Group 1 02-17-14 In Class.ipynb

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import collections\n",
      "import numpy as np\n",
      "import matplotlib.pyplot as plt\n",
      "import re\n",
      "import scipy.stats"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "STOP_SEQS = ('TAA', 'TAG', 'TGA')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Estimate the probabilities of each character from the dataset.\n",
      "data = open('data/chrIV.txt').read()\n",
      "counts = collections.Counter(data)\n",
      "total = sum(counts.itervalues())\n",
      "probs = dict([(l, count / float(total)) for l, count in counts.items()])\n",
      "print probs"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "{'A': 0.31121530771907124, 'C': 0.1888705315441341, 'T': 0.30972046427617916, 'G': 0.19019369646061546}\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Compute the probability that 3 random characters are a stop sequence.\n",
      "\n",
      "def prob_seq(seq, probs):\n",
      "    \"\"\"Probability of a sequence of characters like 'TAA' assuming independence.\"\"\"\n",
      "    return np.product([probs[c] for c in seq])\n",
      "\n",
      "prob_stop_seq = sum(prob_seq(stop_seq, probs) for stop_seq in STOP_SEQS)\n",
      "print 'Estimated probability of 3 random characters being a stop sequence:', prob_stop_seq"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Estimated probability of 3 random characters being a stop sequence: 0.0666634111351\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def split(string, offset):\n",
      "    \"\"\"Split a string into triples starting at the given offset.\"\"\"\n",
      "    return [string[start:(start+3)] for start in range(offset, len(string), 3)]\n",
      "\n",
      "test_data = \"test data\"\n",
      "for i in xrange(3):\n",
      "    print split(test_data, i)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "['tes', 't d', 'ata']\n",
        "['est', ' da', 'ta']\n",
        "['st ', 'dat', 'a']\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Get all possible ORFs.\n",
      "\n",
      "def get_possible_orfs(data, offset):\n",
      "    \"\"\"Return a list of all possible ORFs (i.e., all strings between stop sequences) after\n",
      "    splitting the data into triples starting at the given offset.\"\"\"\n",
      "    possible_orfs = []\n",
      "    orf_triples = []\n",
      "    for triple in split(data, offset):\n",
      "        if triple in STOP_SEQS:\n",
      "            possible_orfs.append(''.join(orf_triples))\n",
      "            orf_triples = []\n",
      "        else:\n",
      "            orf_triples.append(triple)\n",
      "    possible_orfs.append(''.join(orf_triples))\n",
      "    return possible_orfs\n",
      "\n",
      "all_possible_orfs = []\n",
      "for offset in xrange(3):\n",
      "    possible_orfs = get_possible_orfs(data, offset)\n",
      "    all_possible_orfs.extend(possible_orfs)\n",
      "    \n",
      "print len(all_possible_orfs), 'possible ORFs. First 10:'\n",
      "for i in xrange(10):\n",
      "    print all_possible_orfs[i]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "85727 possible ORFs. First 10:\n",
        "ACACCACACCCACACCACACCCACACACACCACACCCACACACCACACCCACACCCACACACCCACACCCACACACCACACACCACACCACACCACACCCACACCCACACCACACCCACACCCACACACCACACCCACACCCACACACCACACACTACCCCTAACACTACCCTATTC\n",
        "CCC\n",
        "TTTTACCTGTCTCCAAACCTACCCTCACATTACCCTACCTCCCCACTCGTTACCCTGCCCCACTCAACCATCCACTCCCAACCACCATCCATCTCTCTACTTACCACTAACCACCGTCCACCA\n",
        "CCGTTACCCTCCAATTACCCATATCCAACTCCACTACCACTTACCCTACTATTACCCTACCATCCACCATGTCCTACTCACTGTACTGTTGTTCTACCCTCCATATTGAAACGT\n",
        "CAAATGATCGTAAATAATACATACATACTTACCCTACCACTCTATACCACCACTACCACCACCGCCACTTGCCACACTCACCTTCACTTCTAC\n",
        "TATGTCATACGCACACGGATGCTACAGTATATACCATCTCAAACTTACCCTACTTTCATATTCCACACCATGGCCCCATTCTCACTAAATCAGTACCAAATGCACTCACATCATTATTCACGGCACTTGCCTCAGCGGTCTATACCCTGTGCCATTTACGCA\n",
        "CTCCCACGATTATCCACATTTTATTACCTATATCTCATTCGGCGGCCCCAAATATTGTATAACTGCTCT\n",
        "TACATACGTTATACCACTTTTACACCGTATACTAACCACTCAATTTATATACACTTATGCCAATATTACAAAAAAATCACCACTAAAATCACC\n",
        "ACA\n",
        "AAATATTCTATTCTTCAACCA\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Get the CDF for a geometric distribution based on the probability of seeing a stop sequence.\n",
      "cdf = scipy.stats.geom(prob_stop_seq).cdf"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Compute the p-values of each possible ORF.\n",
      "probs = 1 - np.array([cdf(len(possible_orf)) for possible_orf in all_possible_orfs])\n",
      "print probs[:10]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[  4.97482250e-06   8.13045545e-01   2.06400075e-04   3.84030066e-04\n",
        "   1.63516321e-03   1.40024238e-05   8.56331878e-03   1.63516321e-03\n",
        "   8.13045545e-01   2.34857331e-01]\n"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Plot a histogram of the p-values.\n",
      "plt.hist(probs, bins=100)\n",
      "plt.title('Stuff and things based on numbers')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 9,
       "text": [
        "<matplotlib.text.Text at 0xbee2a4c>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAuQAAAIACAYAAADKcXnxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0xaa51c2c>"
       ]
      }
     ],
     "prompt_number": 9
    }
   ],
   "metadata": {}
  }
 ]
}