figures/titer-tree-heatmap.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import pickle\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import baltic as bt\n",
    "from itertools import combinations, product\n",
    "import matplotlib.gridspec as gridspec\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "titer_file = '../data/titers_katzelnick2015/normalized_titers.csv'\n",
    "tree_file = '../data/trees/titered_annotated_tree.nexus'\n",
    "colors = pickle.load(open('./colors.p', 'rb'))\n",
    "out_path = './png/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree = bt.loadNexus(tree_file, absoluteTime=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def label_mrcas(tree):\n",
    "    serotypes = ['DENV1', 'DENV2', 'DENV3', 'DENV4']\n",
    "    \n",
    "    for serotype in serotypes:\n",
    "        serotype_tips = [k.numName for k in tree.leaves if k.numName.startswith(serotype)]\n",
    "\n",
    "        mrca = sorted([node for node in tree.nodes \n",
    "                       if all([l in node.leaves for l in serotype_tips])], \n",
    "                     key = lambda n: node.height)[-1]\n",
    "        mrca.traits['mrca'] = serotype\n",
    "    \n",
    "label_mrcas(tree)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "titers = pd.read_csv(titer_file, sep=',', index_col=None)\n",
    "titers = titers.pivot(index='virus', columns='serum', values='titer')\n",
    "\n",
    "titered_strains = set(titers.columns.values).union(set(titers.index.values))\n",
    "tip_order = [ k.numName for k in sorted(tree.Objects, key = lambda k: -1*k.y) if k.branchType=='leaf']\n",
    "# serotype_tip_order = { s: [k for k in tip_order if k.startswith('DENV%d'%s) and k in titered_strains] \n",
    "#                       for s in [1,2,3,4]}\n",
    "\n",
    "titers = titers.reindex(index =tip_order,columns=tip_order)\n",
    "\n",
    "titers = titers.dropna(axis=0, how='all')\n",
    "titers = titers.dropna(axis=1, how='all')\n",
    "titers.index.name = None\n",
    "titers.columns.name = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "titers.to_csv('./attachments/titer_heatmap.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_tree(tree, ax):\n",
    "    \n",
    "    ysf = 1.0#2.5 #yscale factor\n",
    "    \n",
    "    branchWidth=2 ## default branch width\n",
    "    ll = tree\n",
    "    plt.yticks(size=0)\n",
    "        \n",
    "    for k in ll.Objects: ## iterate over objects in tree\n",
    "        x=k.x ## or use absolute time instead\n",
    "        y=k.y*ysf ## get y position from .drawTree that was run earlier, but could be anything else\n",
    "\n",
    "        xp=k.parent.x ## get x position of current object's parent\n",
    "        if x==None: ## matplotlib won't plot Nones, like root\n",
    "            x=0.0\n",
    "        if xp==None:\n",
    "            xp=x\n",
    "            \n",
    "        if isinstance(k,bt.leaf) or k.branchType=='leaf': ## if leaf...\n",
    "            s=30 ## tip size can be fixed\n",
    "            \n",
    "            plt.plot((x,tree.treeHeight*1.1), (y,y), c='lightgray', linestyle='--')\n",
    "#             plt.text(tree.treeHeight*1.35, y, k.numName, fontsize=9, va='center', ha='right')\n",
    "            plt.scatter(x,y,s=s,facecolor='k',edgecolor='none',zorder=11) ## plot circle for every tip\n",
    "            plt.scatter(x,y,s=s+0.8*s,facecolor='k',edgecolor='none',zorder=10) ## plot black circle underneath\n",
    "                        \n",
    "        elif isinstance(k,bt.node) or k.branchType=='node': ## if node...\n",
    "            plt.plot([x,x],[k.children[-1].y*ysf,k.children[0].y*ysf],lw=branchWidth,color='k',ls='-',zorder=9)\n",
    "            \n",
    "            if 'mrca' in k.traits:\n",
    "                ax.text((xp+x)/2., y+0.5, k.traits['mrca'])\n",
    "            \n",
    "        plt.plot([xp,x],[y,y],lw=branchWidth,color='k',ls='-',zorder=9)\n",
    "\n",
    "        \n",
    "    ax.spines['top'].set_visible(False) ## no axes\n",
    "    ax.spines['right'].set_visible(False)\n",
    "    ax.spines['left'].set_visible(False)\n",
    "    ax.spines['bottom'].set_bounds(0,tree.treeHeight)\n",
    "    y0,y1 = ax.get_ylim()\n",
    "    xticks = ax.xaxis.get_major_ticks()\n",
    "    xticks[0].label1.set_visible(False)\n",
    "    xticks[-1].label1.set_visible(False)\n",
    "    xticks[-2].label1.set_visible(False)\n",
    "    ax.set_xlabel('Substitutions per site', fontsize=14)\n",
    "    for tick in ax.xaxis.get_major_ticks():\n",
    "        tick.label.set_fontsize(14) \n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABEcAAAIxCAYAAACvjdZoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nD/+V2fCFq9wEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107dfabd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def add_subfigure_label(ax, label, x=-0.05, y=1.0):\n",
    "    ax.text(x, y, label, va='top', fontdict={'fontsize': 18, 'fontweight': 'bold'}, transform=ax.transAxes)\n",
    "\n",
    "sns.set(style='white', font_scale=1.)\n",
    "fig = plt.figure(figsize=(8.5,7))\n",
    "tree_ax = fig.add_axes([0,0,1,1])\n",
    "heatmap_ax = fig.add_axes([0.9,0.045,0.7,0.935])\n",
    "cbar_ax = fig.add_axes([1.65,0.05,0.03,0.9])\n",
    "\n",
    "plt.sca(tree_ax)\n",
    "plot_tree(tree, tree_ax)\n",
    "plt.sca(heatmap_ax)\n",
    "\n",
    "heatmap_ax.set_facecolor('whitesmoke')\n",
    "sns.heatmap(titers, ax=heatmap_ax, cbar_kws={'label': r'$log_2$(titers)'}, \n",
    "            cmap='viridis',  yticklabels=False, xticklabels = False,\n",
    "            cbar_ax=cbar_ax,linewidths=1,linecolor='whitesmoke')\n",
    "heatmap_ax.figure.axes[-1].yaxis.label.set_size(14)\n",
    "cbar_ax.tick_params(labelsize=14)\n",
    "\n",
    "    \n",
    "serotype_order = ['DENV4', 'DENV2', 'DENV3', 'DENV1']\n",
    "serotype_proportion = {s: len([t for t in titers.columns.values if t.startswith(s)])/float(len(titers.columns.values)) for s in serotype_order}\n",
    "\n",
    "x0, xf = heatmap_ax.get_xlim()#0.,1.\n",
    "y0, yf = heatmap_ax.get_ylim()\n",
    "x=x0\n",
    "for serotype in serotype_order:\n",
    "    length = xf*serotype_proportion[serotype]*0.95\n",
    "    heatmap_ax.plot((x, x+length), (yf-1,yf-1), c='darkgray', lw=3)#, transform=heatmap_ax.transAxes)\n",
    "    heatmap_ax.text((x+x+length)/2., -1, serotype, ha='center', va='bottom', fontsize=14)#transform=heatmap_ax.transAxes)\n",
    "    x += length*1.1\n",
    "heatmap_ax.set_ylim(y0,yf-1)\n",
    "heatmap_ax.set_ylabel('')\n",
    "heatmap_ax.set_xlabel('')\n",
    "\n",
    "tree_ax.text(0, 1, 'A', va='top', fontdict={'fontsize': 18, 'fontweight': 'bold'}, transform=tree_ax.transAxes)\n",
    "heatmap_ax.text(-0.05, 1.03, 'B', va='top', fontdict={'fontsize': 18, 'fontweight': 'bold'}, transform=heatmap_ax.transAxes)\n",
    "plt.savefig('./png/titer_tree_heatmap.png', dpi=300, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def split_serotypes(df, serotype_tip_order=serotype_tip_order):\n",
    "#     '''Return dictionary of {virus_rows: {sera_column: df}} for plotting heatmap small multiples'''\n",
    "#     columns = {}\n",
    "#     rows = {}\n",
    "#     for serotype in [1,2,3,4]:\n",
    "#         columns[serotype] = [s for s in serotype_tip_order[serotype] if s in df.columns.values]\n",
    "#         extra_sera = [ s for s in df.columns.values if s.startswith('DENV%d'%serotype) and s not in columns[serotype] ]\n",
    "#         columns[serotype] += extra_sera\n",
    "        \n",
    "#         rows[serotype] = [v for v in serotype_tip_order[serotype] if v in df.index.values]\n",
    "#         extra_viruses = [v for v in df.index.values if v.startswith('DENV%d'%serotype) and v not in rows[serotype]]\n",
    "# #         rows[serotype] += extra_viruses\n",
    "    \n",
    "# #     columns = { s : [ i for i in df.columns.values if i.startswith('DENV%d'%s) ] for s in [1,2,3,4] }\n",
    "# #     rows = { s : [i for i in df.index.values if i.startswith('DENV%d'%s) ] for s in [1,2,3,4] }\n",
    "\n",
    "#     split_serotypes = {1: {}, 2: {}, 3: {}, 4: {}} # { serotype_1_viruses: {vs_serotype_1_sera: partial_df, vs_2: ...}, ...}\n",
    "    \n",
    "#     for serum_serotype in [1,2,3,4]:\n",
    "#         for virus_serotype in [1,2,3,4]:\n",
    "#             partial_design_matrix = df.loc[rows[virus_serotype], columns[serum_serotype]]\n",
    "#             split_serotypes[serum_serotype][virus_serotype] = partial_design_matrix\n",
    "#     return split_serotypes\n",
    "\n",
    "# def plot_small_multiples(titer_df, cmap=colors['cmap']):\n",
    "#     data = split_serotypes(titer_df)\n",
    "#     sns.set(style='dark', font_scale=1.2)\n",
    "\n",
    "#     ## Set subplot size ratios\n",
    "#     width_ratios = [ len(data[i][1].columns.values) for i in [1,2,3,4] ]\n",
    "#     max_w = max(width_ratios)\n",
    "#     width_ratios = [ i / float(max_w) for i in width_ratios ]\n",
    "#     height_ratios = [ len(data[1][i].index.values) for i in [1,2,3,4]]\n",
    "#     max_h = max(height_ratios)\n",
    "#     height_ratios = [ i / float(max_h) for i in height_ratios ]    \n",
    "\n",
    "#     fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(7.5/2,7.5/2), facecolor='white',\n",
    "#                              gridspec_kw={'wspace': 0.1, 'hspace':0.1,\n",
    "#                             'width_ratios': width_ratios, 'height_ratios': height_ratios})\n",
    "\n",
    "#     ## Set colors\n",
    "#     cmap = plt.get_cmap(cmap)\n",
    "\n",
    "#     all_values = []\n",
    "#     for (virus, serum) in product([1,2,3,4], repeat=2):\n",
    "#         all_values += [i for i in data[virus][serum].values.flatten() if not np.isnan(i) ]\n",
    "#     vmin_val = float(min(all_values)) # Anchor color map for entire figure at global min/max values\n",
    "#     vmax_val = float(max(all_values))\n",
    "#     cbar_axis = fig.add_axes([1.0, 0.15, .03, 0.7])\n",
    "\n",
    "#     for virus, row in enumerate(axes, start=1):\n",
    "#         for serum, axis in enumerate(row, start=1):\n",
    "#             if virus == 1 and serum == 4:\n",
    "#                 cbar_bool = True\n",
    "#             else:\n",
    "#                 cbar_bool = False\n",
    "\n",
    "#             plot = sns.heatmap(data[serum][virus], ax=axis, \n",
    "#                         vmin=vmin_val, vmax=vmax_val, cmap=cmap, xticklabels = False, yticklabels=False, \n",
    "#                        cbar = cbar_bool, cbar_ax=cbar_axis,\n",
    "#                               cbar_kws={'label': r'Mean $log_2$(normalized titer)'})\n",
    "#             axis.set_xlabel('')\n",
    "#             axis.set_ylabel('')\n",
    "#             if virus == 1:\n",
    "#                 plot.set_title('DENV%d'%serum, fontsize=12)\n",
    "                \n",
    "#             if serum == 1:\n",
    "#                 plot.set_ylabel('DENV%d'%virus, fontsize=12)\n",
    "#                 if virus == 2:\n",
    "#                     plt.text(-.75, 0, 'Virus',  transform=axis.transAxes, fontsize=14, ha='center', va='center', rotation='vertical')\n",
    "#     plt.suptitle(' Sera', va='bottom', ha='center', fontsize=14)\n",
    "#     return fig"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fig = plot_small_multiples(titers)\n",
    "# tree_ax = fig.add_axes([1.8,0.05,0.03,0.9])\n",
    "\n",
    "# # plt.savefig('./png/titer_heatmap.png', dpi=300, bbox_inches='tight')\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}