Write_An_Assembler.ipynb

{
 "metadata": {
  "name": "Write_An_Assembler"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Write A Genome Assembler with ``blasr`` and ``(I)Python``\n",
      "=========================================================="
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Introduction\n",
      "============\n",
      "\n",
      "I am not sure it is right thing to do some coding in a vacation. (See \n",
      "[this tweet](https://twitter.com/pm/status/284102222416056320), not sure if I would agree.)\n",
      "Anyway, before the vacation, I decided to organize whole bunch of random papers\n",
      "I collected in the last few years in my laptop. I eventually felt that I should\n",
      "read some of some theoretical papers about genome assembly again. I grabbed \n",
      "[Gene Meyer's paper](http://bioinformatics.oxfordjournals.org/content/21/suppl_2/ii79.abstract),\n",
      "and started to think about the problem about constructing unitigs\n",
      "(high-confident contigs). Before I tried to read the paper in detail, I just\n",
      "felt maybe that it was useful to write some quick code to check out what real\n",
      "data looked like. I started writing some code visualizing the local overlapping\n",
      "and generating the global overlapping graph. It was pretty straight forward and quite\n",
      "inspiring from the visualization. The visualization motived me to write\n",
      "more ad-hoc code in IPython to go beyond generating simple visualization during the\n",
      "vacation. I started to implement a very simple greedy algorithm to connect the\n",
      "input DNA fragments.  Eventually, I found I can atucally get the whole genome \n",
      "assembly right!! This Notebook shows the work step by step toward a simple genome \n",
      "assembler for long read data using ``IPython`` and ``Python``.\n",
      "\n",
      "The input data was the E. coli K12 data from PacBio(R) \n",
      "[DevNet Site](https://github.com/PacificBiosciences/DevNet/wiki/Hierarchical-Genome-Assembly-Process-%28HGAP%29).\n",
      "I generated a new set of pre-assembled reads as the input DNA fragment using\n",
      "[HBAR-DTK](https://github.com/PacificBiosciences/HBAR-DTK). The statistics of\n",
      "the input reads is shown below:\n",
      "\n",
      "    Pre-assembled Read Stats\n",
      "    #Seqs   17497\n",
      "    Min     448\n",
      "    1st Qu. 5365\n",
      "    Median  6149\n",
      "    Mean    6193\n",
      "    3rd Qu. 7189\n",
      "    Max     14762\n",
      "    Total   108363627\n",
      "    n50     6521\n",
      "    n90     4694\n",
      "    n95     3864\n",
      "\n",
      "This is about 23x of average length of 6.2kb reads of E. coli K12 genome. The\n",
      "mean sequence identity when aligning the pre-assembled reads to the canonical\n",
      "E. coli K12 reference is 99.787% (median identity = 99.921%). The pre-assembled\n",
      "read data is available for download from \n",
      "[this link to pre_assembled_read.fa](http://files.figshare.com/1004577/pre_assembled_reads.fa). According the\n",
      "Lander-Waterman statistics, this set of reads is enough to get single contig\n",
      "for the single chromosome. The only thing one needs to worry about is that\n",
      "whether the repeats in genome can be uniquely. Since it is shown one can use\n",
      "Celera Assembler to assemble this data set into one single contig, this implies\n",
      "that all of the repeats of the rRNA operon can be resolved from this data set.\n",
      "Typically, one might need multiple libraries, especially those long jumping\n",
      "libraries, to resolve long repeats from short read (~100bp) data. For example,\n",
      "ALLPATHS-LG can use multiple short-read libraries and long read data to get\n",
      "perfect assemblies for bacteria\n",
      "[Paper](http://genome.cshlp.org/content/early/2012/07/24/gr.141515.112.abstract) \n",
      "[ALLPATHS-LG Blog](http://www.broadinstitute.org/software/allpaths-lg/blog/?page_id=14). However,\n",
      "when one can get long range information from one single library, the assembly\n",
      "process can be probably greatly simplified.  As demonstrated here, one can \n",
      "actually get pretty good assembly using Python code with blasr for overlapping and\n",
      "Quiver for polishing the residual errors.\n",
      "\n",
      "For long read data, the de Bruijn graph with some sophisticated graph\n",
      "operations to get around the limit of short read data to assemble genomes may\n",
      "not be the right approach. The Overlap-Layout-Consensus can be much more\n",
      "efficient for assembling long reads. Also, when the reads are getting longer,\n",
      "the overlapping between the reads becomes more and more specific. This will\n",
      "significantly reduce the complexity of the overlapping graph between the reads.\n",
      "(Here we define an overlapping graph as a graph where the nodes are reads and\n",
      "the edges are constructed from the pairs of proper overlapped reads.) When the\n",
      "overlapping graph is simple, it becomes possible to use some naive algorithm to\n",
      "get a reasonably good assembly. To make this clear, one can image if the read\n",
      "lengths are as long as the genome, the number of overlapping one needs to worry\n",
      "about will be in single digit and there will be no ambiguity at all. The\n",
      "overlapping and layout will be trivial in such extreme case.\n",
      "\n",
      "When I use [Celera Assembler (CA)](http://sourceforge.net/projects/wgs-assembler/)\n",
      "to assemble genomes, I always try to peek the\n",
      "intermediate data in order to understand the process better and maybe fix few\n",
      "things manually. [Michael Schatz](http://schatzlab.cshl.edu/) ([@mike_schatz](https://twitter.com/mike_schatz)) \n",
      "and [Adam Phillippy](https://twitter.com/aphillippy) both pointed to me\n",
      "some very useful information about CA.  (Thanks, Michael and Adam.)\n",
      "``tigStore`` is a very useful command to see the relationship between the\n",
      "generated unitigs and the initial DNA fragments.  The file\n",
      "``4-unitigger/best.edges`` contains the information best overlapped reads which\n",
      "one can use to construct the overlapping graph and one can learn a lot subtle\n",
      "points about genome assembly by visualizing those.  This notebook is definitely\n",
      "inspired by the great work done by the CA's developers.\n",
      "\n",
      "This IPython notebook is just a \"Notebook\". I write these code for fun and for\n",
      "(self-)educational purposes. It shows that given the good long read data and\n",
      "the proper tools [``blasr``](https://github.com/PacificBiosciences/blasr),\n",
      "[``pbdagcon``](https://github.com/PacificBiosciences/pbdagcon), \n",
      "[``HBAR-DTK``](https://github.com/PacificBiosciences/HBAR-DTK), and \n",
      "[``quiver``](https://github.com/PacificBiosciences/GenomicConsensus) one can learn how to\n",
      "do genome assembly with a \"not-so-high-performance\" language [``Python``](http://python.org).  The great \n",
      "[``IPython Notebook``](http://ipython.org/) allows me to try out different\n",
      "code snippet and different algorithm ideas without a lot of overhead. One\n",
      "should notice this code in this ``IPython Notebook`` is mainly for fun, not for\n",
      "\"production\". I chose the least resistant path for coding.  Coding style and\n",
      "code performance were considered but not taking seriously as I was doing a lot\n",
      "of experiments for different algorithm approaches when I coded this notebook.\n",
      "\n",
      "I consider this Notebook is as a personal project for fun. The data and the\n",
      "used software can be all downloaded publicly.  Although I am an employee of\n",
      "PacBio(R), this notebook is developed in my own time. PacBio will not be \n",
      "responsible for any information associated with this Notebook.  \n",
      "\n",
      "--[Jason Chin](https://twitter.com/infoecho), Apr 2, 2013\n",
      "\n"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Get Data\n",
      "--------\n",
      "I put the pre-assembled read file in [Figshare](http://figshare.com). Here is [the link to the pre_assembled_read.fa](http://files.figshare.com/1004577/pre_assembled_reads.fa).\n",
      "For most of part of the Notebook, one does not need the fasta file. You can also visit [the project data page](http://figshare.com/articles/Write_a_Genome_Assembler_in_Python_IPython/666737) to see other data files and this code."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Overlapping\n",
      "-----------\n",
      "The first step to assemble the genome is to use ``blasr`` to find the overlapping between the reads. This can be done by executing this commend:\n",
      "\n",
      "    blasr pre_assembled_reads.fa pre_assembled_reads.fa -m 4 -nproc 32 -bestn 32 -maxLCPLength 15 -nCandidates 36 -out pr_pr_bn32.m4 -noSplitSubreads\n",
      "\n",
      "We ask ``blasr`` to output the coordinates of the alignment between the reads by specified the ``-m 4`` option.\n",
      "The ``-nproc 32`` will let ``blasr`` to use 32 threads.  For each reads, we allow ``blasr`` to find the other 32 best overlaps by using ``-bestn 32`` and scaning 36 potential hit candidateswith ``-nCandidates 36``. The ``-noSplitSubreads`` has only cosmatic effect here. It makes the ouput query identifier cleaner. \n",
      "\n",
      "``blasr`` is a general purpose aligner. It is not really tuned to do overlapping work. One will see sometimes it does not handle the ends of the alignment as one would like it to do. The ``containment_tolerance`` variables used below was introduced for being less picky at the ends of the alignment. However, one would need to worry about potential mis-assemblies if the tolerance is too big.\n",
      "\n",
      "The aligned results can be found in the FigShare project page mentioned above. Or, one can download it from [this link]( http://files.figshare.com/1004576/pr_pr_bn32.m4)."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "\"``blasr -m 4``\" output format\n",
      "------------------------------\n",
      "The cells below show the content of the ``m4``  file. It containes the alignment information between the reads. The fields are ``query identifier``, ``target identifier``, ``alignment score (more negative -> better alignment)``, ``alignment identity``, ``query strand``, ``query start``, ``query end``, ``query length``, ``target strand``, ``target start``, ``target end``, ``target length``.   The rest of the fields are not used.  One trap here is that the \"``target start``\" and \"``target end``\" are not relative to the original sequence if \"``target strand``\"  is ``1`` .  When \"``target strand= 1``\", it means that the reverse-complimentary strand is aligned and the \"``target start``\" and \"``target end``\" are the coordinates in the reverse complimentary strand.  In the stand ``blasr`` output, \"``query strand``\" is always ``0``.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "!head -10 pr_pr_bn32.m4"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "0xc4b462ba_0096978 0xc4b462ba_0096978 -14495 100 0 0 2899 2899 0 0 2899 2899 50 60847 -2716.48 0 21\r\n",
        "0xc4b462ba_0096978 0x79c77458_0006261 -14400 99.4505 0 0 2899 2899 1 2502 5411 10399 50 60890 -2683.04 0 21\r\n",
        "0xc4b462ba_0096978 0x72a8c4f1_0018952 -14395 99.4164 0 0 2899 2899 0 483 3393 5755 50 60893 -2658.33 0 21\r\n",
        "0xc4b462ba_0096978 0x4dd7c9d5_0133333 -14390 99.4162 0 0 2899 2899 1 2718 5626 6023 50 60894 -2683.04 0 21\r\n",
        "0xc4b462ba_0096978 0x726e8550_0094067 -14385 99.3821 0 0 2899 2899 0 27 2936 7163 50 60897 -2658.76 0 21\r\n",
        "0xc4b462ba_0096978 0x833174a6_0078354 -14385 99.3821 0 0 2899 2899 1 2676 5585 6916 50 60897 -2683.28 0 21\r\n",
        "0xc4b462ba_0096978 0xc52dd432_0103092 -14380 99.3819 0 0 2899 2899 1 2565 5472 6025 50 60900 -2666.7 0 21\r\n",
        "0xc4b462ba_0096978 0x48e2fc97_0054635 -14375 99.3477 0 0 2899 2899 0 2621 5529 7539 50 60902 -2658.24 0 21\r\n",
        "0xc4b462ba_0096978 0xea0a0bab_0128842 -14375 99.3477 0 0 2899 2899 0 7684 10592 11923 50 60901 -2657.5 0 21\r\n",
        "0xc4b462ba_0096978 0x4949e7c9_0060870 -14360 99.3132 0 0 2899 2899 1 2927 5832 6621 50 60907 -2666.12 0 21\r\n"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Convert ``blasr``'s output to python internal data\n",
      "--------------------------------------------------\n",
      "\n",
      "We deinfe ``read_node`` class for the objects that holds the local overlapping information of each read. It simply holds a ``python`` dictionary (hash_map) where the key is the ``target identifier`` and the elements from ``m4`` line are converted to the right types and stored. Only \"non-contaminated\"\" alignments are stored. If one read is full aligned within the other read, such alignment is not stored in a ``read_node``. The ``get_overlap_data`` function processes the ``m4`` file and determine whether the alignments are ``contaminated`` with some tolerence of the ends of the alignments. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import sys\n",
      "import os\n",
      "class read_node_0:  #experimenting on keeping the data in a list rather than a dictionary, not used\n",
      "\n",
      "    def __init__(self, name, length):\n",
      "        self.name = name\n",
      "        self.length = length\n",
      "        self.hits = [] \n",
      "        self.hit_map = None\n",
      "        \n",
      "    def add_hit(self, aln_info):\n",
      "        self.hits.append ( aln_info )\n",
      "\n",
      "    def __getitem__(self, target_name):\n",
      "        if self.hit_map == None:\n",
      "            self.hit_map = dict(zip( [h[0] for h in self.hits], self.hits ) )\n",
      "        return self.hit_map[target_name]\n",
      "\n",
      "class read_node:\n",
      "\n",
      "    def __init__(self, name, length):\n",
      "        self.name = name\n",
      "        self.length = length\n",
      "        self.hit_map = {}\n",
      "        \n",
      "    def add_hit(self, aln_info):\n",
      "        self.hit_map[aln_info[0]] = aln_info\n",
      "\n",
      "    def __getitem__(self, target_name):\n",
      "        return self.hit_map[target_name]\n",
      "    \n",
      "    @property\n",
      "    def hits(self):  #another convenient representation of the same data\n",
      "        return self.hit_map.values()\n",
      "\n",
      "\n",
      "def get_overlap_data(m4_filename):\n",
      "    containment_tolerance = 50 \n",
      "    permitted_error_pct = 2\n",
      "    overlap_data = {}\n",
      "    contained_reads = set()\n",
      "    with open(m4_filename) as m4_f:\n",
      "        for l in m4_f:\n",
      "            l = l.strip().split()\n",
      "            q_name, t_name =l[0:2]\n",
      "            if q_name == t_name:\n",
      "                continue\n",
      "            aln_score = int(l[2])\n",
      "            aln_idt = float(l[3])\n",
      "            \n",
      "            if aln_idt < 100-permitted_error_pct:\n",
      "                continue\n",
      "    \n",
      "            q_strand, q_start, q_end, q_len = ( int(x) for x in l[4:8])\n",
      "            t_strand, t_start, t_end, t_len = ( int(x) for x in l[8:12])\n",
      "    \n",
      "            if q_len - (q_end - q_start) < containment_tolerance:\n",
      "                contained_reads.add(q_name)\n",
      "    \n",
      "            if t_len - (t_end - t_start) < containment_tolerance:\n",
      "                contained_reads.add(t_name)\n",
      "    \n",
      "            if q_name not in overlap_data:\n",
      "                overlap_data[ q_name ] = read_node(q_name, q_len)\n",
      "    \n",
      "            assert q_strand == 0\n",
      "            if t_name not in [x[0] for x in overlap_data[ q_name ].hits]:\n",
      "                overlap_data[ q_name ].add_hit( (t_name, \n",
      "                                                 aln_score, \n",
      "                                                 aln_idt, \n",
      "                                                 (q_strand, q_start, q_end, q_len),\n",
      "                                                 (t_strand, t_start, t_end, t_len) ) )\n",
      "    \n",
      "            #symmetrized the alignment record\n",
      "            if t_name not in overlap_data:\n",
      "                overlap_data[ t_name ] = read_node(t_name, t_len)\n",
      "    \n",
      "            if q_name not in [x[0] for x in overlap_data[ t_name ].hits]:\n",
      "                if t_strand == 1: \n",
      "                    t_start, t_end = t_len - t_end, t_len - t_start\n",
      "                    q_start, q_end = q_len - q_end, q_len - q_start\n",
      "                    overlap_data[ t_name ].add_hit( (q_name, \n",
      "                                                 aln_score, \n",
      "                                                 aln_idt, \n",
      "                                                 (0, t_start, t_end, t_len),\n",
      "                                                 (1, q_start, q_end, q_len) ) )\n",
      "                else:\n",
      "                    overlap_data[ t_name ].add_hit( (q_name, \n",
      "                                                 aln_score, \n",
      "                                                 aln_idt, \n",
      "                                                 (0, t_start, t_end, t_len),\n",
      "                                                 (0, q_start, q_end, q_len) ) )\n",
      "    return overlap_data, contained_reads"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "overlap_data, contained_reads = get_overlap_data(\"pr_pr_bn32.m4\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A simple function to convert the overlapping data to a GML file\n",
      "---------------------------------------------------------------\n",
      "\n",
      "The ``generate_overlap_gml`` takes the ``overlap_data`` and the ``contained_reads`` to generate a overlap graph in ``GML`` format. (You will need to install ``networkx`` in your ``Python`` environment to use this code.)  \n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def generate_overlap_gml(overlap_data, contained_reads, gml_filename):\n",
      "    containment_tolerance = 50 \n",
      "    permitted_error_pct = 2\n",
      "    import networkx as nx\n",
      "    G=nx.DiGraph()\n",
      "    node_in_graph = set()\n",
      "    for q_name in overlap_data:\n",
      "        if q_name in contained_reads:\n",
      "            continue\n",
      "        if q_name not in node_in_graph:\n",
      "            G.add_node(q_name)\n",
      "        targets = overlap_data[ q_name ].hits\n",
      "        targets_3prime = [ h for h in targets if h[4][1] < containment_tolerance and h[0] not in contained_reads]\n",
      "        targets_5prime = [ h for h in targets if h[3][1] < containment_tolerance and h[0] not in contained_reads]\n",
      "        targets_3prime.sort(key = lambda k:k[1])\n",
      "        targets_5prime.sort(key = lambda k:k[1])\n",
      "\n",
      "        if len(targets_3prime) > 0:\n",
      "            t = targets_3prime[0]\n",
      "            t_name = t[0]\n",
      "            if t_name not in node_in_graph:\n",
      "                G.add_node(t_name)\n",
      "            G.add_edge(q_name, t_name)\n",
      "\n",
      "        if len(targets_5prime) > 0:\n",
      "            t = targets_5prime[0]\n",
      "            t_name = t[0]\n",
      "            if t_name not in node_in_graph:\n",
      "                G.add_node(t_name)\n",
      "            G.add_edge(q_name, t_name)\n",
      "    nx.write_gml(G, gml_filename)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "generate_overlap_gml(overlap_data, contained_reads, \"overlap_graph.gml\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Overlapping graph visualized by Gephi\n",
      "-------------------------------------\n",
      "\n",
      "Ther generated overlap_graph.gml can be visualized by [Gephi](https://gephi.org/). One can \n",
      "use \"YifanHu's Multilevel\" graph layout algorithm to do the rough\n",
      "layout followed by \"ForceAtlas 2\" algorithm to smooth the graph. The cell below shows the\n",
      "global overlapping structure. One can see that it is topologically a circle plus\n",
      "a number of singletons. Namely, the read data does capture the whole genome in\n",
      "one single circular layout."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import Image\n",
      "Image(filename = \"overlap_graph01.png\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAABAAAAAQACAYAAAB/HSuDAACAAElEQVR42uyd93Md15Vu3z85VTOe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\ngIXtX1aYJbPnBt9hAAAAAElFTkSuQmCC\n",
       "prompt_number": 6,
       "text": [
        "<IPython.core.display.Image at 0x11c389e10>"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "We can also see some local \"bubbles\" in the graph."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "Image(filename = \"overlap_graph02.png\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAOgAAADuCAYAAAAz8EtlAAAEJGlDQ1BJQ0MgUHJvZmlsZQAAOBGF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\nJNCQiWJEwA0BCdSNK2SICIQE/gfcDv5dap6FqgAAAABJRU5ErkJggg==\n",
       "prompt_number": 7,
       "text": [
        "<IPython.core.display.Image at 0x10fc49510>"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "Image(filename = \"overlap_graph03.png\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAZ0AAADqCAYAAAB5j5hvAAAEJGlDQ1BJQ0MgUHJvZmlsZQAAOBGF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\nzyYAAAAASUVORK5CYII=\n",
       "prompt_number": 8,
       "text": [
        "<IPython.core.display.Image at 0x11c401550>"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Some utility constants, functions and a class for viewing the local overlaping with ``SVG`` files\n",
      "--------------------------------------------------------------------------------------------------\n",
      "\n",
      "The cell below defines some contants and utility funtions so we can generate ``SVG`` to show the local overlapping alignments.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "arrow_defs = \"\"\"<defs>\n",
      "    <marker id=\"Triangle_green\" stroke=\"green\" fill=\"green\"\n",
      "      viewBox=\"0 0 10 10\" refX=\"0\" refY=\"5\"\n",
      "      markerUnits=\"strokeWidth\"\n",
      "      markerWidth=\"4\" markerHeight=\"3\"\n",
      "      orient=\"auto\">\n",
      "      <path d=\"M 0 0 L 10 5 L 0 10 z\" />\n",
      "    </marker>\n",
      "    <marker id=\"Triangle_blue\" stroke=\"blue\" fill=\"blue\"\n",
      "      viewBox=\"0 0 10 10\" refX=\"0\" refY=\"5\"\n",
      "      markerUnits=\"strokeWidth\"\n",
      "      markerWidth=\"4\" markerHeight=\"3\"\n",
      "      orient=\"auto\">\n",
      "      <path d=\"M 0 0 L 10 5 L 0 10 z\" />\n",
      "    </marker>\n",
      "    <marker id=\"Triangle_black\" stroke=\"black\" fill=\"black\"\n",
      "      viewBox=\"0 0 10 10\" refX=\"0\" refY=\"5\"\n",
      "      markerUnits=\"strokeWidth\"\n",
      "      markerWidth=\"4\" markerHeight=\"3\"\n",
      "      orient=\"auto\">\n",
      "      <path d=\"M 0 0 L 10 5 L 0 10 z\" />\n",
      "    </marker>\n",
      "    <marker id=\"Triangle_red\" stroke=\"red\" fill=\"red\"\n",
      "      viewBox=\"0 0 10 10\" refX=\"0\" refY=\"5\"\n",
      "      markerUnits=\"strokeWidth\"\n",
      "      markerWidth=\"4\" markerHeight=\"3\"\n",
      "      orient=\"auto\">\n",
      "      <path d=\"M 0 0 L 10 5 L 0 10 z\" />\n",
      "    </marker>\n",
      "  </defs>\"\"\"\n",
      "\n",
      "def svg_arrow( x1, y1, x2, y2, col, w):\n",
      "    return \"\"\"<g stroke = \"%s\" stroke-width = \"%f\" fill = \"none\">\\\n",
      "<path d = \"M %f %f L %f %f\" marker-end=\"url(#Triangle_%s)\"/> </g>\"\"\" % (col, w, x1, y1, x2, y2, col)\n",
      "\n",
      "def svg_line( x1, y1, x2, y2, col, w):\n",
      "    return \"\"\"<g stroke = \"%s\" stroke-width = \"%f\" fill = \"none\">\\\n",
      "<path d = \"M %f %f L %f %f\"/> </g>\"\"\" % (col, w, x1, y1, x2, y2)\n",
      "\n",
      "class Overlap_SVG_view:\n",
      "    def __init__(self):\n",
      "        self.header = \"\"\"<?xml version=\"1.0\" standalone=\"no\"?>\n",
      "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\" \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
      "<svg xmlns:svg=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\" width=\"900px\" height=\"300px\">\n",
      "\"\"\"\n",
      "        self.footer = \"\"\"</svg>\"\"\"\n",
      "        self.elements = []\n",
      "        self._svg_data = None\n",
      "        \n",
      "    def add_element(self, elm):\n",
      "        self.elements.append(elm)\n",
      "        \n",
      "    def _construct_SVG(self):\n",
      "        SVG_str = []\n",
      "        SVG_str.append( self.header )\n",
      "        for elm in self.elements:\n",
      "            SVG_str.append( elm )\n",
      "        SVG_str.append( self.footer )\n",
      "        self._svg_data = \"\\n\".join(SVG_str)\n",
      "        self._svg_data = self._svg_data.decode('utf-8')\n",
      "\n",
      "    def output(self, filename):\n",
      "        if self._svg_data is None:\n",
      "            self._construct_SVG()\n",
      "        with open(filename, \"w\") as out_f:\n",
      "            print >> out_f, self._svg_data\n",
      "            \n",
      "    def _repr_svg_(self):\n",
      "        if self._svg_data is None:\n",
      "            self._construct_SVG()\n",
      "        return self._svg_data\n",
      "        \n",
      "    @property\n",
      "    def svg(self):\n",
      "        return SVG(self._repr_svg_())\n",
      "            "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "``generate_overlap_view``: a function taking the overlapping data to generate a ``SVG`` file\n",
      "---------------------------------------------------------------------------------------------\n",
      "\n",
      "The ``generate_overlap_view`` simple does several coordinate transformations to create a ``SVG`` file so we can see the local overlap around a given DNA fragment. The blank arrow indicates the fragment in interesting. Green and blue arrows indicate the forward and reversed proper overlapped alignments. The red arrows indicate containment alignments or improper alignments (locally inconsistent alignments due to repeats, mis-mapping or sequencing artifacts). "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.display import SVG\n",
      "\n",
      "import os\n",
      "\n",
      "def generate_overlap_view(overlap_data, q_name):\n",
      "    containment_tolerance = 10\n",
      "    \n",
      "    hits = overlap_data[q_name].hits\n",
      "    SV = Overlap_SVG_view()\n",
      "    SV.add_element(arrow_defs)\n",
      "    x_offset = 10000\n",
      "    x_scaling = 0.02\n",
      "    x1 = 0\n",
      "    x2 = overlap_data[q_name].length\n",
      "    x1 += x_offset \n",
      "    x2 += x_offset\n",
      "    x1 *= x_scaling\n",
      "    x2 *= x_scaling\n",
      "    y = 200\n",
      "    SV.add_element(svg_arrow(x1, y, x2, y, \"black\", 4))\n",
      "    hits.sort(key = lambda k: (k[3][1],k[3][3],k[3][2]))\n",
      "    for h in hits:\n",
      "        t_name, t_score, t_idt = h[0], h[1], h[2]\n",
      "        if t_name in contained_reads:\n",
      "            continue\n",
      "        q_strand, q_start, q_end, q_len = h[3]\n",
      "        t_strand, t_start, t_end, t_len = h[4]\n",
      "        y -=10\n",
      "        x1 = q_start\n",
      "        x2 = q_end\n",
      "        col = None\n",
      "        \n",
      "        if q_start < containment_tolerance and t_len - t_end < containment_tolerance:\n",
      "            x1 = -t_start\n",
      "        elif q_len - q_end < containment_tolerance and t_start < containment_tolerance:\n",
      "            x2 = q_end + (t_len - t_end)\n",
      "        else:\n",
      "            col = \"red\"\n",
      "            if t_strand == 1:\n",
      "                x1, x2 = x2, x1\n",
      "            print h\n",
      "        if t_strand == 0 and col == None:\n",
      "            col = \"green\"\n",
      "        elif col == None:\n",
      "            col = \"blue\"\n",
      "            x1, x2 = x2, x1\n",
      "        #print h\n",
      "        #print min(x1, x2), max(x1, x2), abs(x1-x2)\n",
      "        x1 += x_offset\n",
      "        x2 += x_offset\n",
      "        x1 *= x_scaling\n",
      "        x2 *= x_scaling\n",
      "        SV.add_element(svg_arrow(x1, y, x2, y, col, 3))\n",
      "    return SV\n",
      "    \n",
      "q_name = overlap_data.keys()[0]\n",
      "OSV = generate_overlap_view(overlap_data, q_name)\n",
      "display(OSV.svg)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "('0x9ba164ba_0123315', -32790, 99.8631, (0, 0, 6571, 6571), (0, 38, 6610, 7974))\n"
       ]
      },
      {
       "output_type": "display_data",
       "svg": [
        "<svg height=\"300px\" version=\"1.1\" width=\"900px\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:svg=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
        "\n",
        "<defs>\n",
        "    <marker fill=\"green\" id=\"Triangle_green\" markerHeight=\"3\" markerUnits=\"strokeWidth\" markerWidth=\"4\" orient=\"auto\" refX=\"0\" refY=\"5\" stroke=\"green\" viewBox=\"0 0 10 10\">\n",
        "      <path d=\"M 0 0 L 10 5 L 0 10 z\"/>\n",
        "    </marker>\n",
        "    <marker fill=\"blue\" id=\"Triangle_blue\" markerHeight=\"3\" markerUnits=\"strokeWidth\" markerWidth=\"4\" orient=\"auto\" refX=\"0\" refY=\"5\" stroke=\"blue\" viewBox=\"0 0 10 10\">\n",
        "      <path d=\"M 0 0 L 10 5 L 0 10 z\"/>\n",
        "    </marker>\n",
        "    <marker fill=\"black\" id=\"Triangle_black\" markerHeight=\"3\" markerUnits=\"strokeWidth\" markerWidth=\"4\" orient=\"auto\" refX=\"0\" refY=\"5\" stroke=\"black\" viewBox=\"0 0 10 10\">\n",
        "      <path d=\"M 0 0 L 10 5 L 0 10 z\"/>\n",
        "    </marker>\n",
        "    <marker fill=\"red\" id=\"Triangle_red\" markerHeight=\"3\" markerUnits=\"strokeWidth\" markerWidth=\"4\" orient=\"auto\" refX=\"0\" refY=\"5\" stroke=\"red\" viewBox=\"0 0 10 10\">\n",
        "      <path d=\"M 0 0 L 10 5 L 0 10 z\"/>\n",
        "    </marker>\n",
        "  </defs>\n",
        "<g fill=\"none\" stroke=\"black\" stroke-width=\"4.000000\"><path d=\"M 200.000000 200.000000 L 331.420000 200.000000\" marker-end=\"url(#Triangle_black)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"green\" stroke-width=\"3.000000\"><path d=\"M 43.260000 190.000000 L 268.600000 190.000000\" marker-end=\"url(#Triangle_green)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"blue\" stroke-width=\"3.000000\"><path d=\"M 271.320000 180.000000 L 104.440000 180.000000\" marker-end=\"url(#Triangle_blue)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"blue\" stroke-width=\"3.000000\"><path d=\"M 285.540000 170.000000 L 136.260000 170.000000\" marker-end=\"url(#Triangle_blue)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"green\" stroke-width=\"3.000000\"><path d=\"M 159.180000 160.000000 L 296.400000 160.000000\" marker-end=\"url(#Triangle_green)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"blue\" stroke-width=\"3.000000\"><path d=\"M 330.340000 150.000000 L 170.080000 150.000000\" marker-end=\"url(#Triangle_blue)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"red\" stroke-width=\"3.000000\"><path d=\"M 200.000000 140.000000 L 331.420000 140.000000\" marker-end=\"url(#Triangle_red)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"blue\" stroke-width=\"3.000000\"><path d=\"M 399.180000 130.000000 L 224.280000 130.000000\" marker-end=\"url(#Triangle_blue)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"green\" stroke-width=\"3.000000\"><path d=\"M 284.840000 120.000000 L 463.480000 120.000000\" marker-end=\"url(#Triangle_green)\"/> </g>\n",
        "<g fill=\"none\" stroke=\"green\" stroke-width=\"3.000000\"><path d=\"M 309.160000 110.000000 L 475.560000 110.000000\" marker-end=\"url(#Triangle_green)\"/> </g>\n",
        "</svg>"
       ],
       "text": [
        "<IPython.core.display.SVG at 0x11c40e310>"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Generate the \"best overlap graph\"\n",
      "---------------------------------\n",
      "\n",
      "Given that we see a beautiful circle overlapping graph, we like to see if can choose a seeding frgament and extend the overlapping to form a contigs.\n",
      "The code below scans through the ``overlap_data`` and assign the best score alignments for both 3' and 5' of a read as the best partner. The return of the ``get_best_overlap_graph`` is a dictionary where the key is a fragment identifier and the value is the alignment information of the 3' and 5' best overlapping alignments."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def filter_overlap_hit(overlap_hit):\n",
      "    containment_tolerance = 50\n",
      "    q_strand, q_start, q_end, q_len = overlap_hit[3]\n",
      "    t_strand, t_start, t_end, t_len = overlap_hit[4]\n",
      "    if q_len - (q_end - q_start) < containment_tolerance and  t_len - (t_end - t_start) / t_len < containment_tolerance:\n",
      "        return False\n",
      "    elif overlap_hit[0] in contained_reads:\n",
      "        return False\n",
      "    else:\n",
      "        return True\n",
      "\n",
      "def get_best_overlap_graph(overlap_data):\n",
      "    best_overlap_graph = {}\n",
      "    containment_tolerance = 50\n",
      "    \n",
      "    for q_name in overlap_data:\n",
      "        if q_name in contained_reads:\n",
      "            continue\n",
      "        \n",
      "        if q_name not in best_overlap_graph:\n",
      "            best_overlap_graph[q_name] = {\"5p\":None, \"3p\":None}\n",
      "        \n",
      "        \n",
      "        targets = overlap_data[ q_name ].hits\n",
      "        \n",
      "        targets_5prime = [ h for h in targets if h[3][1] < containment_tolerance and filter_overlap_hit(h)]\n",
      "        targets_3prime = [ h for h in targets if h[4][1] < containment_tolerance and filter_overlap_hit(h)]\n",
      "    \n",
      "        targets_5prime.sort(key = lambda k:k[1])\n",
      "        targets_3prime.sort(key = lambda k:k[1])\n",
      "    \n",
      "        if len(targets_5prime) > 0:\n",
      "            best_overlap_graph[q_name][\"5p\"] = targets_5prime[0]\n",
      "            t_name = targets_5prime[0][0]\n",
      "    \n",
      "                \n",
      "        if len(targets_3prime) > 0:\n",
      "            best_overlap_graph[q_name][\"3p\"] = targets_3prime[0]\n",
      "            t_name = targets_3prime[0][0]\n",
      "    \n",
      "    return best_overlap_graph"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 11
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Check the output of ``get_best_overlap_graph``\n",
      "----------------------------------------------\n",
      "The cell below find the best alignment from the read of the first key in the ``best_overlap_graph``. It shows the best 3' and 5' partners."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "best_overlap_graph = get_best_overlap_graph(overlap_data)\n",
      "q_name = best_overlap_graph.keys()[0]\n",
      "print q_name, \"has best partners:\", best_overlap_graph[q_name]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "0xf6bb6a22_0085668 has best partners: {'3p': ('0xdcf35477_0042197', -37729, 99.9206, (0, 1032, 8586, 8586), (1, 0, 7557, 9323)), '5p': ('0xd9132340_0068296', -40160, 99.373, (0, 0, 8128, 8586), (1, 436, 8525, 8525))}\n"
       ]
      }
     ],
     "prompt_number": 12
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Find a best overlapping path from a seeding fragment\n",
      "----------------------------------------------------\n",
      "\n",
      "The two functions defined below allows us to extend the overlapping through the chain of overlapped fragments. The extension is terminated when (1) some inconsistent overlapping is detected (2) the overlapped fragment is already used by other contigs (3) no overlapped fragment is found and (4) a circle is detected in the path. Such rules may not be the best way to generate unitigs but they are very simple to implement. At least, these rules do work well for this E. coli data set. One can image some more complicated scenarios that it will be necessary to refine these rules or add more rules to get the best results and avoid mis-assemblies. \n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def proper_overlap_hit(overlap_hit):\n",
      "    containment_tolerance = 50\n",
      "    q_strand, q_start, q_end, q_len = overlap_hit[3]\n",
      "    t_strand, t_start, t_end, t_len = overlap_hit[4]\n",
      "    if q_start < containment_tolerance:\n",
      "        if t_len - t_end > containment_tolerance:\n",
      "            return False\n",
      "        else:\n",
      "            return True\n",
      "    if t_start < containment_tolerance:\n",
      "        if q_len - q_end > containment_tolerance:\n",
      "            return False\n",
      "        else:\n",
      "            return True\n",
      "        \n",
      "def find_path( q_name_end, frag_used = set() ):\n",
      "    reverse_end = {\"3p\":\"5p\", \"5p\":\"3p\"}\n",
      "    path = []\n",
      "    path_q_name = set()\n",
      "    q_name, end = q_name_end \n",
      "    if end == \"5p\":\n",
      "        reversing_path = True\n",
      "    else:\n",
      "        reversing_path = False\n",
      "        \n",
      "    while 1:\n",
      "        if q_name in frag_used:\n",
      "            if reversing_path:\n",
      "                path.reverse()\n",
      "            return path, \"frag_used\"\n",
      "        \n",
      "        path.append( (q_name, end) )\n",
      "        path_q_name.add(q_name)\n",
      "        if q_name not in best_overlap_graph:\n",
      "            if reversing_path:\n",
      "                path.reverse()\n",
      "            return path, \"terminate_1\" \n",
      "        next_hit = best_overlap_graph[q_name][end]\n",
      "        #print next_hit\n",
      "        if next_hit == None:\n",
      "            if reversing_path:\n",
      "                path.reverse()\n",
      "            return path, \"terminate_2\"\n",
      "        \n",
      "        if next_hit[0] in best_overlap_graph: #if not mutual good hits, break the path\n",
      "            \n",
      "            # Using mutual best hit might be to strigent, \n",
      "            # bh = []\n",
      "            #if best_overlap_graph[next_hit[0]][\"5p\"]:\n",
      "            #    bh.append( best_overlap_graph[next_hit[0]][\"5p\"][0] )\n",
      "            #if best_overlap_graph[next_hit[0]][\"3p\"]:\n",
      "            #    bh.append( best_overlap_graph[next_hit[0]][\"3p\"][0] )\n",
      "            \n",
      "            bh = [h[0] for h in overlap_data[next_hit[0]].hits if proper_overlap_hit(h)]\n",
      "       \n",
      "            if q_name not in bh:\n",
      "                if reversing_path:\n",
      "                    path.reverse()\n",
      "                return path, \"branch\"\n",
      "        \n",
      "        q_name = next_hit[0]\n",
      "        \n",
      "        if q_name in path_q_name:\n",
      "            if reversing_path:\n",
      "                path.reverse()\n",
      "            return path, \"circle\"\n",
      "        \n",
      "        if next_hit[4][0] == 1: #reversed strand\n",
      "            end = reverse_end[end]\n",
      "\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Examples of the best overlapping paths\n",
      "--------------------------------------\n",
      "\n",
      "Here we show how to call the ``find_path`` function to get the overlapping path for both 5' and 3' from a seed fragment. We pick the longest fragment in the data set as the seed.  One will notice that the total fragments in from both 5'-parh and 3'-path are greater than the total number of fragments in the ``best_overlap_graph``. It mean that there are significant overlapping between the 5'-path and 3'-path. This is consistent with the genome is circular."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "len_qname = [ (overlap_data[x].length, x) for x in best_overlap_graph.keys() ]\n",
      "len_qname.sort()\n",
      "q_name = len_qname[-1][1]\n",
      "#q_name = \"0xa4919d66_0061381\"\n",
      "print \"The longest fragment is\", q_name\n",
      "\n",
      "path_5p, s_5p = find_path( (q_name, \"5p\") )\n",
      "path_3p, s_3p = find_path( (q_name, \"3p\") )\n",
      "print \"The number of the fragment of the 5' path is\", len(path_5p)\n",
      "print \"The number of the fragment of the 3' path is\", len(path_3p)\n",
      "print \"The total number of framgent of both 3' and 5' path is %d.\" % (len(path_3p)+len(path_5p)-1)\n",
      "print \"%d is greater than the total number of fragments %d in the best_overlap_graph.\" % (len(path_3p)+len(path_5p)-1, len(best_overlap_graph))\n",
      "print \"The begin of the 5'-path is\", path_5p[0]\n",
      "print \"The end of the 3'-path is\", path_3p[-1]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "The longest fragment is 0x4eaa5fa7_0001083\n",
        "The number of the fragment of the 5' path is"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 2221\n",
        "The number of the fragment of the 3' path is 2158\n",
        "The total number of framgent of both 3' and 5' path is 4378.\n",
        "4378 is greater than the total number of fragments 3390 in the best_overlap_graph.\n",
        "The begin of the 5'-path is ('0xa4919d66_0061381', '3p')\n",
        "The end of the 3'-path is ('0x24a73353_0082329', '3p')\n"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Build a squence database\n",
      "------------------------\n",
      "Since we get the paths that seems covering the whole genome, we can use the paths to construct draft contigs. In order to do that, we need to get the fragment sequences."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "seq_db = {}\n",
      "with open(\"pre_assembled_reads.fa\") as seq_file:\n",
      "    for l in seq_file:\n",
      "        l = l.strip()\n",
      "        if l[0] == \">\":\n",
      "            name = l[1:]\n",
      "            continue\n",
      "        else:\n",
      "            seq_db[name] = l"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 15
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Build a \"draft\" contig using the best overlapping paths\n",
      "-------------------------------------------------------\n",
      "\n",
      "The ``layout_path`` function goes through a layout path and fill in the base from the fragment sequences. Nothing is really special here. However, it is alwaye tricky to track the right orientation of the fragements. The function also tracks the fragments used in the layout. Each fragement can only be assigned to one layout/contig."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def rc_seq(seq):\n",
      "    rev_map = dict(zip(\"acgtACGTNn-\",\"tgcaTGCANn-\"))\n",
      "    return \"\".join([rev_map[c] for c in seq[::-1]])\n",
      "\n",
      "def layout_path(full_path, frag_used, out_fn, tig_name):\n",
      "    \n",
      "    if len(full_path) == 0 or full_path[0][0] in frag_used:\n",
      "        return None\n",
      "    \n",
      "    if len(full_path) == 1:\n",
      "        with open(\"singleton_\"+out_fn, \"w\") as out_seq_file:\n",
      "            seq = seq_db[full_path[0][0]]\n",
      "            print >>out_seq_file, \">%s\" % ( \"singleton_\" + tig_name  )\n",
      "            print >>out_seq_file, seq\n",
      "            frag_used.add(full_path[0][0])\n",
      "        return None\n",
      "    \n",
      "    first = full_path[0]\n",
      "    offset = 0\n",
      "    current_orientation = first[1]\n",
      "    revserse_orientation = {\"5p\":\"3p\", \"3p\":\"5p\"}\n",
      "    tig_seq = []\n",
      "    frag_in_layout = set()\n",
      "    frag_in_layout.add(first[0])\n",
      "    for second in full_path[1:]:\n",
      "        overlap_hit = overlap_data[first[0]][second[0]]\n",
      "        \n",
      "        q_strand, q_start, q_end, q_len = overlap_hit[3]\n",
      "        t_strand, t_start, t_end, t_len = overlap_hit[4]\n",
      "        #print overlap_hit, offset, current_orientation\n",
      "        seq = seq_db[first[0]]\n",
      "        if current_orientation == \"3p\":\n",
      "            seq = rc_seq(seq)\n",
      "        del tig_seq[offset:-1]\n",
      "        tig_seq.extend(seq)\n",
      "        \n",
      "        if current_orientation == \"5p\":\n",
      "            offset += q_start\n",
      "        elif current_orientation == \"3p\":\n",
      "            offset += q_len - q_end\n",
      "        if t_strand == 1:\n",
      "           current_orientation  = revserse_orientation[current_orientation] \n",
      "        frag_in_layout.add(first[0])\n",
      "        first = second\n",
      "        if second[0] in frag_in_layout:\n",
      "            break\n",
      "        if second[0] in frag_used:\n",
      "            break\n",
      "\n",
      "    seq = seq_db[first[0]]\n",
      "    if current_orientation == \"3p\":\n",
      "        seq = rc_seq(seq)\n",
      "    del tig_seq[offset:-1]\n",
      "    tig_seq.extend(seq)\n",
      "    frag_in_layout.add(first[0])\n",
      "    \n",
      "    frag_used.update( frag_in_layout )\n",
      "    \n",
      "    tig_seq = tig_seq[0:offset+len(seq)]\n",
      "    tig_seq = \"\".join(tig_seq)\n",
      "    with open(out_fn, \"w\") as out_seq_file:\n",
      "        print >>out_seq_file, \">%s\" % tig_name\n",
      "        print >>out_seq_file, tig_seq"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 16
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Loop through all fragments to generate all contigs and singletons\n",
      "-----------------------------------------------------------------\n",
      "\n",
      "Ok, ready for the primer time!!\n",
      "\n",
      "The code below will pick the longest fragment as the first seed and generate a contig from its overlapping path. Then, we will pick another longest one from those reads that have not been used in a contig. This process repeats until no more fragement is avaiable. \n",
      "\n",
      "In the data set, we generate one big contig (4650011 bp) and 49 singleton reads."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "frag_used = set()\n",
      "\n",
      "all_best_overlap_frags = set(best_overlap_graph.keys())\n",
      "\n",
      "unused_frag = all_best_overlap_frags - frag_used\n",
      "i = 0\n",
      "while len(unused_frag) != 0:\n",
      "\n",
      "    len_qname = [ (overlap_data[x].length, x) for x in unused_frag  ]\n",
      "    len_qname.sort()\n",
      "    q_name = len_qname[-1][1]\n",
      "    print \"iteration: \",i\n",
      "    print \"frag is used?\", q_name in frag_used\n",
      "    if q_name in frag_used:\n",
      "        continue\n",
      "\n",
      "    path_5p, s_5p = find_path( (q_name, \"5p\"), frag_used  )\n",
      "    path_3p, s_3p = find_path( (q_name, \"3p\"), frag_used  )\n",
      "    print \"seeding frag:\", q_name, \"Length:\", overlap_data[q_name].length\n",
      "    print \"number of unused frag:\", len(unused_frag), \"total overlapped frag:\", len(path_3p)+len(path_5p)\n",
      "    print len(path_5p), s_5p, len(path_3p), s_3p\n",
      "    print \"--------------------------\"\n",
      "    #print path_5p[0], path_3p[-1]\n",
      "    if len(path_5p) + len(path_3p) == 0:\n",
      "        out_fn = \"tig_%05d.fa\" % i\n",
      "        tig_name =\"tig%05d\" % i\n",
      "        with open(\"singleton_\"+out_fn, \"w\") as out_seq_file:\n",
      "            seq = seq_db[q_name]\n",
      "            print >>out_seq_file, \">%s\" % ( \"singleton_\" + tig_name  )\n",
      "            print >>out_seq_file, seq\n",
      "        frag_used.add(q_name)\n",
      "        unused_frag = all_best_overlap_frags - frag_used\n",
      "        i += 1\n",
      "        continue\n",
      "        \n",
      "    if len(path_5p) > 0:\n",
      "        #assert path_5p[-1] == path_3p[0]\n",
      "\n",
      "        full_path = path_5p + path_3p[1:]\n",
      "    else:\n",
      "        full_path = path_3p\n",
      "    layout_path(full_path, frag_used, \"tig_%05d.fa\" % i, \"tig%05d\" % i)\n",
      "    unused_frag = all_best_overlap_frags - frag_used\n",
      "    i += 1\n",
      "\n"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "iteration:  0\n",
        "frag is used? False\n",
        "seeding frag:"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 0x4eaa5fa7_0001083 Length: 14762\n",
        "number of unused frag: 3390 total overlapped frag: 4379\n",
        "2221 terminate_2 2158 terminate_2\n",
        "--------------------------\n",
        "iteration: "
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 1\n",
        "frag is used? False\n",
        "seeding frag: 0x720a6867_0101011 Length: 9562\n",
        "number of unused frag: 64 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  2\n",
        "frag is used? False\n",
        "seeding frag: 0xbb3ff171_0040732 Length: 8653\n",
        "number of unused frag: 63 total overlapped frag: 2\n",
        "1 branch 1 branch\n",
        "--------------------------\n",
        "iteration:  3\n",
        "frag is used? False\n",
        "seeding frag: 0xf9f76d57_0075336 Length: 8601\n",
        "number of unused frag: 62 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  4\n",
        "frag is used? False\n",
        "seeding frag: 0x24a73353_0082329 Length: 8529\n",
        "number of unused frag: 61 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  5\n",
        "frag is used? False\n",
        "seeding frag: 0x99a801d6_0119226 Length: 8390\n",
        "number of unused frag: 60 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  6\n",
        "frag is used? False\n",
        "seeding frag: 0x3176999f_0092788 Length: 8012\n",
        "number of unused frag: 59 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  7\n",
        "frag is used? False\n",
        "seeding frag: 0xd1f5cb9c_0041132 Length: 7742\n",
        "number of unused frag: 58 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  8\n",
        "frag is used? False\n",
        "seeding frag: 0x6a67d4d1_0054113 Length: 7635\n",
        "number of unused frag: 57 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  9\n",
        "frag is used? False\n",
        "seeding frag: 0x5e62b551_0087810 Length: 7612\n",
        "number of unused frag: 56 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  10\n",
        "frag is used? False\n",
        "seeding frag: 0x47c817fa_0037671 Length: 7592\n",
        "number of unused frag: 55 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  11\n",
        "frag is used? False\n",
        "seeding frag: 0xd34bbcb8_0096026 Length: 7586\n",
        "number of unused frag: 54 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  12\n",
        "frag is used? False\n",
        "seeding frag: 0xde53c97d_0130793 Length: 7568\n",
        "number of unused frag: 53 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  13\n",
        "frag is used? False\n",
        "seeding frag: 0xbab42971_0019685 Length: 7462\n",
        "number of unused frag: 52 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  14\n",
        "frag is used? False\n",
        "seeding frag: 0x95dcccbf_0021944 Length: 7446\n",
        "number of unused frag: 51 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  15\n",
        "frag is used? False\n",
        "seeding frag: 0x7b49a930_0025802 Length: 7402\n",
        "number of unused frag: 50 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  16\n",
        "frag is used? False\n",
        "seeding frag: 0xec2de5ef_0014019 Length: 7335\n",
        "number of unused frag: 49 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  17\n",
        "frag is used? False\n",
        "seeding frag: 0xb48bdbdf_0062004 Length: 7033\n",
        "number of unused frag: 48 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  18\n",
        "frag is used? False\n",
        "seeding frag: 0x201a3501_0131298 Length: 6986\n",
        "number of unused frag: 47 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  19\n",
        "frag is used? False\n",
        "seeding frag: 0xa3d2663c_0086930 Length: 6961\n",
        "number of unused frag: 46 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  20\n",
        "frag is used? False\n",
        "seeding frag: 0xe092bfc0_0025000 Length: 6941\n",
        "number of unused frag: 45 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  21\n",
        "frag is used? False\n",
        "seeding frag: 0x101cf700_0133174 Length: 6913\n",
        "number of unused frag: 44 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  22\n",
        "frag is used? False\n",
        "seeding frag: 0x27b1b79d_0098081 Length: 6687\n",
        "number of unused frag: 43 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  23\n",
        "frag is used? False\n",
        "seeding frag: 0x746a0885_0091093 Length: 6663\n",
        "number of unused frag: 42 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  24\n",
        "frag is used? False\n",
        "seeding frag: 0xf16e1447_0062854 Length: 6502\n",
        "number of unused frag: 41 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  25\n",
        "frag is used? False\n",
        "seeding frag: 0x2e1dde37_0073961 Length: 6444\n",
        "number of unused frag: 40 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  26\n",
        "frag is used? False\n",
        "seeding frag: 0x39af8897_0112579 Length: 6439\n",
        "number of unused frag: 39 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  27\n",
        "frag is used? False\n",
        "seeding frag: 0x4252156_0076315 Length: 6326\n",
        "number of unused frag: 38 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  28\n",
        "frag is used? False\n",
        "seeding frag: 0x8b32b9de_0082459 Length: 6255\n",
        "number of unused frag: 37 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  29\n",
        "frag is used? False\n",
        "seeding frag: 0xc0298763_0082661 Length: 6235\n",
        "number of unused frag: 36 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  30\n",
        "frag is used? False\n",
        "seeding frag: 0xfe2ddc8f_0006613 Length: 6219\n",
        "number of unused frag: 35 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  31\n",
        "frag is used? False\n",
        "seeding frag: 0xf2c9d7d9_0019647 Length: 6209\n",
        "number of unused frag: 34 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  32\n",
        "frag is used? False\n",
        "seeding frag: 0x5f497891_0044312 Length: 6178\n",
        "number of unused frag: 33 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  33\n",
        "frag is used? False\n",
        "seeding frag: 0x9d1fd3d5_0003951 Length: 6142\n",
        "number of unused frag: 32 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  34\n",
        "frag is used? False\n",
        "seeding frag: 0xd3405c1f_0120478 Length: 6056\n",
        "number of unused frag: 31 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  35\n",
        "frag is used? False\n",
        "seeding frag: 0xd84402bc_0135092 Length: 6022\n",
        "number of unused frag: 30 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  36\n",
        "frag is used? False\n",
        "seeding frag: 0x8e9f3eb7_0077492 Length: 6020\n",
        "number of unused frag: 29 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  37\n",
        "frag is used? False\n",
        "seeding frag: 0x6560473d_0129910 Length: 6009\n",
        "number of unused frag: 28 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  38\n",
        "frag is used? False\n",
        "seeding frag: 0x8e21ba01_0028034 Length: 5945\n",
        "number of unused frag: 27 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  39\n",
        "frag is used? False\n",
        "seeding frag: 0x328d5dfc_0076264 Length: 5867\n",
        "number of unused frag: 26 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  40\n",
        "frag is used? False\n",
        "seeding frag: 0x8c4c9750_0115911 Length: 5854\n",
        "number of unused frag: 25 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  41\n",
        "frag is used? False\n",
        "seeding frag: 0x99d966a0_0043182 Length: 5848\n",
        "number of unused frag: 24 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration: "
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        " 42\n",
        "frag is used? False\n",
        "seeding frag: 0xebaae05c_0001651 Length: 5718\n",
        "number of unused frag: 23 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  43\n",
        "frag is used? False\n",
        "seeding frag: 0x6015c8f3_0137444 Length: 5607\n",
        "number of unused frag: 22 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  44\n",
        "frag is used? False\n",
        "seeding frag: 0x820c3016_0097966 Length: 5594\n",
        "number of unused frag: 21 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  45\n",
        "frag is used? False\n",
        "seeding frag: 0xfb73db8e_0019839 Length: 5523\n",
        "number of unused frag: 20 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  46\n",
        "frag is used? False\n",
        "seeding frag: 0x5842b8b1_0141185 Length: 5483\n",
        "number of unused frag: 19 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  47\n",
        "frag is used? False\n",
        "seeding frag: 0x2876a699_0009758 Length: 5403\n",
        "number of unused frag: 18 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  48\n",
        "frag is used? False\n",
        "seeding frag: 0xf936c2fd_0094203 Length: 5372\n",
        "number of unused frag: 17 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  49\n",
        "frag is used? False\n",
        "seeding frag: 0xe5322dba_0130928 Length: 5027\n",
        "number of unused frag: 16 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  50\n",
        "frag is used? False\n",
        "seeding frag: 0x2ac1140b_0130957 Length: 4894\n",
        "number of unused frag: 15 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  51\n",
        "frag is used? False\n",
        "seeding frag: 0x53d92d4b_0116355 Length: 4758\n",
        "number of unused frag: 14 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  52\n",
        "frag is used? False\n",
        "seeding frag: 0xe09961a7_0000768 Length: 4746\n",
        "number of unused frag: 13 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  53\n",
        "frag is used? False\n",
        "seeding frag: 0x6bdaa7bf_0024191 Length: 4530\n",
        "number of unused frag: 12 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  54\n",
        "frag is used? False\n",
        "seeding frag: 0x2d6f1fdb_0064584 Length: 4419\n",
        "number of unused frag: 11 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  55\n",
        "frag is used? False\n",
        "seeding frag: 0xe6c70a1b_0009216 Length: 4386\n",
        "number of unused frag: 10 total overlapped frag: 2\n",
        "1 frag_used 1 frag_used\n",
        "--------------------------\n",
        "iteration:  56\n",
        "frag is used? False\n",
        "seeding frag: 0x1bed69cc_0052813 Length: 4332\n",
        "number of unused frag: 9 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  57\n",
        "frag is used? False\n",
        "seeding frag: 0xc48ec0c3_0089324 Length: 4173\n",
        "number of unused frag: 8 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  58\n",
        "frag is used? False\n",
        "seeding frag: 0x2afb8806_0097596 Length: 3840\n",
        "number of unused frag: 7 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  59\n",
        "frag is used? False\n",
        "seeding frag: 0xf953ded0_0012908 Length: 2975\n",
        "number of unused frag: 6 total overlapped frag: 2\n",
        "1 frag_used 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  60\n",
        "frag is used? False\n",
        "seeding frag: 0x3b256f1f_0102519 Length: 2397\n",
        "number of unused frag: 5 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  61\n",
        "frag is used? False\n",
        "seeding frag: 0x3f31dab0_0019006 Length: 2040\n",
        "number of unused frag: 4 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  62\n",
        "frag is used? False\n",
        "seeding frag: 0xcc88cea2_0015902 Length: 1651\n",
        "number of unused frag: 3 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n",
        "iteration:  63\n",
        "frag is used? False\n",
        "seeding frag: 0x2708c7d8_0113402 Length: 1617\n",
        "number of unused frag: 2 total overlapped frag: 2\n",
        "1 terminate_2 1 terminate_2\n",
        "--------------------------\n",
        "iteration:  64\n",
        "frag is used? False\n",
        "seeding frag: 0x612f720b_0078835 Length: 1275\n",
        "number of unused frag: 1 total overlapped frag: 2\n",
        "1 terminate_2 1 frag_used\n",
        "--------------------------\n"
       ]
      }
     ],
     "prompt_number": 17
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "What are the singletons?\n",
      "================================\n",
      "\n",
      "One can align the contigs to the canoincal E. coli K12 reference to evaluate the quality of the assembly. One can see from the alignment data below, it seems that some fragments have lower accuracy that does not pass the 98% identity threshold to be called \"overlapped\" with other fragments. Three of them does not have full alginments. In this case, one will need to examine the raw reads that corresponds to these three fragments to see whether they are minor variants or some lower quality reads are not filtered."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "!cat singleton_tig_000* > singletons.fa\n",
      "!blasr singletons.fa ecoli_k12_MG1655.fasta -m 4 -noSplitSubreads -nCandidates 20 -bestn 1 | awk '{print $1\" \"$2\" \"$4\" \"$6\" \"$7\" \"$8\" \"$8-($7-$6)}'"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00001 Escherichia_coli_K12 97.4017 0 9562 9562 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00002 Escherichia_coli_K12 97.3956 1431 8653 8653 1431\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00003 Escherichia_coli_K12 98.432 0 8601 8601 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00004 Escherichia_coli_K12 98.3834 0 8529 8529 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00005 Escherichia_coli_K12 99.4192 0 8390 8390 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00006 Escherichia_coli_K12 97.3447 0 8012 8012 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00007 Escherichia_coli_K12 98.1237 0 7742 7742 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00008 Escherichia_coli_K12 97.7821 0 7635 7635 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00009 Escherichia_coli_K12 98.2443 0 7612 7612 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00010 Escherichia_coli_K12 99.1766 2 7592 7592 2\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00011 Escherichia_coli_K12 97.1927 0 7586 7586 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00012 Escherichia_coli_K12 96.5121 0 7558 7568 10\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00013 Escherichia_coli_K12 97.4269 0 7462 7462 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00014 Escherichia_coli_K12 98.3228 0 7446 7446 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00015 Escherichia_coli_K12 99.4492 0 7402 7402 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00016 Escherichia_coli_K12 98.9734 0 7335 7335 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00017 Escherichia_coli_K12 97.9114 0 7033 7033 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00018 Escherichia_coli_K12 97.6773 0 6986 6986 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00019 Escherichia_coli_K12 96.437 0 6961 6961 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00020 Escherichia_coli_K12 97.4028 0 6941 6941 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00021 Escherichia_coli_K12 97.5844 0 6913 6913 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00022 Escherichia_coli_K12 99.1106 0 6687 6687 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00023 Escherichia_coli_K12 98.9452 0 6663 6663 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00024 Escherichia_coli_K12 97.4351 0 6502 6502 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00025 Escherichia_coli_K12 97.3736 0 6444 6444 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00026 Escherichia_coli_K12 96.5362 0 6439 6439 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00027 Escherichia_coli_K12 97.9105 0 6326 6326 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00028 Escherichia_coli_K12 97.7177 0 6255 6255 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00029 Escherichia_coli_K12 98.0028 0 6235 6235 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00030 Escherichia_coli_K12 98.09 0 6219 6219 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00031 Escherichia_coli_K12 95.9777 0 6209 6209 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00032 Escherichia_coli_K12 97.8922 0 6178 6178 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00033 Escherichia_coli_K12 97.8164 0 6142 6142 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00034 Escherichia_coli_K12 99.2136 0 6056 6056 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00035 Escherichia_coli_K12 97.2532 0 6022 6022 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00036 Escherichia_coli_K12 97.854 0 6020 6020 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00037 Escherichia_coli_K12 97.1526 0 6009 6009 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00038 Escherichia_coli_K12 96.5394 0 5945 5945 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00039 Escherichia_coli_K12 97.2628 0 5867 5867 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00040 Escherichia_coli_K12 97.3697 0 5854 5854 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00041 Escherichia_coli_K12 97.9883 0 5848 5848 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00042 Escherichia_coli_K12 97.7751 0 5718 5718 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00043 Escherichia_coli_K12 97.9378 0 5607 5607 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00044 Escherichia_coli_K12 99.2899 0 5594 5594 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00045 Escherichia_coli_K12 98.4484 0 5523 5523 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00046 Escherichia_coli_K12 97.7524 0 5483 5483 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00047 Escherichia_coli_K12 83.9108 0 5403 5403 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00048 Escherichia_coli_K12 97.5472 0 5372 5372 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00049 Escherichia_coli_K12 97.8931 0 5027 5027 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00050 Escherichia_coli_K12 97.9968 0 4894 4894 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00051 Escherichia_coli_K12 97.739 0 4758 4758 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00052 Escherichia_coli_K12 94.8541 58 4746 4746 58\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00053 Escherichia_coli_K12 97.3543 0 4530 4530 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00054 Escherichia_coli_K12 95.1304 0 4419 4419 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00055 Escherichia_coli_K12 96.9268 0 4386 4386 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00056 Escherichia_coli_K12 97.6502 0 4332 4332 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00057 Escherichia_coli_K12 97.4755 0 4173 4173 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00058 Escherichia_coli_K12 97.8816 0 3840 3840 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00059 Escherichia_coli_K12 97.7303 0 2975 2975 0\r\n",
        "singleton_tig00060 Escherichia_coli_K12 96.2158 0 2397 2397 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00061 Escherichia_coli_K12 97.5586 0 2040 2040 0\r\n",
        "singleton_tig00062 Escherichia_coli_K12 97.5768 0 1651 1651 0\r\n",
        "singleton_tig00063 Escherichia_coli_K12 97.8221 0 1617 1617 0\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "singleton_tig00064 Escherichia_coli_K12 97.032 0 1275 1275 0\r\n"
       ]
      }
     ],
     "prompt_number": 18
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "How good is the main contig?\n",
      "============================\n",
      "\n",
      "The main contig is 4650011bp. It seems that it should cover the whole genome.  We can see the whole genome alignment using \n",
      "[``gepard``]( http://www.helmholtz-muenchen.de/en/mips/services/analysis-tools/gepard/index.html ). We can see there is no larger scale mis-assembly. Promising!! (The strand of the K12 for this data is actuall slight different from the canonical one. We do expect to see some small structure variations.)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "Image(filename=\"tig_align.jpeg\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "jpeg": "/9j/4AAQSkZJRgABAgAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0a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\nFFABRRRQAUUUUAFFFFABRRRQAUUUUAf/2Q==\n",
       "output_type": "pyout",
       "prompt_number": 19,
       "text": [
        "<IPython.core.display.Image at 0x11c401d90>"
       ]
      }
     ],
     "prompt_number": 19
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Accuracy?\n",
      "---------\n",
      "We don't expect to get high accuracy from the \"draft\" assembly. The way we construct the contig is just to copy-and-paste from the pre-assembled reads which have mean identity 99.789% to the reference genome. For a 4.7Mb genome, we will expect 4700000 * (1-0.99787) ~ 10k differences. One can do another round of consensus using the pre-assembled reads. Or, we can align the raw reads on top of the draft contig and apply the \"Quiver\" algorithm to get the best accuracy from all raw data. Let's see how \"Quiver\" helps to remove the errors in the draft contig."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "First, let's evalute the differece between the draft contigs to the canonical reference with the ``dnadiff`` from \n",
      "[Mummer3 package](http://mummer.sourceforge.net/)"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "#!dnadiff ecoli_k12_MG1655.fasta tig_00000.fa -p diff_tig0\n",
      "!echo the number of SNPs is `cat diff_tig0.snps  | wc | awk '{print $1}'`"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "the number of SNPs is 10026\r\n"
       ]
      }
     ],
     "prompt_number": 20
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The draft contig has about 10k SNPs as what we expect from the estimation earlier. We apply ``Quvier`` consensus algorithm on the the draft contig using the following script::\n",
      "\n",
      "    #!/bin/bash\n",
      "    export SEYMOUR_HOME=/PathTo/PacBio/SMRTAnalysis\n",
      "    . $SEYMOUR_HOME/etc/setup.sh \n",
      "    cd /Path/To/WorkingDirectory\n",
      "    cp tig_00000.fa asm.ctg.fasta\n",
      "    referenceUploader -c -p $PWD -n assembly -f asm.ctg.fasta --skipIndexUpdate\n",
      "\n",
      "    compareSequences.py --info --useGuidedAlign --algorithm=blasr --nproc=24 --noXML --h5mode=w --h5fn=out.cmp.h5 --minAccuracy=0.70 --minLength=200 \\\n",
      "    -x -nCandidates 50 -x -minMatch 12 -x -bestn 1 -x -minPctIdentity 70.0 /Path/To/WorkingDirectory/input.fofn assembly/\n",
      "    \n",
      "    loadPulses /Path/To/WorkingDirectory/input.fofn out.cmp.h5\\\n",
      "    -metrics DeletionQV,IPD,InsertionQV,PulseWidth,QualityValue,MergeQV,SubstitutionQV,DeletionTag -byread \\\n",
      "    \n",
      "    cmph5tools.py sort out.cmp.h5 --tmp /scratch\n",
      "    \n",
      "    variantCaller.py --algorithm quiver -j 16 --referenceFilename assembly/sequence/assembly.fasta \\\n",
      "    --parameters best -o output.gff  -o output.fasta -o output.fastq -q 0  -X 80 -x 5 --mapQvThreshold 0 out.cmp.h5\n",
      "\n",
      "This will generate the consensus sequence using Quiver. The consensus output is in ``output.fasta``. We find 26 SNPs in the final consensus results. This gives a lower bound of the final assembly phred QV about -10*log10(26/4639675) = 52.5.  One can see that the ``Quiver`` algorithm which utilizes all important information from the raw trace signal is able to correct the ~10k errors in the\n",
      "draft assembly. This allows us more freedom while constructing the initial draft assembly.  We do not need to construct perfect draft assembly from perfect reads. The goal for constructing the draft is to get the contiguity right. Given the long reads, one can expect that most mapping is unambiguous and that will lead to almost perfect consensus using ``Quiver``.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "#dnadiff ecoli_k12_MG1655.fasta output.fasta -p diff_quiver_tig0\n",
      "!echo the number of SNPs is `cat diff_quiver_tig0.snps | wc | awk '{print $1}'`\n",
      "from math import log\n",
      "print \"The phred scale QV of the assembly is about %0.1f\" % (-10*log(26.0/4639675)/log(10))\n",
      "print \"The concordnace is about %0.5f%%\" % (100*(1-26.0/4639675))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "the number of SNPs is 26\r\n"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "The phred scale QV of the assembly is about 52.5\n",
        "The concordnace is about 99.99944%\n"
       ]
      }
     ],
     "prompt_number": 21
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Final Thought\n",
      "=============\n",
      "\n",
      "I had some fun with this exercise. I think this approach is general but it will\n",
      "need more tests for sure. We only have n=1 successful story for now. However,\n",
      "the reality is when we can generate longer and longer reads, the assembly\n",
      "problem will become easier and easier. DNA sequencing is about connecting\n",
      "different bases across distance. Some problems can be only solved by enough\n",
      "read length spans. I do like to see more fundamental theories published about\n",
      "genome assembly and haplotyping (e.g. the classical Lander-Waterman paper and\n",
      "the recent careful analysis done ([preprint](http://arxiv.org/abs/1301.0068)) \n",
      "by Davie Tse's group).  Getting good consensus from long reads is\n",
      "probably a relative simpler problem than dealing with combinatorial explosion\n",
      "when one tries to infer long range information using short reads.  Taking\n",
      "bacteria assembly as an example, it is likely impossible to assemble the E.\n",
      "coli K12 genome to single contig with single-end short read library.  Without\n",
      "any long pair-end or jumping libraries, the single-end short read data simply\n",
      "does not the power to resolve various scales of repeat structure in the genome.\n",
      "Such constraints can not be solved by simply sequencing more (see Tse's argument).\n",
      "Actually, regardless how cheap DNA sequencing is, without proper understanding\n",
      "the fundamental theories and mathematical constraints, one could just waste\n",
      "money collecting tons of useless data.  Cheap data still costs money. Cheap data\n",
      "that can not solve problems is actually expansive. One can sequence E coli. K12 genome\n",
      "to 10000x coverage with short read technologies cheaply. However, without\n",
      "proper libraries to get long range information, 10000x coverage from short fragments \n",
      "will still lead to fragmented assemblies. One the other hand, one can see from\n",
      "the example in this ``IPython`` Notebook. Once we can generate good quality long reads,\n",
      "even an assembly newbie like me can put together a prototype of an assembler with python\n",
      "in a few days. By the way, it only takes 20 seconds on my 2012 MacBook Air (Intel i7) to \n",
      "get the draft assembly from the ``m4`` file.\n"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "!time python Write_An_Assembler.py > /dev/null"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\r\n",
        "real\t0m21.142s\r\n",
        "user\t0m20.326s\r\n",
        "sys\t0m0.745s\r\n"
       ]
      }
     ],
     "prompt_number": 22
    }
   ],
   "metadata": {}
  }
 ]
}