYWG  1.000000e-05      3  \n",
       "FFTDYWMNWVGSYYGMDYWG  4.647704e-09      3  \n",
       "FFSDYWMNHVGSYYGMDYWG  1.000000e-05      3  \n",
       "FFSDYWMCWVGSYYGMDYWG  1.000000e-05      3  \n",
       "\n",
       "[2807 rows x 3 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#one hot encode automatically using the custom ProteinOneHotEncoder class\n",
    "enc = ProteinOneHotEncoder()\n",
    "#minimum number of mutations in a site\n",
    "threshold = 2\n",
    "#minimum number of observed mutations needed to include\n",
    "min_count = 10\n",
    "varying_sites = ProteinOneHotEncoder.get_varying_sites(binary_data.index.values,threshold=threshold,min_count=min_count)\n",
    "print(f'varying sites are {varying_sites}')\n",
    "binary_data.index = ProteinOneHotEncoder.transform_varying_sites(binary_data.index.values,threshold=threshold,min_count=min_count)\n",
    "binary_data\n",
    "continuous_data.index = ProteinOneHotEncoder.transform_varying_sites(continuous_data.index.values,threshold=threshold,min_count=min_count)\n",
    "continuous_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "78e72eed-3f0b-40cf-a26c-cb0e64d2e569",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<640x337 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 12800 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_encoded = enc.fit_transform(binary_data.index.values)\n",
    "X_encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "b3c2124b-b54f-490d-999b-89572f3fbe74",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_binary = binary_data['label'].values\n",
    "y_binary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "15b065fa-14a0-457c-a88a-38ff5e5d9402",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.02137247e-09, 7.18517314e-06, 3.35045073e-06, 1.04859909e-09,\n",
       "       2.03546242e-09, 2.00439237e-09, 1.51844583e-08, 3.08242085e-07,\n",
       "       1.05253740e-09, 3.45016700e-06, 1.40732248e-08, 1.51627072e-09,\n",
       "       1.00734390e-08, 3.01813946e-06, 1.00000000e-05, 1.00000000e-05,\n",
       "       2.28209870e-07, 1.97254749e-07, 8.50172517e-07, 1.32379896e-07,\n",
       "       3.33569212e-06, 2.94232330e-08, 3.79496604e-07, 6.97419048e-06,\n",
       "       2.20424907e-06, 1.61165298e-08, 4.90927268e-09, 7.26975197e-06,\n",
       "       6.28492423e-06, 2.41046439e-08, 1.00000000e-05, 3.30749355e-09,\n",
       "       3.34096389e-06, 1.75716838e-07, 8.56607430e-08, 1.23727251e-07,\n",
       "       2.76965480e-08, 1.00000000e-05, 4.75937867e-06, 8.13705441e-07,\n",
       "       9.83170333e-06, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.17368832e-08, 3.86029975e-06, 7.12007214e-06, 1.96943958e-07,\n",
       "       8.44095935e-10, 6.71189168e-06, 1.00000000e-05, 1.07669001e-08,\n",
       "       2.55973308e-09, 1.00000000e-05, 6.93507408e-07, 5.20413246e-09,\n",
       "       4.13811509e-09, 3.41121904e-06, 5.13350583e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.37377642e-09, 1.72938279e-08,\n",
       "       2.19988601e-09, 6.42652634e-08, 7.43517716e-07, 5.23472663e-09,\n",
       "       2.32224227e-08, 1.13871500e-09, 4.26213212e-09, 1.39737027e-09,\n",
       "       3.16227766e-10, 3.28743598e-09, 4.12466195e-09, 1.15649149e-09,\n",
       "       4.23187987e-09, 8.25432955e-09, 6.75211194e-08, 5.01592220e-07,\n",
       "       3.83458163e-08, 6.67129451e-06, 4.36928000e-09, 1.62824290e-08,\n",
       "       3.33533900e-06, 4.42254276e-09, 1.66858997e-06, 1.02889570e-09,\n",
       "       1.77993202e-09, 3.89354749e-09, 6.66814707e-06, 3.28899265e-09,\n",
       "       5.12928932e-09, 3.20071420e-09, 1.00000000e-05, 1.57207435e-06,\n",
       "       6.74259061e-06, 2.43858112e-08, 9.55470756e-06, 6.67673317e-06,\n",
       "       4.52334912e-08, 8.33364013e-06, 1.00000000e-05, 1.18285602e-06,\n",
       "       1.08160157e-06, 4.20974868e-06, 5.35286487e-06, 1.50880138e-07,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.09043447e-06, 1.82822844e-08,\n",
       "       3.75179765e-08, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       5.77856708e-06, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 6.67129451e-06, 1.00000000e-05, 5.00941010e-06,\n",
       "       1.00000000e-05, 2.89611118e-08, 1.58509349e-08, 6.01352346e-06,\n",
       "       1.41645081e-06, 1.41301923e-08, 2.79418294e-09, 1.14899034e-08,\n",
       "       6.24245475e-09, 2.88803697e-09, 4.10203356e-06, 9.63798980e-06,\n",
       "       5.38169733e-06, 7.24036220e-06, 2.46976464e-08, 1.00000000e-05,\n",
       "       9.14028728e-06, 1.00000000e-05, 1.00000000e-05, 8.33379827e-06,\n",
       "       4.55508820e-09, 1.79377048e-07, 2.91294677e-09, 6.01350838e-10,\n",
       "       6.57162635e-07, 3.25928321e-06, 1.66761903e-06, 1.00558648e-06,\n",
       "       2.79457411e-06, 2.72273680e-06, 1.00000000e-05, 3.61663748e-10,\n",
       "       1.54468978e-08, 2.70922750e-09, 3.53153096e-10, 5.69778113e-10,\n",
       "       6.25417430e-10, 5.30403461e-10, 1.41064279e-08, 1.90104439e-07,\n",
       "       4.43605817e-06, 2.52447127e-08, 1.00000000e-05, 2.83288824e-06,\n",
       "       1.15324747e-06, 2.22209483e-09, 2.97997186e-09, 6.12423098e-10,\n",
       "       1.00000000e-05, 1.12262508e-09, 3.16537136e-06, 3.21981784e-10,\n",
       "       4.50397610e-10, 1.27955232e-09, 4.86082944e-09, 4.13528707e-09,\n",
       "       4.84265550e-10, 7.45218333e-10, 3.16227766e-10, 5.66728325e-09,\n",
       "       3.16227766e-10, 4.57902163e-10, 5.08538010e-10, 4.31801410e-09,\n",
       "       1.69168786e-09, 2.04114227e-09, 4.67871131e-09, 8.28057456e-06,\n",
       "       2.75414743e-06, 1.00000000e-05, 4.83000884e-07, 3.70520413e-09,\n",
       "       1.00000000e-05, 1.89624547e-09, 1.08184366e-09, 4.11946364e-09,\n",
       "       8.67531992e-08, 4.59234869e-09, 1.36558328e-09, 1.42331179e-09,\n",
       "       2.69982499e-09, 1.00000000e-05, 5.85279462e-07, 4.68990319e-09,\n",
       "       6.77392659e-06, 1.92805718e-07, 2.74095027e-09, 4.32772199e-06,\n",
       "       1.00000000e-05, 3.37217901e-06, 1.00000000e-05, 4.03256416e-09,\n",
       "       1.88329716e-08, 1.00000000e-05, 9.97631608e-07, 3.62901950e-09,\n",
       "       3.50477572e-09, 2.06194337e-09, 1.00000000e-05, 1.73753048e-09,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 3.93941761e-09,\n",
       "       1.84220459e-08, 5.64954883e-07, 1.14294639e-06, 1.18949404e-06,\n",
       "       1.55699816e-07, 1.00000000e-05, 7.65097125e-08, 6.25907056e-09,\n",
       "       9.17670177e-07, 6.67565895e-06, 3.42866890e-09, 8.60131216e-07,\n",
       "       2.51791718e-08, 6.77150186e-08, 3.38716992e-07, 1.84585714e-07,\n",
       "       7.75046470e-08, 3.76901601e-06, 3.11915778e-09, 1.26575993e-08,\n",
       "       3.00623571e-06, 5.18062279e-09, 2.67460327e-07, 1.76238750e-07,\n",
       "       1.18263575e-09, 1.21966939e-08, 2.69475026e-09, 1.66286275e-09,\n",
       "       6.09596930e-08, 8.60674801e-09, 5.27378961e-09, 1.68336656e-06,\n",
       "       1.95649656e-09, 5.64177270e-07, 4.99757342e-09, 1.86888268e-06,\n",
       "       8.30904687e-10, 1.00000000e-05, 1.00000000e-05, 8.27523709e-08,\n",
       "       1.30576359e-08, 4.27507043e-09, 1.48130652e-08, 1.23896909e-09,\n",
       "       1.16003774e-08, 2.68421325e-07, 6.38296321e-09, 3.33932695e-06,\n",
       "       9.23930505e-06, 6.66734725e-06, 7.40862368e-06, 1.64048208e-08,\n",
       "       3.84333230e-09, 6.40610108e-07, 6.97266381e-09, 2.33270371e-09,\n",
       "       4.88297356e-09, 3.23885470e-08, 1.00000000e-05, 9.52017237e-06,\n",
       "       6.67574234e-06, 3.90035162e-09, 6.67528412e-06, 4.40105148e-10,\n",
       "       7.92888778e-10, 3.36686932e-08, 6.75720892e-09, 2.11853085e-08,\n",
       "       3.03589225e-09, 1.25558541e-09, 2.34377798e-09, 3.16227766e-10,\n",
       "       8.33486799e-10, 8.25962833e-09, 3.21832765e-10, 4.58711611e-06,\n",
       "       1.00000000e-05, 4.50801587e-08, 3.87779992e-10, 7.26114950e-10,\n",
       "       1.04811399e-09, 1.00000000e-05, 2.35943841e-09, 7.22616142e-10,\n",
       "       3.86579149e-10, 2.09334344e-09, 4.81234651e-09, 5.09777519e-08,\n",
       "       8.32236763e-10, 7.42465588e-09, 6.73875728e-06, 1.07665910e-08,\n",
       "       1.32967427e-06, 1.00000000e-05, 6.68356636e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 2.79499388e-09, 1.33464392e-09, 4.82870620e-09,\n",
       "       2.17503056e-09, 6.07922675e-09, 6.69095394e-06, 1.00000000e-05,\n",
       "       8.85546992e-08, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       7.31208500e-06, 3.19675283e-09, 1.00000000e-05, 8.71576234e-07,\n",
       "       5.17026507e-09, 1.14374169e-08, 1.09309545e-08, 3.65239722e-09,\n",
       "       1.47286473e-08, 5.12224980e-09, 2.47589158e-09, 1.06208209e-09,\n",
       "       1.00000000e-05, 6.66686295e-06, 4.06851221e-07, 2.84119114e-06,\n",
       "       9.33421692e-09, 1.04803261e-08, 1.62770746e-09, 1.00000000e-05,\n",
       "       1.21947215e-08, 6.72230559e-06, 2.38990438e-09, 3.33715410e-06,\n",
       "       1.00000000e-05, 8.34428169e-06, 7.62151237e-06, 3.46677899e-06,\n",
       "       1.21061035e-07, 6.66786998e-06, 6.06076768e-09, 4.29599714e-07,\n",
       "       3.36830776e-06, 3.63613035e-09, 3.54315369e-06, 3.36628243e-06,\n",
       "       7.88536956e-08, 6.67842589e-06, 2.47234315e-09, 8.33344219e-06,\n",
       "       1.20128109e-09, 8.12171944e-06, 1.11208766e-08, 2.29684337e-09,\n",
       "       1.08618235e-09, 2.25530441e-09, 7.86638783e-09, 1.00000000e-05,\n",
       "       1.45356708e-09, 1.28640796e-09, 1.77881284e-09, 1.50408533e-08,\n",
       "       4.19672585e-09, 5.75518908e-06, 1.30609475e-09, 1.14494537e-09,\n",
       "       5.26144186e-09, 8.23685268e-10, 1.88033556e-09, 6.66915197e-06,\n",
       "       1.20202987e-08, 4.33466390e-09, 2.35731935e-06, 1.39476654e-07,\n",
       "       8.18883470e-09, 1.16760852e-09, 2.48987933e-09, 1.00000000e-05,\n",
       "       4.28882557e-09, 1.62584598e-09, 3.29798854e-06, 1.40540699e-09,\n",
       "       3.62947651e-08, 1.00000000e-05, 1.98224349e-09, 3.99008436e-09,\n",
       "       2.18178209e-09, 1.57342988e-09, 4.51613184e-10, 2.31289606e-09,\n",
       "       2.63702387e-09, 1.93880316e-09, 3.35418923e-06, 1.00000000e-05,\n",
       "       3.59630640e-08, 1.95567179e-09, 8.43641794e-09, 5.57857880e-10,\n",
       "       1.00000000e-05, 2.30785859e-06, 5.55972352e-06, 1.07747725e-09,\n",
       "       2.34164665e-09, 1.99352967e-09, 8.28945510e-09, 3.17502836e-08,\n",
       "       2.76586687e-09, 7.70168071e-09, 5.58643534e-09, 2.78746500e-09,\n",
       "       7.02590175e-06, 2.97691217e-09, 1.45173669e-09, 1.44618631e-07,\n",
       "       1.35043814e-09, 2.11608248e-06, 1.25351838e-09, 2.54039749e-09,\n",
       "       5.22796046e-09, 7.07546572e-07, 2.07801932e-09, 2.10108984e-09,\n",
       "       1.66231724e-09, 9.10726245e-10, 1.23971037e-09, 9.55862384e-10,\n",
       "       6.37770209e-10, 7.43976261e-07, 4.13893066e-09, 3.13353273e-08,\n",
       "       1.41413189e-09, 1.34307308e-08, 3.43901101e-09, 2.28254174e-09,\n",
       "       2.45460890e-09, 1.65966271e-09, 8.52147196e-10, 1.41710561e-09,\n",
       "       1.63760123e-09, 3.88181368e-08, 3.23753932e-08, 3.33565593e-06,\n",
       "       2.30956560e-09, 1.58589996e-09, 1.03682430e-09, 2.74004913e-09,\n",
       "       1.70021654e-08, 8.48807974e-09, 3.33407661e-06, 5.58061168e-10,\n",
       "       1.00583978e-09, 2.48806059e-09, 1.90747434e-09, 1.65048583e-08,\n",
       "       2.05191000e-07, 1.77977379e-07, 1.03863449e-08, 2.34463682e-09,\n",
       "       6.69221987e-07, 4.13406323e-09, 1.00000000e-05, 1.29327160e-09,\n",
       "       3.99089622e-09, 8.40322287e-10, 5.53981080e-10, 3.83222059e-09,\n",
       "       9.95873897e-10, 7.03407830e-09, 8.17222466e-09, 3.92053345e-09,\n",
       "       2.12401420e-09, 1.44656712e-09, 2.69588308e-09, 1.07050817e-08,\n",
       "       7.91226708e-06, 5.54051344e-09, 1.05762629e-09, 5.60409463e-07,\n",
       "       6.82788686e-09, 1.00000000e-05, 3.33748101e-06, 3.83854350e-07,\n",
       "       8.53726225e-08, 1.24912918e-09, 1.73577061e-08, 3.62475853e-09,\n",
       "       4.64770362e-09, 6.74662776e-06, 7.62801050e-06, 7.48962670e-06,\n",
       "       6.82347847e-06, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       4.00474566e-06, 1.00000000e-05, 3.42705558e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 6.67367119e-06, 1.00000000e-05, 5.34581264e-06,\n",
       "       4.87707694e-06, 2.98451597e-06, 9.52017237e-06, 1.00000000e-05,\n",
       "       6.73511750e-06, 6.86984563e-06, 9.37610172e-06, 6.74662776e-06,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       6.70342694e-06, 1.00000000e-05, 4.41954965e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 9.10941511e-06, 8.34428169e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 3.40208125e-06,\n",
       "       8.98608318e-06, 7.00776433e-06, 1.00000000e-05, 9.67190379e-06,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       7.73296504e-06, 1.00000000e-05, 9.10941511e-06, 7.91226708e-06,\n",
       "       6.67528412e-06, 1.00000000e-05, 1.00000000e-05, 7.56466837e-06,\n",
       "       1.00000000e-05, 8.28057456e-06, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 9.23930505e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       9.23930505e-06, 1.00000000e-05, 1.00000000e-05, 6.69224538e-06,\n",
       "       1.00000000e-05, 1.00000000e-05, 1.00000000e-05, 7.44807627e-06,\n",
       "       6.09995776e-06, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 6.67842589e-06, 1.00000000e-05, 1.00000000e-05,\n",
       "       6.73511750e-06, 1.00000000e-05, 1.00000000e-05, 1.00000000e-05,\n",
       "       6.84984696e-06, 6.74662776e-06, 1.00000000e-05, 1.00000000e-05,\n",
       "       6.73511750e-06, 1.00000000e-05, 1.00000000e-05, 6.67367119e-06,\n",
       "       1.00000000e-05, 3.36669628e-06, 9.23930505e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 9.83170333e-06, 1.00000000e-05,\n",
       "       1.00000000e-05, 6.95866391e-06, 1.00000000e-05, 1.00000000e-05,\n",
       "       1.00000000e-05, 1.00000000e-05, 7.21039097e-06, 1.00000000e-05])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_continuous = continuous_data.loc[binary_data.index,'mean'].values\n",
    "y_continuous"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "52caa6dd-1772-4688-ab41-059d193f2bc6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X is (640, 337), shape of continuous y is (640,), shape of binary y is (640,)\n"
     ]
    }
   ],
   "source": [
    "print(f'Shape of X is {X_encoded.shape}, shape of continuous y is {y_continuous.shape}, shape of binary y is {y_binary.shape}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "89de5b17-86a8-4f0d-a539-16776b4a4dd5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#scale output\n",
    "scaler = StandardScaler()\n",
    "y_continuous = scaler.fit_transform(y_continuous.reshape(-1, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "d882e949-3025-4498-b6fd-13d4b1faa9dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'continuous target value')"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_continuous,y_binary) #box and whisker? or shift the data points with some noise to see the density\n",
    "plt.ylabel('discrete target value')\n",
    "plt.xlabel('continuous target value')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a60e6ccf-9f12-444a-99b1-27a435c40489",
   "metadata": {},
   "source": [
    "# test models automatically"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "0459c418-c4b1-4019-a947-8a094cc8266a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Binary mode : LinearRegression: -0.36\n",
      "Continuous mode : LinearRegression: 0.56\n",
      "Binary mode : Ridge: 0.10\n",
      "Continuous mode : Ridge: 0.69\n",
      "Binary mode : Lasso: 0.04\n",
      "Continuous mode : Lasso: 0.46\n",
      "Binary mode : SVR: 0.18\n",
      "Continuous mode : SVR: 0.60\n",
      "Binary mode : KNeighborsRegressor: -0.07\n",
      "Continuous mode : KNeighborsRegressor: 0.13\n",
      "Binary mode : DecisionTreeRegressor: -0.25\n",
      "Continuous mode : DecisionTreeRegressor: 0.49\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/sw/pkgs/arc/python3.9-anaconda/2021.11/lib/python3.9/site-packages/sklearn/utils/validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Binary mode : RandomForestRegressor: 0.15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/marcase/titeseq/titeseq_modeling.py:73: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  model.fit(self.X_train, self.y_train_continuous)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Continuous mode : RandomForestRegressor: 0.63\n"
     ]
    }
   ],
   "source": [
    "#test regression models potential of modeling classification tasks and regression tasks\n",
    "tester = ModelTester(X_encoded, y_binary,y_continuous)\n",
    "models = [linear_model.LinearRegression(),linear_model.Ridge(alpha=0.5),linear_model.Lasso(alpha=0.01),svm.SVR(),KNeighborsRegressor(), DecisionTreeRegressor(), RandomForestRegressor()]\n",
    "tester.test_models(models)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "6cf8710b-21c5-4cd1-92f3-5d957a8e5ece",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifier mode : RidgeClassifier: 0.80\n",
      "Regressor mode : Ridge: 0.69\n",
      "Classifier mode : DecisionTreeClassifier: 0.80\n",
      "Regressor mode : DecisionTreeRegressor: 0.52\n",
      "Classifier mode : RandomForestClassifier: 0.78\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/marcase/titeseq/titeseq_modeling.py:103: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  regressor.fit(self.X_train, self.y_train_continuous)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regressor mode : RandomForestRegressor: 0.65\n"
     ]
    }
   ],
   "source": [
    "#test classifier models and regression models in tandem\n",
    "classifier_models = [linear_model.RidgeClassifier(alpha=0.5),DecisionTreeClassifier(),RandomForestClassifier()]\n",
    "regressor_models = [linear_model.Ridge(alpha=0.5),DecisionTreeRegressor(),RandomForestRegressor()]\n",
    "tester.test_models_paired(classifier_models,regressor_models)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4ea7e2e-e271-43b3-80f4-01e6b0a84bfd",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "## test linear models (manually)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "175d2c13-94a1-48c6-a162-139b5674f7a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train_binary, y_test_binary = train_test_split(X, y_binary, test_size=0.3, random_state=42)\n",
    "_, _, y_train_continuous, y_test_continuous = train_test_split(X, y_continuous, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04ea68a6-036a-466f-85fb-a43fc9c9d6d0",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "### linear regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "99118fb0-4ee9-4bea-9b76-ebbeb73d37a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = linear_model.LinearRegression().fit(X_train, y_train_continuous)\n",
    "reg_bin = linear_model.LinearRegression().fit(X_train, y_train_binary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "6b29398c-6fed-4acb-8871-27fbbe0ed4a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x15343da30eb0>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train_pred = reg.predict(X_train)\n",
    "plt.scatter(y_train_continuous,y_train_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Training performance')\n",
    "y_train_bin_pred = reg_bin.predict(X_train)\n",
    "plt.scatter(y_train_binary,y_train_bin_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "d13422aa-207a-4a4b-8fdf-75b0f8948ae8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on continuous train: 1.0\n",
      "Accuracy on binary train: 1.0\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy on continuous train: {metrics.r2_score(y_train_continuous,y_train_pred)}')\n",
    "print(f'Accuracy on binary train: {metrics.r2_score(y_train_binary,y_train_bin_pred)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "61f02b5c-dbb0-4891-a79a-a602fd724ad1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x15342c2cab50>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test_pred= reg.predict(X_test)\n",
    "plt.scatter(y_test_continuous,y_test_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Test performance')\n",
    "y_test_bin_pred = reg_bin.predict(X_test)\n",
    "plt.scatter(y_test_binary,y_test_bin_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "6ba1770d-7eee-4bbd-b8e2-1f9fd67dd27e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on continuous test: 0.0002241266893779903\n",
      "Accuracy on binary test: -0.009344008879415666\n"
     ]
    }
   ],
   "source": [
    "print(f'Accuracy on continuous test: {metrics.r2_score(y_test_continuous,y_test_pred)}')\n",
    "print(f'Accuracy on binary test: {metrics.r2_score(y_test_binary,y_test_bin_pred)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a54992a2-98db-4436-864f-40087d2a56b4",
   "metadata": {
    "tags": []
   },
   "source": [
    "### ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "c1570171-77fe-4b93-b91b-2081e80d62ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = linear_model.Ridge(alpha=0.5).fit(X_train,y_train_continuous)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "79390fde-02eb-4f13-b100-33ce656d4f4b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Training performance')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train_pred = reg.predict(X_train)\n",
    "plt.scatter(y_train_continuous,y_train_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Training performance')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "63644ecb-4b3a-4afb-b4d8-f5fa0579eb88",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Test performance')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test_pred = reg.predict(X_test)\n",
    "plt.scatter(y_test_continuous,y_test_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Test performance')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e34e0f0-e1a6-4e47-a352-6c9df768b463",
   "metadata": {},
   "source": [
    "### lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "98214491-a16b-493a-9fd8-46534426d06f",
   "metadata": {},
   "outputs": [],
   "source": [
    "reg = linear_model.Lasso(alpha=0.001).fit(X_train,y_train_continuous)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "18a742df-5eb8-4fb6-8ac1-d1fabc7f02aa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Training performance')"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train_pred = reg.predict(X_train)\n",
    "plt.scatter(y_train_continuous,y_train_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Training performance')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "56fa5808-8da1-4c07-8e54-8740440278ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Test performance')"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEWCAYAAABIVsEJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAiVUlEQVR4nO3de7xVdZ3/8debw7HATCABgSTUIWZKhPKUZNPFhCathEpGKX8xXYaxqSzn96twbLLr5ISP+U0zzeQwdqE0yvJGpcmlzC4/ySOa6CiRqagwcLoIXjBRPr8/1vfgZrMva6999tnn8n4+Hvux1+X7Xd/PXmft/Tnr9l2KCMzMzBo1ot0BmJnZ4OQEYmZmhTiBmJlZIU4gZmZWiBOImZkV4gRiZmaFOIGYtYmkN0m6X9Ijkl7U7njMGuUEYoNS+tHtfe2VtLtk/G0Flne9pHe3ItYaLgTeFxHPiohb+rlts6aNbHcAZkVExLN6hyXdC7w7Ita2L6L8JI2MiCeB5wF3FFxGR0Q81beRmTXGeyA2pEgaIWmppLsl/U7SZZLGpXnPlHRJmv6QpJskTZT0GeAVwBfSHswXKix3mqSQtETSVknbJP3vnO321n2XpC3ATyQ9AnQAv5R0dyr3Z2lP6CFJd0g6tWT5X5X0RUnXSHoUOFHSvZI+JOk2SY9K+lL6PNdKeljSWkljS5bxbUn/I2mnpBskvbBs+f8u6fup7npJR5fMf6GkNZJ+L2m7pL+v97lt6HMCsaHmbGAB8CpgMvAH4N/TvMXAocARwHOAs4DdEXEe8BOePpz0vhrLPxGYDrwWWCppbo52e70K+DPgNSV7ULMi4mhJncB3gdXABOD9wKWSZpTUfyvwGeAQ4Kdp2luAecDzgTcC1wJ/DxxG9v0+u6T+tSn2CcAG4NKy+BYBnwDGAr9ObSHpEGAt8IP02f4EWNfA57ahKiL88mtQv4B7gblp+E7gpJJ5k4A9ZIdr3wn8HDi2wjKuJzsMVq2NaUAAf1oy7XPAl3K021v3qLJlBvAnafgVwP8AI0rmrwQ+noa/Cnytwud+W8n45cAXS8bfD1xV5fOMSe0fWrL8i0vmnwLclYYXAbdUWU7Vz93u7cKv1r98DsSGmucBV0raWzLtKWAi8HWyvY9vShoDXAKcFxF7Glj+/SXD9wEzc7RbqW65ycD9EVFa/z5gSp3620uGd1cYfxZk50zI9igWAuOB3nYOA3am4f8pqftYb12ydXZ3lbhrfe4Hq9SxIcKHsGyouR84OSLGlLyeGREPRsSeiPhERLwAOAF4A/D2VC9vt9RHlAxPBbbWa7ekfK02tgJHSCr9Tk5l/x/hZrrOfiswH5hLdhhvWpquHHXvB46uMa/e57YhygnEhpqLgM9Ieh6ApPGS5qfhEyXNTP+N7yI71NJ7JdN24Kgcy/8HSaPTCeh3AN+q125O64FHgQ9L6pT0arJzGt9sYBm1HAL8EfgdMBr4xwbqfg84XNIHJT1D0iGSjk/zmv3cNog5gdhQ83lgFbBa0sPAjUDvj93hwHfIksedwI/JDmP11jtN0h8k/WuN5f+Y7ATzOuDCiFido926IuIJ4FTgZOC3wH8Ab4+Iu/Iuo46vkR0SexD47xRf3tgeJjtR/0ayw1ybyS4mgCY/tw1uivADpczqkTQNuAfojOweDrNhz3sgZmZWiBOImZkV4kNYZmZWiPdAzMyskGF1I+Fhhx0W06ZNa3cYZmaDys033/zbiBhfPn1YJZBp06bR3d3d7jDMzAYVSfdVmu5DWGZmVogTiJmZFeIEYmZmhTiBmJlZIU4gZmZWiBOImZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJmZoU4gZiZWSFOIGZmVogTiJmZFeIEYmZmhbQlgUgaJ2mNpM3pfWyFMjMk3Vry2iXpgyXz3y9pk6Q7JH2uXz+AmZm1bQ9kKbAuIqYD69L4fiJiU0TMjojZwHHAY8CVAJJOBOYDx0bEC4EL+ytwMzPLtCuBzAdWpOEVwII65U8C7o6I3qdivQe4ICL+CBARO1oRpJmZVdeuBDIxIrYBpPcJdcqfAawsGX8+8ApJ6yX9WNJLWhSnmZlV0bJnoktaCxxeYdZ5DS7nIOBU4NySySOBscAc4CXAZZKOioioUH8JsARg6tSpjTRtZmY1tCyBRMTcavMkbZc0KSK2SZoE1DoEdTKwISK2l0x7ALgiJYxfSNoLHAb0VIhjObAcoKur64AEY2ZmxbTrENYqYHEaXgxcXaPsIvY/fAVwFfAaAEnPBw4Cftu3IZqZWS3tSiAXAPMkbQbmpXEkTZZ0TW8hSaPT/CvK6n8ZOErS7cA3gcWVDl+ZmVnrtOwQVi0R8TuyK6vKp28FTikZfwx4ToVyTwBntjJGMzOrzXeim5lZIU4gZmZWiBOImZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJmZoU4gZiZWSFOIGZmVogTiJmZFeIEYmZmhTiBmJlZIU4gZmZWiBOImZkV4gRiZmaFtCWBSBonaY2kzel9bIUyMyTdWvLaJemDad63SqbfK+nW/v4MZmbDXbv2QJYC6yJiOrAuje8nIjZFxOyImA0cBzwGXJnmnV4y73IOfOStmZm1WLsSyHxgRRpeASyoU/4k4O6IuK90oiQBfwms7OsAzcystnYlkIkRsQ0gvU+oU/4MKieJVwDbI2JztYqSlkjqltTd09NTOGAzM9vfyFYtWNJa4PAKs85rcDkHAacC51aYvYg6ex8RsRxYDtDV1RWNtG1mZtW1LIFExNxq8yRtlzQpIrZJmgTsqLGok4ENEbG9bBkjgTeTnR8xM7N+1q5DWKuAxWl4MXB1jbLV9jLmAndFxAN9HJuZmeXQrgRyATBP0mZgXhpH0mRJ1/QWkjQ6za90lVW18yJmZtYPWnYIq5aI+B3ZlVXl07cCp5SMPwY8p8oy/qpV8ZmZWX2+E93MzApxAjEzs0KcQMzMrBAnEDMzK8QJxMzMCnECMTOzQpxAzMysECcQMzMrxAnEzMwKcQIxM7NCnEDMzKyQqn1hSdoIVH1+RkQc25KIzMxsUKjVmeIb0vt70/vX0/vbyJ5PbmZmw1jVBNL7/HFJL4+Il5fMWirpZ8AnWx2cmZkNXHnOgRws6c97RySdABzcupDMzGwwyPM8kHcBX5Z0KNk5kZ3AO1oalZmZDXhV90BSwiAibo6IWcCxwOyImE2TD6KSNE7SGkmb0/vYCmVmSLq15LVL0gfTvNmSbkzTuyW9tJl4zMyscbUOYa0r/WGPiF0RsVNStUfMNmIpsC4ipgPr0vh+ImJTRPQmrOPITtxfmWZ/DvhEmvexNG5mZv2oVgL5T+BHksb3TpD0VmA58Pom250PrEjDK4AFdcqfBNzde2Kf7FDas9PwocDWJuMxM7MG1boK678kPQ78UNJrgdOBs4ATI+LeJtudGBHbUjvbJE2oU/4MYGXJ+AeB6yRdSJYET6hWUdISYAnA1KlTm4nZzMxK1DyXERFfT0nkFmAL8PKI+F2eBUtaCxxeYdZ5jQQo6SDgVODcksnvAc6JiMsl/SXwJWBupfoRsZxsr4murq6qN0aamVlj8tyJLmA08ByyQ1oCot6d6BFR8Qc9LXu7pElp72MSsKPGok4GNkTE9pJpi4EPpOFvAxfXisXMzPpenjvRW2EVWRK4IL1fXaPsIvY/fAXZOY9XAdcDrwE2932IZmZWS9070VvkAuAySe8iOzS2EEDSZODiiDgljY8G5gF/U1b/r4HPSxoJPE46x2FmZv2nqfs5ikrnUU6qMH0rcErJ+GNkh87Ky/2U7NJeMzNrE3fnbmZmhTiBmJlZIX4eiJmZFeLngZiZWSF+HoiZmRXi54GYmVkhRZ8H8s6WRmVmZgNe3QQSETcDsyQ9G1BE7Gx9WGZmNtDVPYQlaaKkLwHfSs8DeUG6g9zMzIaxPOdAvgpcB0xO478i607dzMyGsTwJ5LCIuAzYCxARTwJPtTQqMzMb8PIkkEclPYd0U6GkOWQn0s3MbBjLcxXW35F1v350uv9jPHBaS6MyM7MBL89VWBskvQqYQfZwqU0RsaflkZmZ2YBWqy+sN1eZ9XxJRMQVLYrJzMwGgVp7IG+sMS8AJxAzs2GsVl9Y72hVo5LGAd8CpgH3An8ZEX8oKzMjlel1FPCxiPgXSbOAi4Bnpfpvi4hdrYrXzMwOlOt5IJJeL+nDkj7W+2qy3aXAuoiYDqxL4/uJiE0RMTsiZpM9ffAx4Mo0+2JgaUTMTNM+1GQ8ZmbWoDx3ol8EnA68n+wk+kLgeU22Ox9YkYZXAAvqlD8JuLvkOe0zgBvS8BrgLU3GY2ZmDcqzB3JCRLwd+ENEfAJ4GXBEk+1OjIhtAOl9Qp3yZwArS8ZvB05NwwtrxSNpiaRuSd09PT1NhGxmZqXyJJDd6f0xSZOBPcCR9SpJWivp9gqv+Y0EKOkgsmTx7ZLJ7wTeK+lm4BDgiWr1I2J5RHRFRNf48eMbadrMzGrIcyPh9ySNAZYBG8iuwLq4XqWImFttnqTtkiZFxDZJk4AdNRZ1MrAhIraXLPsu4LVpWc8HXp/jc5iZWR+quwcSEZ+KiIci4nKycx9/GhH/0GS7q4DFaXgxcHWNsovY//AVkiak9xHAR8muyDIzs36U5yT6e9MeCBHxR2CEpL9tst0LgHmSNgPz0jiSJku6pqTt0Wl++T0niyT9CrgL2Ap8pcl4zMysQYqI2gWkW9OltKXTbomIF7UysFbo6uqK7u7udodhZjaoSLo5IrrKp+c5iT5CkkoW1AEc1JfBmZnZ4JPnJPp1wGXpfpAAzgJ+0NKozMxswMuTQD4CLAHeQ3Yj4WpyXIVlZmZDW57u3PeSXeV0UerD6rkR4ScSmpkNc3muwrpe0rNT8rgV+Iqkf255ZGZmNqDlOYR1aETskvRu4CsRcb6k21od2EAxben3c5ftkHgqgrGjO4mAh3bv2TetlMhOJgkYfVAHjz7x1L5yU8aM4kN/MYMFL5rCVbc8yLLrNvHgQ7v31R0heMbIETy+Zy+TU1mAZddtYutDu/dNW/CiKQfE99GrNrJy/f08FUGHxKLjj+DTC2bm/ny98ZS2k7ftWssoLZ+3jW93b+Fnd/++YhtjR3dy/htfWDOOourF30rtbDuPPzn3+zxZ5aLOM+dM5dMLZvLRqzbyjfVb2FtSbsyoTj5+6gv3bfMfX3UHD+3Onlk3QrA3sr/p43ueYveevfvVe8OsSfzorp6a66Taequ1rZV+54p8V2r56FUbufTGLfSugmeMHMGozg527t6T++/ayLZQ6Tfs3gv65t7rPJfxbiS763sFcF5E3CTptog4tk8i6EeNXsbbSPLoS6M6O3jLcVO4/OYH2b2n9tHCzhECwZ6nnv47jurs4LNvnrnfBvXRqzZyyY1bDqjf+8Wu56pbHuTcKzbuF09nhyBgz97abddaRmn5vG3k0dkhlp02q09/YOvF30rtbDuPWsmj1/QJB7N5x6MV53WOEKe/9Ai+9Yv7G/5blypfJ9XWW6XvV71tLe93pZZq38Nan6FcI9tCrd+wRpJIM5fxfpLsSqxfp+RxFLA5d8vWsN17nmLl+vvrJg/INvbS5NFbf9l1m/abtnL9/RXrV5tebtl1mw6IZ89TccCXrVLbtZZRWj5vG3nseSqqxlFUvfhbqZ1t51EveQBVkwdk2/HK9c0lDzhwnVRbb5W+X/W2tbzflVryLKPe33UgbQt5TqJ/m5KODCPiN7j79JYrP+zVqK0lu+C1lpe3nfLlFSlbb3ojbTQTR18vr6/bGWht95dmt/lepeuk2vop0lZfxNcX37eBtC3keqCU9b+Op+/dLGTymFG5lpe3nfLlFSlbb3ojbTQTR18vr6/bGWht95dmt/lepeuk2vop0lZfxNcX37eBtC04gQxAozo7WHT8EYzq7KhbtnOEsmO3ZfV7Twj2WnR85UemVJte7kN/MeOAeDo7lJ2DqdN2rWWUls/bRh6dHaoaR1H14m+ldradx8gcf6LpEw6uOq9zRHaiusjfulT5Oqm23ip9v+pta3m/K7XkWUa9v+tA2hacQGpo9EqF3v8uxo7uZMyozv2mlVLJ+8EHdexXbsqYUXz2zTP59IKZfPbNM5lS9l/FCMGozhEolV22cBbLTpvFlDGj9k2rdDLt0wtmcuacqfva6ZAaOim44EVT9sWzr+3TZrFsYf22ay2jtHzeNv7l9Nm8/OhxVWMdO7qzz0+g54m/ldrZdh6//uzrayaRM+dMZc3fvZoz50yl/Dd6zKhOli2cxacXzGTZwln7vjvAvrJjR3cyqnPEAfXOnDO15jqptt5Kv1+VtrVSjX5Xaun9HpaugmeMHMGYUZ25/66NbAvVfsNafhWWpL+rVTEiBt29IO5M0cyscdWuwqp1Ev2Q9D4DeAnZMzwA3sjTzyM3M7NhqmoCSc8/R9Jq4MUR8XAa/zj7P17WzMyGoTznQKay/zPHnwCmtSQaMzMbNPIkkK8Dv5D0cUnnA+uBrzXTqKRxktZI2pzex1Ypd46kOyTdLmmlpGc2Ut/MzFonzzPRPwO8A/gD8BDwjoj4xybbXQqsi4jpwLo0vh9JU4Czga6IOAboAM7IW9/MzFor72W8o4FdEfF54AFJRzbZ7nyyvrVI7wuqlBsJjJI0MsWwtcH6ZmbWInm6cz+f7KFS56ZJncAlTbY7MSK2AaT3CeUFIuJB4EJgC7AN2BkRq/PWL4l/iaRuSd09PT1Nhm1mZr3y7IG8CTgVeBQgIrby9CW+VUlam85dlL/m5wksndeYDxwJTAYOlnRmnrqlImJ5RHRFRNf48eMbrW5mZlXkeR7IExERkgJAUvX+CEpExNxq8yRtlzQpIrZJmgTsqFBsLnBPRPSkOlcAJ5Dt/eSpb2ZmLZRnD+QySf8JjJH018Bamn8m+ipgcRpeDFxdocwWYI6k0ZIEnATc2UB9MzNroTzduV8oaR6wi+yu9I9FxJom272ALDG9iyxRLASQNBm4OCJOiYj1kr4DbACeBG4Blteqb2Zm/SfPEwn/KSI+Um/aYOC+sMzMGtfMEwnnVZh2cvMhmZnZYFb1EJak9wB/Cxwt6baSWYcAP291YGZmNrDVOgfyDeBa4LPsf6f3wxHx+5ZGZWZmA17VQ1gRsTMi7gU+D/w+Iu6LiPuAPZKO768AzcxsYMpzDuSLwCMl44+maWZmNozlSSCKkku1ImIv+W5ANDOzISxPAvmNpLMldabXB4DftDowMzMb2PIkkLPIuhB5EHgAOB5Y0sqgzMxs4MtzJ/oOnn4Oh5mZGVD7PpAPR8TnJP0bcMDt6hFxdksjMzOzAa3WHkhvx4Xu+8PMzA5QNYFExHfT+4pqZczMbPiqdQjru1Q4dNUrIk5tSURmZjYo1DqEdWF6fzNwOE8/xnYRcG8LYzIzs0Gg1iGsHwNI+lREvLJk1ncl3dDyyMzMbEDLcx/IeElH9Y5IOhLww8XNzIa5PAnkHOB6SddLuh74EfDBZhqVNE7SGkmb0/vYKuXOkXSHpNslrZT0zDR9YZq+V9IBDzkxM7PWq5tAIuIHwHTgA+k1IyKua7LdpcC6iJgOrGP/7uIBkDQFOBvoiohjgA6evqHxdrJzMz6UZmbWJnUTiKTRwIeA90XEL4Gpkt7QZLvzgd7Lg1cAC6qUGwmMkjQSGA1sBYiIOyNiU5MxmJlZE/IcwvoK8ATwsjT+APDpJtudGBHbANL7hPICEfEg2ZVgW4BtwM6IWN1oQ5KWSOqW1N3T09Nk2GZm1itPAjk6Ij4H7AGIiN2A6lWStDaduyh/zc8TWDovMh84EpgMHCzpzDx1S0XE8ojoioiu8eN97t/MrK/kea7HE5JGkW4qlHQ08Md6lSJibrV5krZLmhQR2yRNAnZUKDYXuCcielKdK8h6Bb6kQlkzM+tnefZAzgd+ABwh6VKyk94fbrLdVcDiNLwYuLpCmS3AHEmjJQk4iaf75zIzszarmUAkjQDGkl3x9FfASrKroq5vst0LgHmSNgPz0jiSJku6BiAi1gPfATYAG1Osy1O5N0l6gOy8zPclNXtVmJmZNUglT6utXEC6oexO9EGrq6srurvdubCZWSMk3RwRB9xzl+cQ1hpJ/0fSEekGwHGSxrUgRjMzG0TynER/Z3p/b8m0AI6qUNbMzIaJPI+0PbI/AjEzs8GlbgJJ/U/9LfDnZHsePwEuiojHWxybmZkNYHkOYX0NeBj4tzS+CPg6sLBVQZmZ2cCXJ4HMiIhZJeM/kvTLVgVkZmaDQ56rsG6RNKd3RNLxwM9aF5KZmQ0GefZAjgfeLmlLGp8K3ClpIxARcWzLojMzswErTwJ5XcujMDOzQSfPZbz39UcgZmY2uOQ5B2JmZnYAJxAzMyvECcTMzApxAjEzs0KcQMzMrBAnEDMzK6QtCSQ9U2SNpM3pfWyVcudIukPS7ZJWpo4dkbRM0l2SbpN0paQx/foBzMysbXsgS4F1ETGd7BnrS8sLSJoCnE32CN1jgA7gjDR7DXBMugv+V8C5/RK1mZnt064EMh9YkYZXAAuqlBsJjJI0EhgNbAWIiNUR8WQqcyPw3NaFamZmlbQrgUyMiG0A6X1CeYGIeBC4ENgCbAN2RsTqCst6J3BttYYkLZHULam7p6enT4I3M7MWJhBJa9O5i/LX/Jz1x5LtqRwJTAYOlnRmWZnzgCeBS6stJyKWR0RXRHSNHz+++AcyM7P95OlMsZCImFttnqTtkiZFxDZJk4AdFYrNBe6JiJ5U5wrgBOCSNL4YeANwUkREn38AMzOrqV2HsFYBi9PwYuDqCmW2AHMkjZYk4CTgTgBJrwM+ApwaEY/1Q7xmZlamXQnkAmCepM3AvDSOpMmSrgGIiPXAd4ANwMYU6/JU/wvAIcAaSbdKuqif4zczG/Y0nI7+dHV1RXd3d7vDMDMbVCTdHBFd5dN9J7qZmRXiBGJmZoU4gZiZWSFOIGZmVogTiJmZFeIEYmZmhTiBmJlZIU4gZmZWiBOImZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJmZoU4gZiZWSFtSSCSxklaI2lzeh9bpdw5ku5Iz1JfKemZafqnJN2WHia1WtLk/v0EZmbWrj2QpcC6iJgOrEvj+5E0BTgb6IqIY4AO4Iw0e1lEHBsRs4HvAR/rl6jNzGyfdiWQ+cCKNLwCWFCl3EhglKSRwGhgK0BE7CopczAwfB6raGY2QIxsU7sTI2IbQERskzShvEBEPCjpQmALsBtYHRGre+dL+gzwdmAncGL/hG1mZr1atgciaW06d1H+mp+z/liyPZUjgcnAwZLO7J0fEedFxBHApcD7aixniaRuSd09PT3NfSgzM9unZQkkIuZGxDEVXlcD2yVNAkjvOyosYi5wT0T0RMQe4ArghArlvgG8pUYcyyOiKyK6xo8f3/wHMzMzoH3nQFYBi9PwYuDqCmW2AHMkjZYk4CTgTgBJ00vKnQrc1cJYzcysgnYlkAuAeZI2A/PSOJImS7oGICLWA98BNgAbU6zLe+unw2G3Aa8FPtDP8ZuZDXuKGD4XMHV1dUV3d3e7wzAzG1Qk3RwRXeXTfSe6mZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJmZoU4gZiZWSFOIGZmVogTiJmZFeIEYmZmhTiBmJlZIU4gZmZWiBOImZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJmZoUMq0faSuoB7mt3HP3oMOC37Q6izbwOMl4PXgdQfB08LyLGl08cVglkuJHUXek5xsOJ10HG68HrAPp+HfgQlpmZFeIEYmZmhTiBDG3L2x3AAOB1kPF68DqAPl4HPgdiZmaFeA/EzMwKcQIxM7NCnECGEEnjJK2RtDm9j61S7l5JGyXdKqm7v+NsBUmvk7RJ0q8lLa0wX5L+Nc2/TdKL2xFnK+VYB6+WtDP93W+V9LF2xNlKkr4saYek26vMHw7bQb110GfbgRPI0LIUWBcR04F1abyaEyNi9lC4Ll5SB/DvwMnAC4BFkl5QVuxkYHp6LQG+2K9BtljOdQDwk/R3nx0Rn+zXIPvHV4HX1Zg/pLeD5KvUXgfQR9uBE8jQMh9YkYZXAAvaF0q/einw64j4TUQ8AXyTbF2Umg98LTI3AmMkTervQFsozzoY8iLiBuD3NYoM9e0gzzroM04gQ8vEiNgGkN4nVCkXwGpJN0ta0m/Rtc4U4P6S8QfStEbLDGZ5P9/LJP1S0rWSXtg/oQ0oQ307yKtPtoORfRmRtZ6ktcDhFWad18BiXh4RWyVNANZIuiv91zJYqcK08uvT85QZzPJ8vg1kfRo9IukU4CqyQznDyVDfDvLos+3AeyCDTETMjYhjKryuBrb37o6n9x1VlrE1ve8AriQ7/DGYPQAcUTL+XGBrgTKDWd3PFxG7IuKRNHwN0CnpsP4LcUAY6ttBXX25HTiBDC2rgMVpeDFwdXkBSQdLOqR3GHgtUPFqjUHkJmC6pCMlHQScQbYuSq0C3p6uwpkD7Ow93DdE1F0Hkg6XpDT8UrLv/+/6PdL2GurbQV19uR34ENbQcgFwmaR3AVuAhQCSJgMXR8QpwETgyrT9jAS+ERE/aFO8fSIinpT0PuA6oAP4ckTcIemsNP8i4BrgFODXwGPAO9oVbyvkXAenAe+R9CSwGzgjhlhXFJJWAq8GDpP0AHA+0AnDYzuAXOugz7YDd2ViZmaF+BCWmZkV4gRiZmaFOIGYmVkhTiBmZlaIE4iZmRXiBGJWJvVWekKTy3ikr+IpW+60ar2s1qjzVUmn1ZqeenK+RdIBl7VWq2/mBGJ2oFcDTSWQwUTSoWT3jyyPiK+0Ox4bPJxAbFiQdFXqPPKO0g4k0zM0NqSO5dZJmgacBZyTnpXwivL/wHv3LiQ9K9XZoOz5KjV7v029AHw/tXW7pNPT9JdI+nma/gtJh6Q9jZ+kZW+otEckqUPSMkk3KXu2xd+k6ZL0BUn/Len7VO9UE+BZwLVkN5R+sUB9G8Z8J7oNF++MiN9LGgXcJOlysn+g/gt4ZUTcI2lcKnMR8EhEXAiQ7uyv5HHgTRGxK/UldKOkVTXu6n0dsDUiXp+We2jqduRbwOkRcZOkZ5PdHbwDmBcRj0uaDqwEyp/d8i6yrjheIukZwM8krQZeBMwAZpL1PPDfwJerxPTPZL0U/N+SaW9qoL4NY94DseHibEm/BG4k60xvOjAHuCEi7gGIiEafoSDgHyXdBqwl6xZ8Yo3yG4G5kv5J0isiYifZD/W2iLgpxbArIp4k63rivyRtBL5N9pCocq8l69fpVmA98Jz0uV4JrIyIp1LHmT+sEdMPgfnKembu1Uh9G8a8B2JDnqRXA3OBl0XEY5KuB55JlgDy9OXzJOmfrdQJ3UFp+tuA8cBxEbFH0r1puRVFxK8kHUfWF9Nn097CVVViOAfYDsxKbT9e6aMB74+I6/abmHXRnbePom8CPwWukXRiRDzcG27O+jaMeQ/EhoNDgT+k5PGnZHseAP8PeJWkIyG7EilNfxg4pKT+vcBxaXg+qWO6tNwdKXmcCDyvVhCpU8vHIuIS4ELgxcBdwGRJL0llDpE0Mi17W0TsBf4XWQeJ5a4j6xSvM9V9vrIelm8AzkjnSCYBJ9aKKyL+hewRyFemQ2oN1bfhy3sgNhz8ADgrHWraRHYYi4joSSfUr5A0gnTeAfgu8J10Uvz9ZOdJrpb0C7If2kfTci8FviupG7iVLBnUMhNYJmkvsAd4T0Q8kU6m/1s6P7ObbG/pP4DLJS0EflTSZqmLgWnAhrRn1EP2GOMrgdeQHTL7FfDjeisoIj4i6SvA14G3Nlrfhif3xmtmZoX4EJaZmRXiBGJmZoU4gZiZWSFOIGZmVogTiJmZFeIEYmZmhTiBmJlZIf8fLJZYKxgJ/cAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_test_pred = reg.predict(X_test)\n",
    "plt.scatter(y_test_continuous,y_test_pred)\n",
    "plt.xlabel('actual scaled Kd')\n",
    "plt.ylabel('predicted scaled Kd')\n",
    "plt.title('Test performance')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39b2d082-5f4f-4a1a-b7d1-ab542f96eb1c",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "c19fac7e-70da-42e2-952d-a21431675c82",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/sw/pkgs/arc/python3.9-anaconda/2021.11/lib/python3.9/site-packages/sklearn/utils/validation.py:63: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  return f(*args, **kwargs)\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "Unknown label type: 'continuous'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_3487971/3670520311.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mreg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlinear_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLogisticRegression\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrandom_state\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/sw/pkgs/arc/python3.9-anaconda/2021.11/lib/python3.9/site-packages/sklearn/linear_model/_logistic.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight)\u001b[0m\n\u001b[1;32m   1345\u001b[0m                                    \u001b[0morder\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"C\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1346\u001b[0m                                    accept_large_sparse=solver != 'liblinear')\n\u001b[0;32m-> 1347\u001b[0;31m         \u001b[0mcheck_classification_targets\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1348\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclasses_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munique\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1349\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/sw/pkgs/arc/python3.9-anaconda/2021.11/lib/python3.9/site-packages/sklearn/utils/multiclass.py\u001b[0m in \u001b[0;36mcheck_classification_targets\u001b[0;34m(y)\u001b[0m\n\u001b[1;32m    181\u001b[0m     if y_type not in ['binary', 'multiclass', 'multiclass-multioutput',\n\u001b[1;32m    182\u001b[0m                       'multilabel-indicator', 'multilabel-sequences']:\n\u001b[0;32m--> 183\u001b[0;31m         \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Unknown label type: %r\"\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0my_type\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    184\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    185\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Unknown label type: 'continuous'"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6f3eee4c-856a-4cdb-8f1a-b313ba86ec34",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}