notebooks/response_reviewers_ev.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Response to Reviewer 2: EV\n",
    "\n",
    "Reviewer 2 argued whether the dorsal stream can be considered similar to EV. In this notebook, we will plot the distances of each ROI in the dorsal and ventral stream to the EV rois, and then compute the difference `dist(ventral, EV) - dist(dorsal, EV)` in each hemisphere. We're going to do this analysis with the task data, so that we can run some statistics across subjects.\n",
    "\n",
    "[Skip to the analyses](#Load-data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Import and define functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set_context('notebook')\n",
    "sns.set_style('ticks')\n",
    "sns.set_palette('dark')\n",
    "\n",
    "import sys\n",
    "p = ['../']\n",
    "for pp in p:\n",
    "    if pp not in sys.path:\n",
    "        sys.path.insert(1, pp)\n",
    "\n",
    "import expdir\n",
    "from mvpa2.suite import *\n",
    "from mds_rois import xboost\n",
    "fns = expdir.expdir()\n",
    "from scipy.spatial import distance as dist\n",
    "import seaborn as sns\n",
    "from os.path import join as pjoin\n",
    "import pandas as pd\n",
    "from scipy.spatial.distance import cdist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "img_dir = './img_response_reviewers'\n",
    "try:\n",
    "    os.makedirs(img_dir)\n",
    "except OSError:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def load_dsm(subnr):\n",
    "    fn = pjoin(fns.mvpadir(subnr), 'roi_rsa_jimroi_5r_fisher_nooverlap/dsm_condition.hdf5')\n",
    "    ds = h5load(fn)\n",
    "    return ds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def make_ds_distance(ds, make_symmetric=False, rank=False):\n",
    "    if rank:\n",
    "        ds_ = np.apply_along_axis(rankdata, 1, ds)\n",
    "    else:\n",
    "        ds_ = ds.samples\n",
    "    if make_symmetric:\n",
    "        # add transpose and take only upper triangular with diagonal\n",
    "        ds_new = []\n",
    "        for row in ds_:\n",
    "            tmp = row.reshape(8, 8)\n",
    "            tmp += tmp.T\n",
    "            tmp /= 2\n",
    "            ds_new.append(tmp[np.triu_indices_from(tmp)])\n",
    "        ds_ = np.array(ds_new)\n",
    "        \n",
    "    ds_dist = squareform(pdist(ds_, metric='correlation'))\n",
    "    return Dataset(ds_dist, sa=ds.sa, fa={'roi': ds.sa.targets})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_average_distance_ds(roi_rdm):\n",
    "    \"\"\"Generate a group distance matrix dataset\"\"\"\n",
    "    dsms_avg = np.mean([pdist(ds, 'correlation') for ds in roi_rdm], axis=0)\n",
    "    dsms_avg = squareform(dsms_avg)\n",
    "    dsms_avg = Dataset(dsms_avg,\n",
    "                       sa=roi_rdm[0].sa,\n",
    "                       fa={'roi': roi_rdm[0].sa.targets})\n",
    "    return dsms_avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def reorder_dataset(ds, labels_ord):\n",
    "    labels_orig = ds.sa.targets\n",
    "    ds_ord = ds.copy()\n",
    "    idx_reorder = [np.where(lbl == labels_orig)[0][0] for lbl in labels_ord]\n",
    "\n",
    "    ds_ord = ds_ord[:, idx_reorder][idx_reorder, :]\n",
    "    return ds_ord"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def test_reorder_dataset():\n",
    "    # make a symmetric dataset\n",
    "    ds_ = np.zeros((10, 10))\n",
    "    ds_[np.triu_indices_from(ds_, k=1)] = np.arange(5*9) + 1\n",
    "    ds_ += ds_.T\n",
    "    ds = Dataset(ds_, sa={'targets': np.arange(10)})\n",
    "    ds_ord = reorder_dataset(ds, np.arange(10)[::-1])\n",
    "    assert(np.array_equal(ds.samples, ds_ord.samples[::-1, ::-1]))\n",
    "    assert(np.array_equal(np.unique(ds_ord.samples - ds_ord.samples.T), [0.]))\n",
    "\n",
    "    ds_ord = reorder_dataset(ds, [0, 2, 1] + range(3, 10))\n",
    "    # check it is still symmetric\n",
    "    assert(np.array_equal(np.unique(ds_ord.samples - ds_ord.samples.T), [0.]))\n",
    "    assert(np.array_equal(ds[1, 2], ds_ord[2, 1]))\n",
    "    assert(np.array_equal(ds[2, 1], ds_ord[1, 2]))\n",
    "    \n",
    "    # make another dataset with a \"cross\"\n",
    "    ds_ = np.zeros((10, 10))\n",
    "    ds_[0] = 1.\n",
    "    ds_[:, 0] = 1.\n",
    "    ds = Dataset(ds_, sa={'targets': np.arange(10)})\n",
    "\n",
    "    ds_ord = reorder_dataset(ds, [1, 0] + range(2, 10))\n",
    "    assert(np.array_equal(ds_ord.samples[1], np.ones(10)))\n",
    "    assert(np.array_equal(ds_ord.samples[:, 1], np.ones(10)))\n",
    "test_reorder_dataset()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def get_labels_color(labels):\n",
    "    #assign different colors depending on system\n",
    "    # using colorbrewer2.org qualitative 6-class palette\n",
    "    early_visual = '#0000FF'\n",
    "    dorsal_core_system = '#b1551c'#'#e4761a'\n",
    "    ventral_core_system = '#e41a1a'\n",
    "    anterior_core_system = '#e4a11a'#'#4daf4a'\n",
    "    theory_of_mind = '#008100'\n",
    "    precuneus = '#099997'\n",
    "\n",
    "    label2color = {\n",
    "    'EV1 - L': early_visual,\n",
    "    'EV1 - R': early_visual,\n",
    "    'EV2 - L': early_visual,\n",
    "    'EV2 - R': early_visual,\n",
    "    'IFG1 - L': anterior_core_system,\n",
    "    'IFG2 - L': anterior_core_system,\n",
    "    'IFG2 - R': anterior_core_system,\n",
    "    'MPFC - L': theory_of_mind,\n",
    "    'MPFC - R': theory_of_mind,\n",
    "    'OccFus - L': ventral_core_system,\n",
    "    'OccFus - R': ventral_core_system,\n",
    "    'TPJ - L': theory_of_mind,\n",
    "    'TPJ - R': theory_of_mind,\n",
    "    'aFus - R': ventral_core_system,\n",
    "    'aMTG - L': dorsal_core_system,\n",
    "    'aMTG - R': dorsal_core_system,\n",
    "    'dPreCun - L': precuneus,\n",
    "    'dPreCun - R': precuneus,\n",
    "    'mFus - L': ventral_core_system,\n",
    "    'mFus - R': ventral_core_system,\n",
    "    'mMTG - L': dorsal_core_system,\n",
    "    'mMTG - R': dorsal_core_system,\n",
    "    'mPreCun - L': precuneus,\n",
    "    'mPreCun - R': precuneus,\n",
    "    'pFus - L': ventral_core_system,\n",
    "    'pFus - R': ventral_core_system,\n",
    "    'pMTG - L': dorsal_core_system,\n",
    "    'pMTG - R': dorsal_core_system,\n",
    "    'vPreCun - L': precuneus,\n",
    "    'vPreCun - R': precuneus,\n",
    "    }\n",
    "\n",
    "    labels_color = [label2color[lbl] for lbl in labels]\n",
    "    assert(len(labels) == len(labels_color))\n",
    "    return labels_color"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def plot_mds_together(coordinates1, coordinates2, labels, labels_color=None):\n",
    "    \"\"\"Plot the two MDS together with lines going from coordinates2 to coordinates2\"\"\"\n",
    "    fig = plt.figure(figsize=(16, 12))\n",
    "    nd = coordinates1.shape[1]\n",
    "    ax = fig.add_subplot(111)\n",
    "\n",
    "    # I should receive a set of coordinates, a set of labels, and an adjacency matrix\n",
    "    assert(len(coordinates1) == len(labels))\n",
    "    assert(len(coordinates2) == len(labels))\n",
    "    ax.scatter(coordinates1[:, 0], coordinates1[:, 1], s=1, c='k')\n",
    "    ax.scatter(coordinates2[:, 0], coordinates2[:, 1], s=1, c='k')\n",
    "\n",
    "    \n",
    "    # Add lines between the two points\n",
    "    common_kwargs = dict(\n",
    "        alpha=0.7,\n",
    "        #color=m.to_rgba(value),\n",
    "        #alpha=norm_alpha(value),\n",
    "        #linewidth=norm_size(value)#value*linewidth_mult\n",
    "        #linewidth=4\n",
    "    )\n",
    "    for (xs, ys), (xe, ye), color in zip(coordinates2, coordinates1, labels_color):\n",
    "        #ax.plot([xs, xe], [ys, ye], color=color, **common_kwargs)\n",
    "        ax.arrow(xs, ys, xe - xs, ye - ys, color=color, \n",
    "                 width=0.0002, **common_kwargs)\n",
    "\n",
    "\n",
    "    # Add the labels only for the first dataset\n",
    "    text_kwargs = dict(fontsize=16,\n",
    "                    horizontalalignment='center',\n",
    "                    verticalalignment='center',\n",
    "                    weight='bold')\n",
    "    \n",
    "    if labels_color is None:\n",
    "        labels_color = ['black'] * len(labels)\n",
    "\n",
    "    for ((x, y), label, color) in zip(coordinates2, labels, labels_color):\n",
    "        ax.text(x, y, label, color='white', #color, \n",
    "                bbox=dict(facecolor=color, alpha=0.6, boxstyle=\"round,pad=0.3\"), **text_kwargs)\n",
    "\n",
    "        \n",
    "    pretty_plot(ax)\n",
    "    ax.spines['left'].set_visible(False)\n",
    "    ax.spines['bottom'].set_visible(False)\n",
    "    ax.xaxis.set_visible(False)\n",
    "    ax.yaxis.set_visible(False)\n",
    "    \n",
    "    return fig#, ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def idx_labels(which, labels):\n",
    "    \"\"\"Returns a list of indices of `which` in `labels`\"\"\"\n",
    "    labels_ = np.array(labels)\n",
    "    idx = []\n",
    "    for w in which:\n",
    "        idx.append(np.where(labels_ == w)[0][0])\n",
    "    return idx\n",
    "\n",
    "def test_idx_labels():\n",
    "    which = list('abc')\n",
    "    labels = list('defgbac')\n",
    "    idx = idx_labels(which, labels)\n",
    "    for i, w in zip(idx, which):\n",
    "        assert(w == labels[i])\n",
    "test_idx_labels()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def plot_distance(roi_dist, which_rois, color, axs=None, title=True, order=None):\n",
    "    if axs is None:\n",
    "        fig, axs = plt.subplots(1, 3, sharey=True, figsize=(10, 3))\n",
    "    if order is None:\n",
    "        order = which_rois.keys()\n",
    "    for i, roi in enumerate(order):\n",
    "        ax = axs[i]\n",
    "        compareto = which_rois[roi]\n",
    "        i_row = idx_labels([roi], labels_original)\n",
    "        i_cols = idx_labels(compareto, labels_original)\n",
    "        idx = np.ix_(i_row, i_cols)\n",
    "        ds_roi = []\n",
    "        for ds in roi_dist:\n",
    "            ds_roi.append(ds.samples[idx].flatten())\n",
    "        sns.pointplot(data=np.array(ds_roi), ax=ax, color=color)\n",
    "        ax.set_xticklabels(compareto);\n",
    "        if title:\n",
    "            ax.set_title(roi)\n",
    "        sns.despine()\n",
    "    axs[0].set_ylabel('Correlation distance')\n",
    "    return axs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Load data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### Task data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# this contains the first level RDMs\n",
    "roi_rdm = [load_dsm(i) for i in range(1, 34)]\n",
    "# this contains the second level RDMs (between ROIs) for each subject\n",
    "roi_dist = [make_ds_distance(ds, make_symmetric=True) for ds in roi_rdm]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# which rois to look at\n",
    "dorsal_rois = {\n",
    "    'EV1 - L': [w + 'MTG - L' for w in 'pma'],\n",
    "    'EV2 - L': [w + 'MTG - L' for w in 'pma'],\n",
    "    'EV1 - R': [w + 'MTG - R' for w in 'pma'],\n",
    "    'EV2 - R': [w + 'MTG - R' for w in 'pma']\n",
    "}\n",
    "ventral_rois = {\n",
    "    'EV1 - L': ['OccFus - L'] + [w + 'Fus - L' for w in 'pm'],\n",
    "    'EV2 - L': ['OccFus - L'] + [w + 'Fus - L' for w in 'pm'],\n",
    "    'EV1 - R': ['OccFus - R'] + [w + 'Fus - R' for w in 'pma'],\n",
    "    'EV2 - R': ['OccFus - R'] + [w + 'Fus - R' for w in 'pma'],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "We will consider the following ROIs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'EV1 - L': ['pMTG - L', 'mMTG - L', 'aMTG - L'],\n",
       " 'EV1 - R': ['pMTG - R', 'mMTG - R', 'aMTG - R'],\n",
       " 'EV2 - L': ['pMTG - L', 'mMTG - L', 'aMTG - L'],\n",
       " 'EV2 - R': ['pMTG - R', 'mMTG - R', 'aMTG - R']}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dorsal_rois"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'EV1 - L': ['OccFus - L', 'pFus - L', 'mFus - L'],\n",
       " 'EV1 - R': ['OccFus - R', 'pFus - R', 'mFus - R', 'aFus - R'],\n",
       " 'EV2 - L': ['OccFus - L', 'pFus - L', 'mFus - L'],\n",
       " 'EV2 - R': ['OccFus - R', 'pFus - R', 'mFus - R', 'aFus - R']}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ventral_rois"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "labels_original = roi_dist[0].sa.targets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Here we are plotting the distances of each ROI (in the dorsal and ventral stream) to the EV ROIs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python2.7/dist-packages/matplotlib/collections.py:571: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  if self._edgecolors == str('face'):\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjQAAAI0CAYAAAAKi7MDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a2dbab690>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjQAAAI0CAYAAAAKi7MDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a2db44610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(2, 2, sharey=True, figsize=(8, 8))\n",
    "plot_distance(roi_dist, dorsal_rois, color='#b1551c', axs=axs[0], \n",
    "              order=['EV1 - L', 'EV1 - R'])\n",
    "plot_distance(roi_dist, ventral_rois, color='#e41a1a', axs=axs[1], title=False,\n",
    "             order=['EV1 - L', 'EV1 - R'])\n",
    "plt.tight_layout()\n",
    "#fig.savefig(pjoin(img_dir, 'distance_ifg.png'), dpi=300)\n",
    "\n",
    "fig, axs = plt.subplots(2, 2, sharey=True, figsize=(8, 8))\n",
    "plot_distance(roi_dist, dorsal_rois, color='#b1551c', axs=axs[0], \n",
    "              order=['EV2 - L', 'EV2 - R'])\n",
    "plot_distance(roi_dist, ventral_rois, color='#e41a1a', axs=axs[1], title=False,\n",
    "             order=['EV2 - L', 'EV2 - R'])\n",
    "plt.tight_layout()\n",
    "fig.savefig(pjoin(img_dir, 'distance_ev.png'), dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Now we compute the mean within each system (dorsal | ventral), and compute an index as follows\n",
    "\n",
    "\n",
    "$$\n",
    "I = \\frac{d(\\text{ventral}, \\text{EV}) - d(\\text{dorsal}, \\text{EV})}{d(\\text{ventral}, \\text{EV}) + d(\\text{dorsal}, \\text{EV})}\n",
    "$$\n",
    "\n",
    "such that\n",
    "\n",
    "$$\n",
    "\\begin{array}*\n",
    "I > 0 & \\text{if EV is closer to the dorsal stream} \\\\\n",
    "I < 0 & \\text{if EV is closer to the ventral stream} \\\\\n",
    "I = 0 & \\text{if EV is equidistant to the dorsal and ventral streams}\\\\\n",
    "\\end{array}\n",
    "$$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# extract means for each subject and each system\n",
    "def compute_distance_from_roi(roi, roistoaverage, labels_original, roi_distances):\n",
    "    \"\"\"\n",
    "    Computes average distances of one roi to a list of rois.\n",
    "    \n",
    "    Arguments\n",
    "    ---------\n",
    "    roi : str\n",
    "    roistoaverage : list of str\n",
    "    labels_original : list of str\n",
    "        the original labels in order as `roi_distances`\n",
    "    roi_distances : list of mvpa2.Dataset\n",
    "        square datasets containing the distances between ROIs,\n",
    "        each dataset is coming from one subject\n",
    "    \n",
    "    Returns\n",
    "    -------\n",
    "    ds_ : array (len(roi_distances), )\n",
    "        distances from roi to roistoaverage\n",
    "    \"\"\"\n",
    "    i_row = idx_labels([roi], labels_original)\n",
    "    i_cols = idx_labels(roistoaverage, labels_original)\n",
    "    ds_ = []\n",
    "    for ds in roi_distances:\n",
    "        ds_.append(ds.samples[np.ix_(i_row, i_cols)].mean())\n",
    "    ds_ = np.array(ds_)\n",
    "    return ds_\n",
    "\n",
    "mean_dorsal = dict()\n",
    "mean_ventral = dict()\n",
    "for roi in dorsal_rois:\n",
    "    compareto_dorsal = dorsal_rois[roi]\n",
    "    compareto_ventral = ventral_rois[roi]\n",
    "    # extract the averages across dorsal and ventral ROIs\n",
    "    mean_dorsal[roi] = compute_distance_from_roi(roi, compareto_dorsal, labels_original, roi_dist)\n",
    "    mean_ventral[roi] = compute_distance_from_roi(roi, compareto_ventral, labels_original, roi_dist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# make a dataframe for plotting\n",
    "df_dorsal = pd.DataFrame(mean_dorsal)\n",
    "df_dorsal['system'] = 'dorsal'\n",
    "\n",
    "df_ventral = pd.DataFrame(mean_ventral)\n",
    "df_ventral['system'] = 'ventral'\n",
    "\n",
    "df = pd.concat((df_dorsal, df_ventral))\n",
    "df_long = pd.melt(df, id_vars='system')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEJCAYAAAB/pOvWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/jgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a8033a810>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1)\n",
    "sns.violinplot(x='variable', y='value', hue='system',\n",
    "               data=df_long, palette=sns.color_palette('Set2'), ax=ax)\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('Correlation distance')\n",
    "sns.despine()\n",
    "fig.savefig(pjoin(img_dir, 'violin_dist_ev.png'), dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# now compute the index\n",
    "def compute_index(a, b):\n",
    "    return (a-b)/(a+b)\n",
    "df_difference = compute_index(df_ventral.iloc[:, :4], df_dorsal.iloc[:, :4])\n",
    "df_difference_long = pd.melt(df_difference)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEJCAYAAABsc6siAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngBnNAwBgdgF6paeh9tKUFwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a27ce4a50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1)\n",
    "sns.violinplot(x='variable', y='value', color='gray',\n",
    "              data=df_difference_long, ax=ax, )\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('Dorsal index')\n",
    "sns.despine()\n",
    "fig.savefig(pjoin(img_dir, 'violin_index_ev.png'), dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZYAAAEJCAYAAAC3yAEAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLEREJCoGCxERiYrBQkREovp/abbKzotvZ08AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a27bf6ed0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1)\n",
    "ax.axhline(color='lightgray', linestyle='dashed')\n",
    "sns.pointplot(x='variable', y='value', color='black', join=False, n_boot=10000, \n",
    "              data=df_difference_long, ax=ax)\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('Dorsal index')\n",
    "sns.despine()\n",
    "fig.savefig(pjoin(img_dir, 'pointplot_index_ev.png'), dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEJCAYAAABsc6siAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npFMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f9a27cb0cd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 1)\n",
    "ax.axhline(color='red', linestyle='dashed')\n",
    "sns.violinplot(x='variable', y='value', color='lightgray',\n",
    "              data=df_difference_long, ax=ax, inner='point')\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('Dorsal index')\n",
    "sns.despine()\n",
    "fig.savefig(pjoin(img_dir, 'pointplot_index_ev.png'), dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Now bootstrap the index to obtain confidence intervals for those values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# bootstrapping\n",
    "nrows = len(df_difference)\n",
    "nbs = 10000\n",
    "df_difference_bs = []\n",
    "np.random.seed(765)\n",
    "for idx in xboost(nrows, nbs):\n",
    "    df_difference_bs.append(df_difference.iloc[idx].mean(axis=0))\n",
    "df_difference_bs = pd.concat(df_difference_bs, axis=1).T\n",
    "\n",
    "cis = np.percentile(df_difference_bs, [2.5, 97.5], axis=0).T\n",
    "avg = df_difference.mean(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EV1 - L: -0.01 [-0.04  0.01]\n",
      "EV1 - R: -0.02 [-0.07  0.01]\n",
      "EV2 - L: -0.05 [-0.08 -0.02]\n",
      "EV2 - R: -0.07 [-0.11 -0.04]\n"
     ]
    }
   ],
   "source": [
    "for lbl, a, ci in zip(avg.index.tolist(), avg, cis):\n",
    "    print(\"{0}: {1} {2}\".format(lbl, np.round(a, 2), np.round(ci, 2)))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}