{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Intersect bed on gene-centric features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fri Mar 13 08:02:12 PDT 2015\r\n"
     ]
    }
   ],
   "source": [
    "!date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "import scipy.stats as stats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature (from nb -03)    \n",
    "\n",
    "**tldr** 4 \"new\" tracks\n",
    "<img src=\"http://eagle.fish.washington.edu/cnidarian/skitch/IGV_and_Directory_Listing_of__halfshell_2015-02-hs-bedgraph__1AA51F1B.png\" width =100% alt=\"IGV_and_Directory_Listing_of__halfshell_2015-02-hs-bedgraph__1AA51F1B.png\"/>\n",
    "```\n",
    "/Users/sr320/data-genomic/tentacle/Cuffdiff_geneexp.sig.gtf\n",
    "/Users/sr320/data-genomic/tentacle/rebuilt.gtf\n",
    "/Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-housekeeping.gff\n",
    "/Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-env-response.gff\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DEGs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`-wb\tWrite the original entry in B for each overlap. Useful for knowing what A overlaps. Restricted by -f and -r.`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 880 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.2M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cuffdiff_geneexp.sig.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 704 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.4M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cuffdiff_geneexp.sig.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 632 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.6M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cuffdiff_geneexp.sig.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "117460 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a /Users/sr320/git-repos/paper-Temp-stress/ipynb/data/array-design/OID40453_probe_locations.gff \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cuffdiff_geneexp.sig.gtf \\\n",
    "| cut -f 11 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 352.138, with p=0.000\n",
      "The uncorrected chi2 value is 352.654, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 2 then Probes\n",
    "obs = array([[880, 10028], [117460, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 547.532, with p=0.000\n",
      "The uncorrected chi2 value is 548.178, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 4 then Probes\n",
    "obs = array([[704, 10148], [117460, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 853.613, with p=0.000\n",
      "The uncorrected chi2 value is 854.371, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 6 then Probes\n",
    "obs = array([[632, 11690], [117460, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Rebuilt (new gtf based on RNAseq data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8768 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.2M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/rebuilt.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7694 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.4M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/rebuilt.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6160 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.6M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/rebuilt.gtf \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1197818 Cufflinks\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a /Users/sr320/git-repos/paper-Temp-stress/ipynb/data/array-design/OID40453_probe_locations.gff \\\n",
    "-b /Users/sr320/data-genomic/tentacle/rebuilt.gtf \\\n",
    "| cut -f 11 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 2184.818, with p=0.000\n",
      "The uncorrected chi2 value is 2185.528, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 2 then Probes\n",
    "obs = array([[8768, 10028], [1197818, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 3052.863, with p=0.000\n",
      "The uncorrected chi2 value is 3053.724, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 4 then Probes\n",
    "obs = array([[7694, 10148], [1197818, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 6233.645, with p=0.000\n",
      "The uncorrected chi2 value is 6234.874, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 6 then Probes\n",
    "obs = array([[6160, 11690], [1197818, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Housekeeping Genes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3210 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.2M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-housekeeping.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3369 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.4M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-housekeeping.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3819 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.6M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-housekeeping.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "251970 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a /Users/sr320/git-repos/paper-Temp-stress/ipynb/data/array-design/OID40453_probe_locations.gff \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-housekeeping.gff \\\n",
    "| cut -f 11 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 34.806, with p=0.000\n",
      "The uncorrected chi2 value is 34.923, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 2 then Probes\n",
    "obs = array([[3210, 10028], [251970, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 17.578, with p=0.000\n",
      "The uncorrected chi2 value is 17.661, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 4 then Probes\n",
    "obs = array([[3369, 10148], [251970, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 28.378, with p=0.000\n",
      "The uncorrected chi2 value is 28.476, with p=0.000\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 6 then Probes\n",
    "obs = array([[3819, 11690], [251970, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Environmental Response Genes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2809 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.2M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-env-response.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2738 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.4M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-env-response.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3216 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a ./data/2014.07.02.colson/genomeBrowserTracks/logFC_HS-preHS/2014.07.02.6M_sig.bedGraph \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-env-response.gff \\\n",
    "| cut -f 6 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "190475 GLEAN\r\n"
     ]
    }
   ],
   "source": [
    "!intersectbed \\\n",
    "-wb \\\n",
    "-a /Users/sr320/git-repos/paper-Temp-stress/ipynb/data/array-design/OID40453_probe_locations.gff \\\n",
    "-b /Users/sr320/data-genomic/tentacle/Cgigas_v9_gene-env-response.gff \\\n",
    "| cut -f 11 \\\n",
    "| sort | uniq -c "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 1.413, with p=0.235\n",
      "The uncorrected chi2 value is 1.439, with p=0.230\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 2 then Probes\n",
    "obs = array([[2809, 10028], [190475, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 0.280, with p=0.597\n",
      "The uncorrected chi2 value is 0.291, with p=0.589\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 4 then Probes\n",
    "obs = array([[2738, 10148], [190475, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CHI SQUARE\n",
      "The corrected chi2 value is 0.141, with p=0.707\n",
      "The uncorrected chi2 value is 0.149, with p=0.700\n"
     ]
    }
   ],
   "source": [
    "# Enter the data comparing Oyster 6 then Probes\n",
    "obs = array([[3216, 11690], [190475, 697753]])\n",
    "\n",
    "# Calculate the chi-square test\n",
    "chi2_corrected = stats.chi2_contingency(obs, correction=True)\n",
    "chi2_uncorrected = stats.chi2_contingency(obs, correction=False)\n",
    "\n",
    "# Print the result\n",
    "print('CHI SQUARE')\n",
    "print('The corrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_corrected[0], chi2_corrected[1]))\n",
    "print('The uncorrected chi2 value is {0:5.3f}, with p={1:5.3f}'.format(chi2_uncorrected[0], chi2_uncorrected[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}