Fig4.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.gridspec import GridSpec\n",
    "import seaborn as sns\n",
    "from brokenaxes import brokenaxes\n",
    "from scipy.optimize import leastsq\n",
    "\n",
    "from sklearn.utils import assert_all_finite\n",
    "from sklearn.utils import check_consistent_length\n",
    "from sklearn.utils import column_or_1d, check_array\n",
    "from sklearn.utils.multiclass import type_of_target\n",
    "from sklearn.utils.extmath import stable_cumsum\n",
    "from sklearn.utils.sparsefuncs import count_nonzero\n",
    "from sklearn.exceptions import UndefinedMetricWarning\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.metrics.ranking import _binary_clf_curve, roc_curve\n",
    "\n",
    "import rpy2\n",
    "from rpy2.robjects.packages import importr\n",
    "# install bootlr\n",
    "# import rpy2.robjects.packages as rpackages\n",
    "# from rpy2.robjects.vectors import StrVector\n",
    "# utils = rpackages.importr('utils')\n",
    "# utils.chooseCRANmirror(ind=1)\n",
    "# packnames = ('binom', 'bootLR')\n",
    "# utils.install_packages(StrVector(packnames))\n",
    "\n",
    "from functools import partial\n",
    "import multiprocessing\n",
    "from multiprocessing import Pool"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Flatten _openen files\n",
    "# n is 5UTR length\n",
    "# t is downstream distance to start codon to include\n",
    "def openen_label(n, t, f):\n",
    "    d = pd.read_csv(f, sep='\\t', skiprows=2, nrows=(n+t), header=None)\n",
    "    d = d.set_index(0).stack().to_frame()\n",
    "    d = d.index.to_frame()\n",
    "    d.columns = [\"Position 'i' centered at the start codon\", \"Subsegment 'l'\"]\n",
    "    d = d.reset_index()[[\"Position 'i' centered at the start codon\", \"Subsegment 'l'\"]]\n",
    "    d[\"Position 'i' centered at the start codon\"] = d[\"Position 'i' centered at the start codon\"] - n\n",
    "    return d\n",
    "\n",
    "# Get opening energy at a specific subsegment 'l'\n",
    "# and position 'i' centered at the start codon\n",
    "def get_openen(n, t, f1, l, i, f2):\n",
    "    label = openen_label(n, t, f1)\n",
    "    d = pd.read_pickle(f2)\n",
    "    d = d.set_index('id')\n",
    "    d = d.T\n",
    "    d = pd.concat([label, d], axis=1)\n",
    "    o = d.loc[(d[\"Subsegment 'l'\"] == l) & (d[\"Position 'i' centered at the start codon\"] == i)]\n",
    "    o = o.T[2:]\n",
    "    o.columns = ['Opening energy']\n",
    "    return o\n",
    "\n",
    "def roc(outcomes, prediction):\n",
    "    fps, tps, thresholds = _binary_clf_curve(outcomes, prediction)\n",
    "    clf = pd.DataFrame([fps, tps, thresholds]).T\n",
    "    clf.columns = ['fps', 'tps', 'thresholds']\n",
    "    clf['fps'] = clf['fps'].astype(int)\n",
    "    clf['tps'] = clf['tps'].astype(int)\n",
    "    fpr, tpr, thresholds = roc_curve(outcomes, prediction, drop_intermediate=False)\n",
    "    r = pd.DataFrame([fpr, tpr, thresholds]).T\n",
    "    r.columns = ['fpr', 'tpr', 'thresholds']\n",
    "    df = pd.merge(clf, r, on='thresholds')\n",
    "    return df\n",
    "\n",
    "def run_bootlr(P, N, d):\n",
    "    r = rpy2.robjects.r\n",
    "    set_seed = r('set.seed')\n",
    "    set_seed(12345)\n",
    "    bootLR = importr('bootLR')\n",
    "    truePos=int(d[1])\n",
    "    trueNeg=N-int(d[0])\n",
    "    lr = bootLR.BayesianLR_test(truePos=truePos, totalDzPos=P, trueNeg=trueNeg, totalDzNeg=N)\n",
    "    lr = [lr.rx('posLR')[0][0], lr.rx('posLR.ci')[0][0], lr.rx('posLR.ci')[0][1],\n",
    "         lr.rx('negLR')[0][0], lr.rx('negLR.ci')[0][0], lr.rx('negLR.ci')[0][1]]\n",
    "    x = d\n",
    "    x.extend(lr)\n",
    "    print(x)\n",
    "\n",
    "# shamelessly taken from https://people.duke.edu/~ccc14/pcfb/analysis.html\n",
    "def logistic4(x, A, B, C, D):\n",
    "    \"\"\"4PL lgoistic equation.\"\"\"\n",
    "    return ((A-D)/(1.0+((x/C)**B))) + D\n",
    "\n",
    "def residuals(p, y, x):\n",
    "    \"\"\"Deviations of data from fitted 4PL curve\"\"\"\n",
    "    A,B,C,D = p\n",
    "    err = y-logistic4(x, A, B, C, D)\n",
    "    return err\n",
    "\n",
    "def peval(x, p):\n",
    "    \"\"\"Evaluated value at x with current parameters.\"\"\"\n",
    "    A,B,C,D = p\n",
    "    return logistic4(x, A, B, C, D)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = get_openen(71, 100, 'data/AaCD00331182_openen', 48, 24, 'data/openen.pET21_NESG.pkl')\n",
    "d = d.reset_index()\n",
    "d.columns = ['Accession', 'Opening energy']\n",
    "c = pd.read_csv('data/class.txt', sep='\\t')\n",
    "c = c.replace(2,1)\n",
    "d = pd.merge(d, c, on='Accession')\n",
    "d = roc(d['Class'], -d['Opening energy'])\n",
    "d = d.values.tolist() #fps tps thresholds fpr tpr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# not run\n",
    "print(['fps', 'tps', 'thresholds', 'fpr', 'tpr', 'posLR', 'posLR.ci.lower', 'posLR.ci.upper', 'negLR negLR.ci.lower', 'negLR.ci.upper'])\n",
    "p = Pool(16)\n",
    "bootlr = partial(run_bootlr, 8780, 2650)\n",
    "p.imap_unordered(bootlr, d)\n",
    "p.close()\n",
    "p.join()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```console\n",
    "python bootlr.py 2>&1 | bootlr.log\n",
    "# remove warning messages regarding invalid confidene intervals.\n",
    "# This happens when sensitivity is 1 and specificity is zero, and vice versa\n",
    "cat \\\n",
    "<(grep posLR bootlr.log | sed 's/,//g;s/ /\\t/g') \\\n",
    "<(grep -v posLR bootlr.log | sed 's/,//g;s/ /\\t/g;s/\\[//;s/\\]//;/^$/d;s/-//g' | sort -k3,3n | awk '!/[a-z]/ && NF>0 {sub(\"\\\\[\",\"\\n\");sub(\"\\\\]\",\"\"); print}' | sort -k8,8nr) \\\n",
    "> bootlr.txt\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read in bootstrap results\n",
    "d = pd.read_csv('data/bootlr.txt', sep='\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fps</th>\n",
       "      <th>tps</th>\n",
       "      <th>thresholds</th>\n",
       "      <th>fpr</th>\n",
       "      <th>tpr</th>\n",
       "      <th>posLR</th>\n",
       "      <th>posLR.ci.lower</th>\n",
       "      <th>posLR.ci.upper</th>\n",
       "      <th>negLR</th>\n",
       "      <th>negLR.ci.lower</th>\n",
       "      <th>negLR.ci.upper</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>4.3981</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.005239</td>\n",
       "      <td>3.470957</td>\n",
       "      <td>1.489764</td>\n",
       "      <td>16.098138</td>\n",
       "      <td>0.996265</td>\n",
       "      <td>0.994209</td>\n",
       "      <td>0.998452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>4.3549</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004897</td>\n",
       "      <td>3.244590</td>\n",
       "      <td>1.383560</td>\n",
       "      <td>14.879459</td>\n",
       "      <td>0.996607</td>\n",
       "      <td>0.994599</td>\n",
       "      <td>0.998772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>4.3619</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.005011</td>\n",
       "      <td>3.320046</td>\n",
       "      <td>1.395184</td>\n",
       "      <td>14.810967</td>\n",
       "      <td>0.996493</td>\n",
       "      <td>0.994462</td>\n",
       "      <td>0.998658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>4.3492</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004670</td>\n",
       "      <td>3.093679</td>\n",
       "      <td>1.313794</td>\n",
       "      <td>14.250706</td>\n",
       "      <td>0.996835</td>\n",
       "      <td>0.994854</td>\n",
       "      <td>0.998981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>4.3336</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004442</td>\n",
       "      <td>2.942768</td>\n",
       "      <td>1.245272</td>\n",
       "      <td>13.618436</td>\n",
       "      <td>0.997063</td>\n",
       "      <td>0.995110</td>\n",
       "      <td>0.999187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>4.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>4.3286</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004214</td>\n",
       "      <td>2.791856</td>\n",
       "      <td>1.149899</td>\n",
       "      <td>12.453189</td>\n",
       "      <td>0.997291</td>\n",
       "      <td>0.995366</td>\n",
       "      <td>0.999400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>4.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>4.2615</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.003872</td>\n",
       "      <td>2.565490</td>\n",
       "      <td>1.047595</td>\n",
       "      <td>11.571868</td>\n",
       "      <td>0.997633</td>\n",
       "      <td>0.995761</td>\n",
       "      <td>0.999720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>11.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>4.7429</td>\n",
       "      <td>0.004151</td>\n",
       "      <td>0.011503</td>\n",
       "      <td>2.771278</td>\n",
       "      <td>1.586098</td>\n",
       "      <td>5.917397</td>\n",
       "      <td>0.992617</td>\n",
       "      <td>0.989402</td>\n",
       "      <td>0.996053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>4.7424</td>\n",
       "      <td>0.004151</td>\n",
       "      <td>0.011390</td>\n",
       "      <td>2.743839</td>\n",
       "      <td>1.569390</td>\n",
       "      <td>5.857779</td>\n",
       "      <td>0.992731</td>\n",
       "      <td>0.989526</td>\n",
       "      <td>0.996161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>4.6590</td>\n",
       "      <td>0.003774</td>\n",
       "      <td>0.009681</td>\n",
       "      <td>2.565490</td>\n",
       "      <td>1.430836</td>\n",
       "      <td>5.796067</td>\n",
       "      <td>0.994070</td>\n",
       "      <td>0.991067</td>\n",
       "      <td>0.997309</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    fps    tps  thresholds       fpr       tpr     posLR  posLR.ci.lower  \\\n",
       "0   4.0   46.0      4.3981  0.001509  0.005239  3.470957        1.489764   \n",
       "1   4.0   43.0      4.3549  0.001509  0.004897  3.244590        1.383560   \n",
       "2   4.0   44.0      4.3619  0.001509  0.005011  3.320046        1.395184   \n",
       "3   4.0   41.0      4.3492  0.001509  0.004670  3.093679        1.313794   \n",
       "4   4.0   39.0      4.3336  0.001509  0.004442  2.942768        1.245272   \n",
       "5   4.0   37.0      4.3286  0.001509  0.004214  2.791856        1.149899   \n",
       "6   4.0   34.0      4.2615  0.001509  0.003872  2.565490        1.047595   \n",
       "7  11.0  101.0      4.7429  0.004151  0.011503  2.771278        1.586098   \n",
       "8  11.0  100.0      4.7424  0.004151  0.011390  2.743839        1.569390   \n",
       "9  10.0   85.0      4.6590  0.003774  0.009681  2.565490        1.430836   \n",
       "\n",
       "   posLR.ci.upper     negLR  negLR.ci.lower  negLR.ci.upper  \n",
       "0       16.098138  0.996265        0.994209        0.998452  \n",
       "1       14.879459  0.996607        0.994599        0.998772  \n",
       "2       14.810967  0.996493        0.994462        0.998658  \n",
       "3       14.250706  0.996835        0.994854        0.998981  \n",
       "4       13.618436  0.997063        0.995110        0.999187  \n",
       "5       12.453189  0.997291        0.995366        0.999400  \n",
       "6       11.571868  0.997633        0.995761        0.999720  \n",
       "7        5.917397  0.992617        0.989402        0.996053  \n",
       "8        5.857779  0.992731        0.989526        0.996161  \n",
       "9        5.796067  0.994070        0.991067        0.997309  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# rows containing Inf were removed due bootLR error messages\n",
    "#Error in run.BayesianLR.test(truePos = truePos, totalDzPos = totalDzPos, : In some of your draws, sensitivity was 1 at the same time that specificity was 0. \n",
    "#This most likely resulted because you had a sensitivity of 1 and a specificity of close to 0, or vice-versa. \n",
    "#The algorithm is not designed to handle this case, and may not ever be. \n",
    "#Please do not just re-run the algorithm until you no longer receive this message, as the confidence intervals so obtained will be invalid.\n",
    "d.sort_values('posLR.ci.upper', ascending=False).head(n=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Least-squares 4 parameter logistic regression fit to positive likelihood ratio\n",
    "p0 = [1, 1, 1, 0]\n",
    "plsq = leastsq(residuals, p0, args=(d['posLR'], d['thresholds']))\n",
    "d['posLR_4PL'] = peval(d['thresholds'],plsq[0])\n",
    "d['tns'] = 2650-d['fps']\n",
    "d['fns'] = 8780-d['tps']\n",
    "d = d.sort_values('thresholds')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fps</th>\n",
       "      <th>tps</th>\n",
       "      <th>thresholds</th>\n",
       "      <th>fpr</th>\n",
       "      <th>tpr</th>\n",
       "      <th>posLR</th>\n",
       "      <th>posLR.ci.lower</th>\n",
       "      <th>posLR.ci.upper</th>\n",
       "      <th>negLR</th>\n",
       "      <th>negLR.ci.lower</th>\n",
       "      <th>negLR.ci.upper</th>\n",
       "      <th>posLR_4PL</th>\n",
       "      <th>tns</th>\n",
       "      <th>fns</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>4.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>4.2615</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.003872</td>\n",
       "      <td>2.565490</td>\n",
       "      <td>1.047595</td>\n",
       "      <td>11.571868</td>\n",
       "      <td>0.997633</td>\n",
       "      <td>0.995761</td>\n",
       "      <td>0.999720</td>\n",
       "      <td>2.383108</td>\n",
       "      <td>2646.0</td>\n",
       "      <td>8746.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>4.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>4.3286</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004214</td>\n",
       "      <td>2.791856</td>\n",
       "      <td>1.149899</td>\n",
       "      <td>12.453189</td>\n",
       "      <td>0.997291</td>\n",
       "      <td>0.995366</td>\n",
       "      <td>0.999400</td>\n",
       "      <td>2.382916</td>\n",
       "      <td>2646.0</td>\n",
       "      <td>8743.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>4.3336</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004442</td>\n",
       "      <td>2.942768</td>\n",
       "      <td>1.245272</td>\n",
       "      <td>13.618436</td>\n",
       "      <td>0.997063</td>\n",
       "      <td>0.995110</td>\n",
       "      <td>0.999187</td>\n",
       "      <td>2.382901</td>\n",
       "      <td>2646.0</td>\n",
       "      <td>8741.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "      <td>41.0</td>\n",
       "      <td>4.3492</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004670</td>\n",
       "      <td>3.093679</td>\n",
       "      <td>1.313794</td>\n",
       "      <td>14.250706</td>\n",
       "      <td>0.996835</td>\n",
       "      <td>0.994854</td>\n",
       "      <td>0.998981</td>\n",
       "      <td>2.382854</td>\n",
       "      <td>2646.0</td>\n",
       "      <td>8739.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>4.3549</td>\n",
       "      <td>0.001509</td>\n",
       "      <td>0.004897</td>\n",
       "      <td>3.244590</td>\n",
       "      <td>1.383560</td>\n",
       "      <td>14.879459</td>\n",
       "      <td>0.996607</td>\n",
       "      <td>0.994599</td>\n",
       "      <td>0.998772</td>\n",
       "      <td>2.382837</td>\n",
       "      <td>2646.0</td>\n",
       "      <td>8737.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fps   tps  thresholds       fpr       tpr     posLR  posLR.ci.lower  \\\n",
       "6  4.0  34.0      4.2615  0.001509  0.003872  2.565490        1.047595   \n",
       "5  4.0  37.0      4.3286  0.001509  0.004214  2.791856        1.149899   \n",
       "4  4.0  39.0      4.3336  0.001509  0.004442  2.942768        1.245272   \n",
       "3  4.0  41.0      4.3492  0.001509  0.004670  3.093679        1.313794   \n",
       "1  4.0  43.0      4.3549  0.001509  0.004897  3.244590        1.383560   \n",
       "\n",
       "   posLR.ci.upper     negLR  negLR.ci.lower  negLR.ci.upper  posLR_4PL  \\\n",
       "6       11.571868  0.997633        0.995761        0.999720   2.383108   \n",
       "5       12.453189  0.997291        0.995366        0.999400   2.382916   \n",
       "4       13.618436  0.997063        0.995110        0.999187   2.382901   \n",
       "3       14.250706  0.996835        0.994854        0.998981   2.382854   \n",
       "1       14.879459  0.996607        0.994599        0.998772   2.382837   \n",
       "\n",
       "      tns     fns  \n",
       "6  2646.0  8746.0  \n",
       "5  2646.0  8743.0  \n",
       "4  2646.0  8741.0  \n",
       "3  2646.0  8739.0  \n",
       "1  2646.0  8737.0  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate R-squared\n",
    "p = plsq[0]\n",
    "res = residuals(p, d['posLR'], d['thresholds'])\n",
    "ss_res = np.sum(res**2)\n",
    "ss_tot = np.sum((d['posLR']-np.mean(d['posLR']))**2)\n",
    "r_squared = 1 - (ss_res / ss_tot)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9452714822630413"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_squared"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distribution residuals is not perfectly symmetrical but R squared > 0.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc8242dd400>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 248,
       "width": 369
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "sns.distplot(res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.collections.PolyCollection at 0x7fc806200438>,\n",
       " <matplotlib.collections.PolyCollection at 0x7fc8062006d8>]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 438,
       "width": 358
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "sps1, sps2 = GridSpec(2,1)\n",
    "fig = plt.figure(figsize=(5,10))\n",
    "sns.set(font_scale=1.2)\n",
    "sns.set_style('whitegrid')\n",
    "\n",
    "bax = brokenaxes(subplot_spec=sps1)\n",
    "bax = plt.subplot2grid((4,1), (1,0)) \n",
    "bax.plot('thresholds', 'tns', data=d, color='blue', label='True Negative')\n",
    "bax.plot('thresholds', 'fns', data=d, color='lightblue', label='False Negative')\n",
    "bax.plot('thresholds', 'tps', data=d, color='red', label='True positive')\n",
    "bax.plot('thresholds', 'fps', data=d, color='orange', label='False positive')\n",
    "bax.set_ylabel('Frequency')\n",
    "# bax.legend(loc=1)\n",
    "\n",
    "bax = brokenaxes(ylims=((0.7, 3.6), (15.5, 16.2)), hspace=.05, subplot_spec=sps2)\n",
    "bax.plot('thresholds', 'posLR_4PL', data=d, label='4 parameter logistic curve')\n",
    "bax.scatter('thresholds', 'posLR', alpha=0.5, color='orchid', data=d, s=10, facecolors='none', label='Positive likelihood ratio')\n",
    "bax.fill_between('thresholds', 'posLR.ci.lower', 'posLR.ci.upper', color='#888888', alpha=0.3, data=d, label='Confidence intervals')\n",
    "# bax.set_xlabel('Opening energy')\n",
    "# bax.set_ylabel('Positive likelihood ratio')\n",
    "# bax.legend(loc=1)\n",
    "# plt.savefig('data/fig5.png', dpi=600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# export table for calculation of posterior probability and score\n",
    "d.to_csv('data/4pl.csv', index=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}