simulations/single_lineage/simulation_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import os\n",
    "import sys\n",
    "from tqdm import tqdm\n",
    "import psix\n",
    "from matplotlib.gridspec import GridSpec\n",
    "\n",
    "cm = 1/2.54\n",
    "\n",
    "plt.rcParams[\"axes.edgecolor\"] = \"black\"\n",
    "plt.rcParams[\"axes.linewidth\"] = 1\n",
    "plt.rcParams[\"axes.facecolor\"] = 'white'\n",
    "\n",
    "import matplotlib as mpl\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "mpl.rcParams[\"mathtext.fontset\"] = \"stix\"\n",
    "mpl.rcParams['pdf.fonttype'] = 42"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For guidance, Nature’s standard figure sizes are 89 mm (single column) and 183 mm (double column) and the full depth of the page is 247 mm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "meta = pd.read_csv('sim_output/meta.tab.gz', sep='\\t')\n",
    "meta.index=['cell_'+str(i+1) for i in range(1000)]\n",
    "\n",
    "pop_1 = meta.loc[meta['pop'] == '3_1'].index\n",
    "pop_2 = meta.loc[meta['pop'] == '3_2'].index\n",
    "labels = [pop_1, pop_2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import kruskal\n",
    "from statsmodels.stats.multitest import multipletests\n",
    "\n",
    "\n",
    "def run_kw(exon_psi, labels):\n",
    "    kw_input = []\n",
    "    for l in labels:\n",
    "        kw_input.append(list(exon_psi[l]))\n",
    "        \n",
    "    if len(labels) == 2:\n",
    "        return kruskal(kw_input[0], kw_input[1], nan_policy='omit')\n",
    "    if len(labels) == 3:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], nan_policy='omit')\n",
    "    if len(labels) == 4:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], kw_input[3], nan_policy='omit')\n",
    "    if len(labels) == 5:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], kw_input[3], kw_input[4], nan_policy='omit')\n",
    "    if len(labels) == 6:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], kw_input[3], kw_input[4], kw_input[5], nan_policy='omit')\n",
    "    if len(labels) == 7:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], kw_input[3], kw_input[4], kw_input[5], kw_input[6], nan_policy='omit')\n",
    "    if len(labels) == 8:\n",
    "        return kruskal(kw_input[0], kw_input[1], kw_input[2], kw_input[3], kw_input[4], kw_input[5], kw_input[6], kw_input[7], nan_policy='omit')\n",
    "    \n",
    "def run_kw_dset(psi_table, labels, exon_list):\n",
    "    kw_output = pd.DataFrame()\n",
    "    kw_score = []\n",
    "    pvals = []\n",
    "    for exon in tqdm(exon_list, position=0, leave=True):\n",
    "        score, p = run_kw(psi_table.loc[exon], labels)\n",
    "        kw_score.append(score)\n",
    "        pvals.append(p)\n",
    "    kw_output['KW_score'] = kw_score\n",
    "    kw_output['pvals'] = pvals\n",
    "    kw_output['qvals'] = multipletests(pvals, method='fdr_bh')[1]\n",
    "    kw_output.index = exon_list\n",
    "    return kw_output\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting splice junctions....\n",
      "Obtaining PSI tables...\n",
      "Reading TPM and transforming to mRNA counts...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:25<00:00, 39.20it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully processed RNA-seq data\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 2591.19it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:35<00:00, 98.91it/s]  "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 25/25 [08:24<00:00, 20.16s/it]  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3486/3486 [00:32<00:00, 108.03it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_single_lineage_1 = psix.Psix()\n",
    "psix_single_lineage_1.junctions2psi(\n",
    "        'processed_tables/splice_junction_counts_0.1.tab.gz',\n",
    "        '',\n",
    "        'processed_tables/tpm_0.1.tab.gz',\n",
    "        save_files_in='psix_output/psix_object_0.1/')\n",
    "\n",
    "psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "                                n_random_exons=2000, n_neighbors=100)\n",
    "\n",
    "psix_results_1 = psix_single_lineage_1.psix_results\n",
    "\n",
    "kw_1 = run_kw_dset(psix_single_lineage_1.adata.uns['psi'][psix_results_1.index].T, \n",
    "                         labels, psix_results_1.index)\n",
    "\n",
    "gearyc_1 = pd.read_csv('gearyc/gearyc_0.1.tab.gz', sep='\\t', index_col=0)\n",
    "gearyc_1['qvals'] = multipletests(gearyc_1.pvals, method='fdr_bh')[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting splice junctions....\n",
      "Obtaining PSI tables...\n",
      "Reading TPM and transforming to mRNA counts...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:22<00:00, 44.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully processed RNA-seq data\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 7989.62it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3244 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3244/3244 [00:32<00:00, 101.15it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 25/25 [08:28<00:00, 20.35s/it]  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3244/3244 [00:29<00:00, 111.71it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_single_lineage_05 = psix.Psix()\n",
    "psix_single_lineage_05.junctions2psi(\n",
    "        'processed_tables/splice_junction_counts_0.05.tab.gz',\n",
    "        '',\n",
    "        'processed_tables/tpm_0.05.tab.gz',\n",
    "        save_files_in='psix_output/psix_object_0.05/')\n",
    "\n",
    "\n",
    "psix_single_lineage_05.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "                                n_random_exons=2000, n_neighbors=100)\n",
    "\n",
    "psix_results_05 = psix_single_lineage_05.psix_results\n",
    "\n",
    "kw_05 = run_kw_dset(psix_single_lineage_05.adata.uns['psi'][psix_results_05.index].T, \n",
    "                         labels, psix_results_05.index)\n",
    "\n",
    "gearyc_05 = pd.read_csv('gearyc/gearyc_0.05.tab.gz', sep='\\t', index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting splice junctions....\n",
      "Obtaining PSI tables...\n",
      "Reading TPM and transforming to mRNA counts...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:17<00:00, 56.87it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully processed RNA-seq data\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 8418.61it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 2097 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 2097/2097 [00:17<00:00, 121.66it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 25/25 [07:47<00:00, 18.68s/it]  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 2097/2097 [00:14<00:00, 144.88it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_single_lineage_01 = psix.Psix()\n",
    "psix_single_lineage_01.junctions2psi(\n",
    "        'processed_tables/splice_junction_counts_0.01.tab.gz',\n",
    "        '',\n",
    "        'processed_tables/tpm_0.01.tab.gz',\n",
    "        save_files_in='psix_output/psix_object_0.01/')\n",
    "\n",
    "\n",
    "psix_single_lineage_01.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "                                n_random_exons=2000, n_neighbors=100)\n",
    "\n",
    "psix_results_01 = psix_single_lineage_01.psix_results\n",
    "\n",
    "kw_01 = run_kw_dset(psix_single_lineage_01.adata.uns['psi'][psix_results_01.index].T, \n",
    "                         labels, psix_results_01.index)\n",
    "\n",
    "gearyc_01 = pd.read_csv('gearyc/gearyc_0.01.tab.gz', sep='\\t', index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.special import logit, expit\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import auc\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "platonic_psi = pd.read_csv('sim_output/psi_platonic.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "platonic_psi.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "\n",
    "platonic_psi = expit(platonic_psi)\n",
    "\n",
    "\n",
    "def get_random_samples(psix_results, kw, gearyc):\n",
    "    \n",
    "    pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results.index\n",
    "    neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results.index\n",
    "\n",
    "    psix_auc = []\n",
    "    kw_auc = []\n",
    "    gearyc_auc = []\n",
    "\n",
    "    for i in tqdm(range(200)):\n",
    "\n",
    "        pos_sample = np.random.choice(pos, int(len(pos)*0.2), replace=False)\n",
    "        neg_sample = np.random.choice(neg, int(len(neg)*0.2), replace=False)\n",
    "\n",
    "        labels = [1]*len(pos_sample) + [0]*len(neg_sample)\n",
    "        order = list(pos_sample) + list(neg_sample)\n",
    "\n",
    "\n",
    "        y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "\n",
    "        pre, rec, thres = precision_recall_curve(y_, psix_results.loc[order].psix_score)\n",
    "        pre_kw, rec_kw, thres_kw = precision_recall_curve(y_, kw.loc[order].KW_score)\n",
    "        pre_c, rec_c, thres_c = precision_recall_curve(y_, gearyc.loc[order].C_score)\n",
    "\n",
    "        psix_auc.append(auc(rec, pre))\n",
    "        kw_auc.append(auc(rec_kw, pre_kw))\n",
    "        gearyc_auc.append(auc(rec_c, pre_c))\n",
    "        \n",
    "    return psix_auc, kw_auc, gearyc_auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:06<00:00, 32.88it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.33it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.79it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_1_auc, kw_1_auc, gearyc_1_auc = get_random_samples(psix_results_1, kw_1, gearyc_1)\n",
    "psix_05_auc, kw_05_auc, gearyc_05_auc = get_random_samples(psix_results_05, kw_05, gearyc_05)\n",
    "psix_01_auc, kw_01_auc, gearyc_01_auc = get_random_samples(psix_results_01, kw_01, gearyc_01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_1_auc, psix_05_auc, psix_01_auc], positions=[1,5,9], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_1_auc, kw_05_auc, kw_01_auc], positions=[1.75,5.75,9.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_1_auc, gearyc_05_auc, gearyc_01_auc], positions=[2.5,6.5,10.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75, 9.75])\n",
    "ax.set_xticklabels(['0.1', '0.05', '0.01'], size=8)\n",
    "ax.set_xlabel('Capture efficiency', size=8)\n",
    "\n",
    "ax.set_yticks([0.85, 0.9, 0.95])\n",
    "ax.set_yticklabels(['0.85', '0.90', '0.95'], size=8)\n",
    "ax.set_ylabel('Precision-recall AUC', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([-0.5, 12])\n",
    "\n",
    "plt.savefig('/mnt/lareaulab/cfbuenabadn/plots_pdf/single_lineage_auc.pdf', bbox_inches='tight', res=20000, dpi =2000)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "\n",
    "ax = fig.add_subplot(111)\n",
    "ax.plot(sorted(np.arange(0+(1/len(neg)), 1+(1/len(neg)), step = 1/len(neg))),\n",
    "            sorted(np.arange(0+(1/len(neg)), 1+(1/len(neg)), step = 1/len(neg))), c='red',\n",
    "        linestyle='--')\n",
    "\n",
    "ax.plot(sorted(np.arange(0+(1/len(neg)), 1+(1/len(neg)), step = 1/len(neg))),\n",
    "            sorted(psix_results_1.loc[neg].pvals), label='Psix', linewidth=2, c='royalblue')\n",
    "\n",
    "ax.plot(sorted(np.arange(0+(1/len(neg)), 1+(1/len(neg)), step = 1/len(neg))),\n",
    "            sorted(kw_1.loc[neg].pvals), label='Kruskal-Wallis', linewidth=2, c='darkorange')\n",
    "\n",
    "ax.plot(sorted(np.arange(0+(1/len(neg)), 1+(1/len(neg)), step = 1/len(neg))),\n",
    "            sorted(gearyc_1.loc[neg].pvals), label=\"Geary's C\", linewidth=2, c='forestgreen')\n",
    "\n",
    "\n",
    "ax.set_xticks([0, 0.25, 0.5, 0.75, 1])\n",
    "ax.set_xticklabels(['0.0', '0.25', '0.5', '0.75', '1.0'], size=8)\n",
    "ax.set_xlabel('Quantiles', size=8)\n",
    "\n",
    "ax.set_yticks([0, 0.25, 0.5, 0.75, 1])\n",
    "ax.set_yticklabels(['0.0', '0.25', '0.5', '0.75', '1.0'], size=8)\n",
    "ax.set_ylabel('p-values (null)', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.legend(frameon=False, fontsize=8)\n",
    "\n",
    "plt.savefig('/mnt/lareaulab/cfbuenabadn/plots_pdf/single_lineage_pvals.pdf', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x75 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rd = pd.read_csv('processed_tables/pc2_rd.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2.5/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "sc = ax.scatter(-rd.PC_1, -rd.PC_2, c=meta.lineage_depth, cmap='inferno', vmin=0, s=20, linewidth=0)\n",
    "cb = plt.colorbar(sc, shrink = 0.8, aspect=10, orientation=\"horizontal\")\n",
    "cb.set_label(label='Pseudotime',size=8)\n",
    "cb.outline.set_visible(False)\n",
    "cb.ax.tick_params(labelsize=0, length=0)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "plt.savefig('../plots/single_lineage.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "underlying_psi = pd.read_csv('sim_output/psi_underlying.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "underlying_psi.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "underlying_psi = expit(underlying_psi)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x75 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rd = pd.read_csv('processed_tables/pc2_rd.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "# pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "# neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2.5/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "sc = ax.scatter(-rd.PC_1, -rd.PC_2, c=psix_single_lineage_1.adata.uns['psi']['gene316'], vmin=0, vmax=1, s=10, linewidth=0)\n",
    "cb = plt.colorbar(sc, shrink = 0.8, aspect=10, orientation=\"horizontal\")\n",
    "cb.set_label(label='$\\Psi$',size=8)\n",
    "cb.outline.set_visible(False)\n",
    "cb.ax.tick_params(labelsize=8, length=2)\n",
    "\n",
    "\n",
    "# plt.title('Positives')\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "plt.savefig('../plots/positive_pca.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x75 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rd = pd.read_csv('processed_tables/pc2_rd.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2.5/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "sc = ax.scatter(-rd.PC_1, -rd.PC_2, c=psix_single_lineage_1.adata.uns['psi']['gene4452'], vmin=0, vmax=1, s=10, linewidth=0)\n",
    "cb = plt.colorbar(sc, shrink = 0.8, aspect=10, orientation=\"horizontal\")\n",
    "cb.set_label(label='$\\Psi$',size=8)\n",
    "cb.outline.set_visible(False)\n",
    "cb.ax.tick_params(labelsize=8, length=2)\n",
    "\n",
    "\n",
    "# plt.title('Positives')\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "plt.savefig('../plots/negative_pca.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x60 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# rd = pd.read_csv('single_lineage/processed_tables/pc2_rd.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "# pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "# neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter(meta.lineage_depth, underlying_psi.loc['gene316'], c='steelblue', s=5, linewidth=0)\n",
    "ax.plot(meta.sort_values('lineage_depth').lineage_depth, \n",
    "        platonic_psi.loc['gene316', meta.sort_values('lineage_depth').index], c='darkred')\n",
    "\n",
    "\n",
    "ax.set_xlabel('Lineage', size=8)\n",
    "ax.set_ylabel('exon $\\Psi$', size=8)\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.set_ylim(-0.05, 1.05)\n",
    "\n",
    "plt.savefig('../plots/positive_psi.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x60 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# rd = pd.read_csv('single_lineage/processed_tables/pc2_rd.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "# pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "# neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter(meta.lineage_depth, underlying_psi.loc['gene4452'], c='steelblue', s=5, linewidth=0)\n",
    "ax.plot(meta.sort_values('lineage_depth').lineage_depth, \n",
    "        platonic_psi.loc['gene4452', meta.sort_values('lineage_depth').index], c='darkred')\n",
    "\n",
    "\n",
    "ax.set_xlabel('Lineage', size=8)\n",
    "ax.set_ylabel('exon $\\Psi$', size=8)\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.set_ylim(-0.05, 1.05)\n",
    "\n",
    "\n",
    "plt.savefig('../plots/negative_psi.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "gene_expression = pd.read_csv('sim_output/gene_counts.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "gene_expression.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "\n",
    "# gene_expression = expit(ungene_expression)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x60 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "# neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter(meta.lineage_depth, np.log1p(gene_expression.loc['gene316']), c='steelblue', s=5, linewidth=0)\n",
    "# ax.plot(meta.sort_values('lineage_depth').lineage_depth, \n",
    "#        gene_expression.loc['gene316', meta.sort_values('lineage_depth').index], c='darkred')\n",
    "\n",
    "\n",
    "ax.set_xlabel('Lineage', size=8)\n",
    "ax.set_ylabel('gene expression', size=8)\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "# ax.set_ylim(-0.05, 1.05)\n",
    "\n",
    "plt.savefig('../plots/positive_gene.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 75x60 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.index\n",
    "# neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.index\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(2.5/2.4, 2/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter(meta.lineage_depth, np.log1p(gene_expression.loc['gene4452']), c='steelblue', s=5, linewidth=0)\n",
    "# ax.plot(meta.sort_values('lineage_depth').lineage_depth, \n",
    "#        gene_expression.loc['gene316', meta.sort_values('lineage_depth').index], c='darkred')\n",
    "\n",
    "\n",
    "ax.set_xlabel('Lineage', size=8)\n",
    "ax.set_ylabel('gene expression', size=8)\n",
    "# ax.tick_params(labelsize=8, length=1)\n",
    "\n",
    "plt.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,\n",
    "    left = False,\n",
    "    right = False, # ticks along the bottom edge are off\n",
    "    top=False,         # ticks along the top edge are off\n",
    "    labelbottom=False,\n",
    "labelleft=False)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "# ax.set_ylim(-0.05, 1.05)\n",
    "\n",
    "plt.savefig('../plots/negative_gene.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os \n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=5)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_5.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=10)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_10.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=20)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_20.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=30)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_30.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=50)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_50.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=100)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_100.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=150)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_150.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=200)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_200.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=250)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_250.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=300)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_300.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=400)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_400.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz', n_jobs=25,\n",
    "#                                 n_random_exons=2000, n_neighbors=500)\n",
    "# psix_single_lineage_1.psix_results.to_csv('k_sensitivity/k_500.tab.gz', sep='\\t', index=True, header=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "psix_5 = pd.read_csv('k_sensitivity/k_5.tab.gz', sep='\\t', index_col=0)\n",
    "psix_10 = pd.read_csv('k_sensitivity/k_10.tab.gz', sep='\\t', index_col=0)\n",
    "psix_20 = pd.read_csv('k_sensitivity/k_20.tab.gz', sep='\\t', index_col=0)\n",
    "psix_30 = pd.read_csv('k_sensitivity/k_30.tab.gz', sep='\\t', index_col=0)\n",
    "psix_50 = pd.read_csv('k_sensitivity/k_50.tab.gz', sep='\\t', index_col=0)\n",
    "psix_100 = pd.read_csv('k_sensitivity/k_100.tab.gz', sep='\\t', index_col=0)\n",
    "psix_150 = pd.read_csv('k_sensitivity/k_150.tab.gz', sep='\\t', index_col=0)\n",
    "psix_200 = pd.read_csv('k_sensitivity/k_200.tab.gz', sep='\\t', index_col=0)\n",
    "psix_300 = pd.read_csv('k_sensitivity/k_300.tab.gz', sep='\\t', index_col=0)\n",
    "psix_400 = pd.read_csv('k_sensitivity/k_400.tab.gz', sep='\\t', index_col=0)\n",
    "psix_500 = pd.read_csv('k_sensitivity/k_500.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "pos = []\n",
    "neg = []\n",
    "for i in range(5000):\n",
    "    if ('gene'+str(i+1)) in psix_5.index:\n",
    "\n",
    "        if (platonic_psi.loc['gene'+str(i+1)].max() - platonic_psi.loc['gene'+str(i+1)].min()) >= 0.2:\n",
    "            pos.append('gene'+str(i+1))\n",
    "        else:\n",
    "        #elif ds[i] == 0:\n",
    "            neg.append('gene'+str(i+1))\n",
    "\n",
    "    \n",
    "labels = [1]*len(pos) + [0]*len(neg)\n",
    "order = list(pos) + list(neg)\n",
    "\n",
    "    \n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "boxplot_list = []\n",
    "for x in [psix_5, psix_10, psix_20, psix_30, psix_50, psix_100, psix_150, psix_200, psix_300, psix_400, psix_500]:\n",
    "    \n",
    "    box = []\n",
    "    for i in range(200):\n",
    "        \n",
    "        pos_sample = np.random.choice(pos, int(len(pos)*0.2), replace=False)\n",
    "        neg_sample = np.random.choice(neg, int(len(neg)*0.2), replace=False)\n",
    "\n",
    "        labels = [1]*len(pos_sample) + [0]*len(neg_sample)\n",
    "        order = list(pos_sample) + list(neg_sample)\n",
    "\n",
    "\n",
    "        y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "        \n",
    "        \n",
    "        pre, rec, thres = precision_recall_curve(y_, x.loc[order].psix_score)\n",
    "    \n",
    "        box.append(auc(rec, pre))\n",
    "\n",
    "    boxplot_list.append(box)\n",
    "\n",
    "    \n",
    "\n",
    "names = pd.Index(['gene' + str(i) for i in range(1, 5001)]) & psix_5.index\n",
    "\n",
    "pos = []\n",
    "neg = []\n",
    "order = []\n",
    "y = []\n",
    "for i in range(5000):\n",
    "    if ('gene'+str(i+1)) in psix_5.index:\n",
    "\n",
    "        if (platonic_psi.loc['gene'+str(i+1)].max() - platonic_psi.loc['gene'+str(i+1)].min()) >= 0.2:\n",
    "            y.append(1)\n",
    "            pos.append('gene'+str(i+1))\n",
    "            order.append('gene'+str(i+1))\n",
    "        else:\n",
    "#         elif ds[i] == 0:\n",
    "            neg.append('gene'+str(i+1))\n",
    "            y.append(0)\n",
    "            order.append('gene'+str(i+1))\n",
    "    \n",
    "    \n",
    "recall_list = []\n",
    "for x in [psix_5, psix_10, psix_20, psix_30, psix_50, psix_100, psix_150, psix_200, psix_300, psix_400, psix_500]:\n",
    "    pre, rec, thres = precision_recall_curve(y, x.loc[order].psix_score)\n",
    "    \n",
    "    recall_list.append(auc(rec, pre))\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.fill_between(np.log10(np.array([0.5, 1, 2, 3, 5, 10, 15, 20, 30, 40, 50])), [np.quantile(x, 0.05) for x in boxplot_list], \n",
    "                 [np.quantile(x, 0.95) for x in boxplot_list], color='skyblue')\n",
    "\n",
    "ax.plot([1, 1], [np.quantile(boxplot_list[0], 0.05), recall_list[5]], linestyle='--', c='gray')\n",
    "\n",
    "ax.scatter(np.log10(np.array([0.5, 1, 2, 3, 5, 10, 15, 20, 30, 40, 50])), recall_list, c='navy', zorder=20, s=20)\n",
    "\n",
    "ax.plot(np.log10(np.array([0.5, 1, 2, 3, 5, 10, 15, 20, 30, 40, 50])), recall_list, c='navy', zorder=20)\n",
    "\n",
    "ax.set_xticks(np.log10(np.array([0.5, 1, 2, 5, 10, 20, 50])))\n",
    "\n",
    "ax.set_xticklabels(['0.5%', '1%', '2%', '5%', '10%', '20%', '50%'], size=8)\n",
    "\n",
    "ax.set_xlabel('Neighborhood size - population percent', size=8)\n",
    "\n",
    "ax.set_yticks([0.7, 0.8, 0.9])\n",
    "ax.set_yticklabels(['0.7', '0.8', '0.9'], size=8)\n",
    "ax.set_ylabel('Precision-recall AUC', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "# ax.set_xlim([-0.5, 12])\n",
    "\n",
    "plt.savefig('../plots/single_lineage_kauc.png', bbox_inches='tight', res=20000, dpi =2000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f9718f05fd0>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8, 4))\n",
    "gs = GridSpec(1,7)\n",
    "gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(gs[0,0:4])\n",
    "ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_1_auc, psix_05_auc, psix_01_auc], positions=[1,7,13], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([psix_1_auc_, psix_05_auc_, psix_01_auc_], positions=[1.75,7.75,13.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='royalblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_1_auc, kw_05_auc, kw_01_auc], positions=[2.5,8.5,14.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='orange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_1_auc_, kw_05_auc_, kw_01_auc_], positions=[3.25,9.25,15.25], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_1_auc, gearyc_05_auc, gearyc_01_auc], positions=[4,10,16], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_1_auc_, gearyc_05_auc_, gearyc_01_auc_], positions=[4.75,10.75,16.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='forestgreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 8.75, 14.75])\n",
    "ax.set_xticklabels(['0.1', '0.05', '0.01'], size=18)\n",
    "ax.set_xlabel('Capture efficiency', size=18)\n",
    "\n",
    "ax.set_yticks([0.85, 0.9, 0.95])\n",
    "ax.set_yticklabels(['0.85', '0.90', '0.95'], size=18)\n",
    "ax.set_ylabel('AU Pre-Rec curve', size=18)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax_2.set_xlim([0, 0])\n",
    "\n",
    "ax_2.hist([-1, -1], color='skyblue', label='Psix')\n",
    "ax_2.hist([-1, -1], color='darkorange', label='Kruskal-Wallis')\n",
    "ax_2.hist([-1, -1], color='limegreen', label=\"Geary's C\")\n",
    "\n",
    "ax_2.spines[\"bottom\"].set_visible(False)\n",
    "ax_2.spines[\"top\"].set_visible(False)\n",
    "ax_2.spines[\"right\"].set_visible(False)\n",
    "ax_2.spines[\"left\"].set_visible(False)\n",
    "\n",
    "ax_2.tick_params(\n",
    "    axis='both',          # changes apply to the x-axis\n",
    "    which='both',      # both major and minor ticks are affected\n",
    "    bottom=False,      # ticks along the bottom edge are off\n",
    "    top=False,  \n",
    "    right=False,\n",
    "    left=False, # ticks along the top edge are off\n",
    "    labelleft=False,\n",
    "    labelbottom=False)\n",
    "\n",
    "ax_2.legend(frameon=False, fontsize=14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Psix performance in differentially expressed genes\n",
    "\n",
    "We run slingshot on the simulated gene trajectory to obtain a lineage pseudotime. Then we run tradeseq to get differentially expressed genes.\n",
    "\n",
    "This code is on ```run_slingshot.Rmd```. The output of tradeseq is on ```processed_tables/ATres.tab```."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.stats.multitest import multipletests\n",
    "\n",
    "ATres = pd.read_csv('processed_tables/ATres.tab.gz', sep='\\t', index_col=0)\n",
    "\n",
    "idx = [int(x.split('gene')[1]) for x in psix_results_1.index]\n",
    "ATres['FDR'] = multipletests(ATres.pvalue)[1]\n",
    "ATres = ATres.loc[idx]\n",
    "ATres['significant'] = (ATres.loc[idx].FDR < 0.05) & (ATres.loc[idx].meanLogFC >= 0.25)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>waldStat</th>\n",
       "      <th>df</th>\n",
       "      <th>pvalue</th>\n",
       "      <th>meanLogFC</th>\n",
       "      <th>FDR</th>\n",
       "      <th>significant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1128</th>\n",
       "      <td>0.535707</td>\n",
       "      <td>4</td>\n",
       "      <td>9.699331e-01</td>\n",
       "      <td>0.012144</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1880</th>\n",
       "      <td>27.276286</td>\n",
       "      <td>4</td>\n",
       "      <td>1.747897e-05</td>\n",
       "      <td>0.404205</td>\n",
       "      <td>3.295231e-02</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2332</th>\n",
       "      <td>15.913010</td>\n",
       "      <td>4</td>\n",
       "      <td>3.138140e-03</td>\n",
       "      <td>0.124870</td>\n",
       "      <td>9.883284e-01</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>705</th>\n",
       "      <td>537.670351</td>\n",
       "      <td>4</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.645671</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1878</th>\n",
       "      <td>69.716086</td>\n",
       "      <td>4</td>\n",
       "      <td>2.609024e-14</td>\n",
       "      <td>0.132263</td>\n",
       "      <td>7.498335e-11</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        waldStat  df        pvalue  meanLogFC           FDR  significant\n",
       "1128    0.535707   4  9.699331e-01   0.012144  1.000000e+00        False\n",
       "1880   27.276286   4  1.747897e-05   0.404205  3.295231e-02         True\n",
       "2332   15.913010   4  3.138140e-03   0.124870  9.883284e-01        False\n",
       "705   537.670351   4  0.000000e+00   0.645671  0.000000e+00         True\n",
       "1878   69.716086   4  2.609024e-14   0.132263  7.498335e-11        False"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ATres.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "DE = ATres.loc[ATres.significant].index\n",
    "nonDE = ATres.loc[~ATres.significant].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "psi_platonic = pd.read_csv('sim_output/psi_platonic.tab.gz', sep='\\t',\n",
    "                          names = range(1000))\n",
    "psi_platonic.index = range(1, 5001)\n",
    "psi_platonic = psi_platonic.loc[np.array(idx)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "nonDS = psi_platonic.loc[\n",
    "    (psi_platonic.var(axis=1) <= 10e-20) | ((psi_platonic.max(axis=1) - psi_platonic.min(axis=1)) < 0.2)].index\n",
    "DS = psi_platonic.loc[\n",
    "    (psi_platonic.var(axis=1) >= 10e-20) & ((psi_platonic.max(axis=1) - psi_platonic.min(axis=1)) >= 0.2)].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "DE_nonDS = DE.intersection(nonDS)\n",
    "DE_DS = DE.intersection(DS)\n",
    "\n",
    "nonDE_nonDS = nonDE.intersection(nonDS)\n",
    "nonDE_DS = nonDE.intersection(DS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.48129359543436906"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(DE_DS)/len(DE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.01589242053789731"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(psix_results_1.loc[['gene'+str(x) for x in DE_nonDS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.14792176039119803"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(kw_1.loc[['gene'+str(x) for x in DE_nonDS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6324110671936759"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(psix_results_1.loc[['gene'+str(x) for x in DE_DS]].qvals <= 0.01).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7852437417654808"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(kw_1.loc[['gene'+str(x) for x in DE_DS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7033985581874357"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(psix_results_1.loc[['gene'+str(x) for x in nonDE_DS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7456230690010298"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(kw_1.loc[['gene'+str(x) for x in nonDE_DS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.013859275053304905"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(psix_results_1.loc[['gene'+str(x) for x in nonDE_nonDS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04797441364605544"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(kw_1.loc[['gene'+str(x) for x in nonDE_nonDS]].qvals <= 0.05).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.special import logit, expit\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import auc\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "platonic_psi = pd.read_csv('sim_output/psi_platonic.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "platonic_psi.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "\n",
    "platonic_psi = expit(platonic_psi)\n",
    "\n",
    "\n",
    "def get_random_samples(psix_results, kw, gearyc):\n",
    "    \n",
    "    pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results.index\n",
    "    neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results.index\n",
    "\n",
    "    psix_auc = []\n",
    "    kw_auc = []\n",
    "    gearyc_auc = []\n",
    "\n",
    "    for i in tqdm(range(200)):\n",
    "\n",
    "        pos_sample = np.random.choice(pos, int(len(pos)*0.2), replace=False)\n",
    "        neg_sample = np.random.choice(neg, int(len(neg)*0.2), replace=False)\n",
    "\n",
    "        labels = [1]*len(pos_sample) + [0]*len(neg_sample)\n",
    "        order = list(pos_sample) + list(neg_sample)\n",
    "\n",
    "\n",
    "        y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "\n",
    "        pre, rec, thres = precision_recall_curve(y_, psix_results.loc[order].psix_score)\n",
    "        pre_kw, rec_kw, thres_kw = precision_recall_curve(y_, kw.loc[order].KW_score)\n",
    "        pre_c, rec_c, thres_c = precision_recall_curve(y_, gearyc.loc[order].C_score)\n",
    "\n",
    "        psix_auc.append(auc(rec, pre))\n",
    "        kw_auc.append(auc(rec_kw, pre_kw))\n",
    "        gearyc_auc.append(auc(rec_c, pre_c))\n",
    "        \n",
    "    return psix_auc, kw_auc, gearyc_auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:06<00:00, 32.29it/s]\n",
      "100%|██████████| 200/200 [00:05<00:00, 33.90it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.53it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_auc, kw_auc, gearyc_auc = get_random_samples(psix_results_1, kw_1, gearyc_1)\n",
    "\n",
    "psix_DE_auc, kw_DE_auc, gearyc_DE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in DE]])\n",
    "\n",
    "psix_nonDE_auc, kw_nonDE_auc, gearyc_nonDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in nonDE]])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import f_oneway, ranksums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RanksumsResult(statistic=-4.407759197952194, pvalue=1.0444557229165906e-05)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ranksums(psix_nonDE_auc, psix_DE_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9468263193958526\n",
      "0.8843380467136673\n",
      "0.8891371324999124\n"
     ]
    }
   ],
   "source": [
    "pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results_1.loc[['gene'+str(x) for x in DE]].index\n",
    "neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results_1.loc[['gene'+str(x) for x in DE]].index\n",
    "\n",
    "    \n",
    "labels = [1]*len(pos) + [0]*len(neg)\n",
    "order = list(pos) + list(neg)\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "\n",
    "pre, rec, thres = precision_recall_curve(y_, psix_results_1.loc[order].psix_score)\n",
    "pre_kw, rec_kw, thres_kw = precision_recall_curve(y_, kw_1.loc[order].KW_score)\n",
    "pre_c, rec_c, thres_c = precision_recall_curve(y_, gearyc_1.loc[order].C_score)\n",
    "\n",
    "\n",
    "print(auc(rec, pre))\n",
    "print(auc(rec_kw, pre_kw))\n",
    "print(auc(rec_c, pre_c))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9425429215687996"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(psix_nonDE_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9472976054035357"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(psix_DE_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_auc, psix_nonDE_auc, psix_DE_auc], positions=[1,5, 9], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_auc, kw_nonDE_auc, kw_DE_auc], positions=[1.75,5.75, 9.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_auc, gearyc_nonDE_auc, gearyc_DE_auc], positions=[2.5,6.5, 10.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75, 9.75])\n",
    "ax.set_xticklabels(['all', 'nonDE', 'DE'], size=8)\n",
    "ax.set_xlabel('Gene expression', size=8)\n",
    "\n",
    "ax.set_yticks([0.85, 0.9, 0.95])\n",
    "ax.set_yticklabels(['0.85', '0.90', '0.95'], size=8)\n",
    "ax.set_ylabel('Precision-recall AUC', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0, 12])\n",
    "\n",
    "plt.savefig('/mnt/lareaulab/cfbuenabadn/plots_pdf/single_lineage_auc_DEvnonDE.pdf', bbox_inches='tight', res=20000, dpi =2000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 8.0)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL8AAACTCAYAAADfhMhLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAWD0lEQVR4nO2deZRU1bWHvx/dNN2AjSIzyuyAIyBKnNLOUYlRE31GjQMG40s0EF8072Wp0TgkDkkMLB8SNUQFTJSoOKDRKDOoDKKIKIoDPplRZOymu2G/P+7p7uqiqrqqqbnOt1atume45+7q3nVq33P32VtmhsdTiLTItAAeT6bwyu8pWLzyewoWr/yegsUrv6dg8crvKVi88icZSeWSpkiaLulNSYMldZF00x6MOTtG23mS2sc5zqOSesVonx6j7UpJeaUvefVhsoTLgWfM7CTgBGCZma0xs7tSdL3zgLiUfw+5kjzTl7z6MFnCduBYSR3MrNbMtkjqJWkCgPs1+IukdyUNk/SMpMWSjnTts917L0mPhg7s+k+XtEDSGZJ6AGcCEyXdKKmjpOclTZM0xp3TW9Jbkp4H+oQLK+knTqb7Q+p+LWmGO2+gpGOAAcDrki4LlyMlf8V0YGb+lcQX0BL4DfAe8BrQBegFTHDtHwGdgW7AGqAUOA4Y7dpnu/dewKNhda3dezvgVXf8KNDPHf8RONYd3wMcC4wBjgeKgfeBXiGyFgNvuffjgOlh1+kHTHTH04HiaHLk4qt4j789nkaYWQ1wO3C7pIuBXwBjQ7qsN7O1AJI+MbMqSauAfcKGUoThvyNppGvrFKG9P3C3JAPaAvMIZvtFZlYraXFY/w7ACte2MKT+MkmXAruASP4vTcmRE3jlTzKSegKr3JdgHbublhbluE7ZS9374RGG/zVQAbQC5ri6GqDIHS8j+IVZ6GQpBk4DjpQ0L8KYG4CekoqAgSH1P3PlvsDDYdepjSJHzuGVP/kMAJ6SVEmgMMNI7O88xdn9b0VoexGYSTCjf+PqXgHGSJoE/A54SFI7gll7OPAH4AlgrXvV42b8vwFzgRkhTfPcdWaGygVMlvRIFDlyDjnbzeMpOPxqj6dg8crvKVi88nsKFq/8noLFK7+nYMnJpU5J5lepPHES6WEh4Gd+TwHjld9TsHjl9xQsOWnze5KPtLtpnO/3VV75PUCDokvKe6WvI6rZ4zZBHBBWd2CsbXAeTy4Ry+b/X2BTWN03wIOpE8fjSR+xzJ4yM1sXWmFm6ySVRjshF4hk20L+27ee3Yk189dK6hhaIakzwWaGnCVkC1/49kNPgRFr5v8V8KykqcBqoDvB7p3r0iGYx5NqYm5mkdSSYBN0V4IvwBtue15GSYZ7Q6pXNXLVvMrD1Z6o7g1RZ35Jl4dVdSbYzf9BkoTKa0IVKA8VKi+IZfaEf2PaA792EQd+29TALg7MYOBtMxsZUn86cAdQCfzUzD6UdBtwPrAReN7M/pTYx/A0h8WL1zJ79hdhtUczZsz8+lLPnu0YOvTA9AqWJhLewytpugXRyGL1GUSg2FdLehAYZ2bzXdssgkBL5cCfzewip/yzzey1OGXIerMnU9dKhJNOe5z1NbvYp8te9XWVlZWUlZUFBYM5k5ZgdmuGJEwKiZs9EUcJ7gFaxtH1W8C/3fFrBPcN9dOJmW0DtknqG3LOPZI2AjeY2TuJyOVpPkP+4wgOPK5HxDYzY86kJWmWKH3EsvnH0ziuTCmwP3B3HOPuDXzqjjcBh4aN3ZkgSFN/VzXazG5zT5THASfGJX2cfPnlZsaNWxRWW8HttzdE6ygtLWbEiCGUlhaWx0fVlh3NassHYv2nbw4rV9VFGouDTQRmDe49NLbLr4B/ACtwAY/M7Gv3/nG0VRJJPwF+Euf1GzF+/LuMe+ZD+h6zX33dycNvYMaXW+vLi6csY8iQ7lRU9GrOJXKSH18xgBHX/4vls1dw1o0nUFbe8PxyxrgFzJ34Ltf87OgMSphaoiq/ma0ILUs6TtKNwBAza2pmfgO4BniKIGLYoyHjvgGc7Gb569zY5Wa2WVKHaDKZ2UPAQ65/wgZ0n0HdOP2nQ6K2r3pndaJD5jxHH92NowZ1ZeFb/8f8k/66W3vP3ntzztn9MiBZeohl9hQDpxKswnQGDgTON7MbmhrUzN6WVOVubt8BvpB0k5nd5eLUnwZ8RfAFAbhP0mEET5z/Z48+UQrI1TX7pvj5L/5FdZdyrn/uNErKGt/KmRkfzPiMc897ktqaWzIkYWqJZfZsAJ4EbjezlZJeNrOP4h04dHnTcZerv6vuOKTvNaSQPn32YdGf3qR0r1Ycc8FhlLYtqW/7ZvUW5oxfxMqPNtCt214Rz89Xd9+a2l30G7L/booPwWftX9GbnbW7MiBZeoil/OcQzPpjJS0A2qRHpORz4YWHsnz519x88zReHv1GxD4vvngxBxyw7x5fa9maZdTuDHN/2hveX/k+4JSqa/+ovyae9BHL5p8FzAKQNBAolvQysM7MrkiTfEnh4YcX8sAji7j0vjM55OQ+tGjRoHjVlTXMf3YpP7z0GebMHMYRR3Ru9nVe/+B1hj4wlLblbRvVt7ykJRUPVACwbes2xv1oHBcPubjZ1/Ekh7jW9cxsEbAIgg0tKZUoBXz9dSX9T+7DYaf23a2tpKwlx19yJMtnfMbGjZV7dJ3NlZvpsH8HOp8R/Qu0duZaNldt3qPrJJP5/1zC8jkrIrblk4kXiYQXtROx+z3Zzf33nc6sWY3dG0aOHMmoUaPqyzf88JB0i5U2CuaJTs2OWio3V0VtT9aNXfWOamyXoRaRN4TXVGXcKbaegQO7MnBg10Z1I0e+xYgR0ZeE84lYS51XRWszs3GpESc1HHRQB35/zxzenbKsvm779kpaty6rL5eWFtO9e3mk0+Pm+H7H07u0Nx8//TEdju/AXiGrR9vXb2fD7A10atmJMw7J3Rxu+URUxzZJUW9qzeyxlEkUB5lybGvqHDNj0oJJXPX4VWyr2haxz8OXP8yw44dR1KIoYnumSeTvkiPhTprl2DYrBYLkNa998BrD/zGcrqd3pXy/3X9Ftq3Zxo0v3kjrktZcMuSSDEiYXHL9+Ucs5b+FwLEt/JtjQFSTKJ/Ytm4dhPxT2wLb1ja4N6moiNYdOtSXt1ZtpbxjeUTFB2jTpQ1turZhS9WWlMnsiZ9Y6/zD0ilItvHx5MlMuehCSlo3PP28tRU8dnDv+vKO7TVc8Opr7F9RkQkRc5JsMpWaXO2RdBpwI7AfQYa/jWb27VQLlmkq16+n/9ElfOf87eEt9UfP/b2cyg0b0itYjpNNplI8S52/I3BEex44lzC/HE9jKrdV8s3n0bNzVm2JvtzqSS/xKP82525sQBWQvw7ee8jgXoMZ3Gkw1V9UN6qfOWsm3z4x+LFsUd6CEw9I6l4dTzOJR/kfc1HaRhMkHn4xtSLlLvu3359Xfv7KbvW6Xkx7eloGJPLEIp74/M+YWZWZPWtmQ4BRTZ6RB5T37MmnH8DHEbawmsHShfDl8hr22n//9AvnSQpxKX8T5byk1xlncM5zr/DvF/Zm/uxWjdrmvN6a2bM684PXZ9H1mGMyJKFnT4nH7GkVVs7pQLXxUr11K0seHkNxUTUzJu9gxuTQ1u3AdpaOe5gOhx5KcWnu/0lClyDrjjO9GpNq4lH+6ZImAHMJQpBMT6lEWcKySZPY+MZkrhxRSUkE3a7cDk89Mp7PzziLfueem34Bk0y+K3okmlR+M7tF0gDgAOAPZvZu6sXKPFZbS/tOiqj4AGWtYe+Oxeyqzemg1QVNkza/pB7AfwLnAUskDU+5VClEUv3Pet2x31JYmMRzwzsO+DPQ1cx2Ajm9/y40Jr+Pz1/YxGPzF7lgsnXlgklfWlNtbA1PzBRCbU3+RjYoBOJR/qmSxgLdJI2iIQZnXtP+4IOZu6KU8WMbwpxs+mYT7fZuV19uUVTM3v3yN6hTvtNUcgoBhxO4MR8MfJQNN7zJ2MzSzOs2y0TKBieuVJKKjUFJJOoNXUwTxmnYXWb2nplNSkTxJd0vaZb7tQitP13Sm5KmSTrY1XWTNFXSXOdF6vGknLgc21yM/YUELs1N7uF18fnbmtmJkh6UdHRdfH7gNwRhEMsJbqQvIghReAvwLoHvUFxx+j25wYerP+SpBU81rhwIt79we32xXVk7rjvlurRu74xH+V9uxriJxuc/HBhpZiZpS13g2mZcN2sIXz4tlKemkRg1bRRPLH+CNp0agv51uboLY1eOrS9/s/QbzjzsTA7qclDa5IrnIddjEOToMrPH4xw30fj8RSFG/CZ3fk4rfyEqeSzKe5TT6YhOUdt3fJb+XACJxO25EohX+ROKz48zp6L0B/YsPr8necy95X+Yd1/jlGn3toA/lzasirXuuC/Dln1Cy9at0y1eQiSi/C8l0Deh+PzAYknHAouBiCbPnsbn9ySHjR8u5dRzazh4QHhLQzCuv/x+I7WVlbmv/JJaE9ygrq1LT9qU+dOM+Pz3EvyqlAE5nf2sECgqhpYlTffLduKZ+V8hmPVXJTJwgvH5vwROSWR8T25RWxXdAXBX7S5qatIfxjEe5d9oZr9PuSSevOXq469m0uhJrNM6Oh7VsVEc0x2bd7D6tdVU9Kugb8fdo2inkniUv1jSs8ASXHZGM/tNSqXy5BWDeg5i0U2LOOTWQyhqU8S+hzQkAVn16iqG9hvK48MeT7t3bTzKf0/KpfDkDC1albFhdQsaL9A18M0GqK3eSYviBtVauXEl5zx4Dm26tWHFjBWsmNE4H8DkysncMeUObhl6S1q/AE1mYJfUAvgB0A/4BHjauTZnjFzz7cknNn3+OU+fVkGfnmuoOLMahTjIrP4CJo8v49g772PAz66trx/x5AieXPkk3Su6R1Tu6q3VfPr0pyy+aTF9OyXd9Gmeb49jPMEurrfd+4QkCeXJQdr16sVFc+fz3oISloV5ej37WCsG33RHI8UHqNlZQ0n7kqizeknbEkpLS6ndld5dcfGYPfuZ2aXu+BVJM2L29uQ11Vu28PIlF9Ctxy5enAgvTgxt3cG8u26j86BB9Dj55EyJGDfxKP9Ktzb/NjAYKLxszZ56Zt4wkrJN8zjrsh1E8kH7YvlWnv3uWfx07QZK2rbdvUMWEY/ZcxmwDDgCWAr8KKUSebKaHV9voPeBkRUfoEc/KCqCnTvS76uTKLHSEg10WRgrgK/dC+DbwNQ0yObJI7au2UqL4uhzbVVV+gP4xjJ7jiBIPxoeVdXwyu9JgMuHXM626m0Qcj87YeIEfnRpgxGxz3H70GvfXmmVK56lzi5mtsb5+HwXmG5m69IiXXSZ0rrUGW2VohCXPadc+D16l7zAIYOi93ngt6348WcrKds3ekb7rN/G6Ki7n78T6AtMSoZEuYQPd5KfxKP8df57HZyPj4/w5MkL4lnqnCVpGnCvi9Nf2dQJnvylRUkr5s8uY+mS6D7NNVVbUVF2ploNpUmbPxvJlHuDB7avX8+aBQsa1Z199tm89FLDXqeyDh3oenTsBD7ZYPPHSkI9ysxGug0pdZ1EENEkownpvPJnF7kat8fP/J49JleVP54ozX8NOZakR5IllceTSeJZ7elTd+Cm2/Rut/F4UkQ8qz0bXEz+uswsX6VWJI8nPcQz818BtAWuBdoAl6dUIo8nTcSj/DuANcCHwBjCoq95PLlKPMo/gSB84MVmVgv4SA6evCAe5e9oZmOB9PucejwpJB7lXyfpIqBM0vnEuZMrRnz+CyXNk/SWpHNd3aOuPF3SJQl/Co+nGcSj/FcBHQm2Me4HXN3UCaHx+YESSaHPuq8HTnKv/wqpv9TMTjKzJ+IT3ePZM2Iudbq0RJPN7MwEx40Vn/8TglUjaAhDbsDjkr4CrjOzxoFdPJ4UEFP5XbKIj53ZE5qZ5dNY5xE7Pv+zBDvEBAxzdb80s68lnQD8EbggfEAfotyTbOLZyfW3sCozs6uaOOdaYL2ZPSXp+wThT0a7tncI9gEDvGRmJ4SdOzu8LsL43rcni0jETyfSrrgU/y+j+vbE2sBeQjADf0gQveG5BDQuanx+gucG2wlMnRJ3rXIz2yzpICIkpvDkD9k0acW64X0S6E5gogwGRsc7qJm9DdTF59+Ji8/vmh8kyMgyF5dsApgoaTbwCEFyOo8n5cTy559qZqeElF83s1PTJlkMvNmTXWR5DNPEzR6gj6TbQ8p968o+RLknH4g181dEO8nMMhqv08/82UWuzvx+J5en2WRg5aY5NMvs8XhikoWKnhDxuDd4PHmJV35PwZKzZk+6k5d5chYzs4jKkpM3vJlC0gIzG5xpObKNXP27eLPHU7B45fcULF75E+OhprsUJDn5d/E2v6dg8TO/p2DJ2aXObEDSdHdYS7Cx/wYzWyvpUaA/DbkMzjCz6vRLmHwknQQ8BiwHioC/EOzfeAv4wHV7wsyy3hTyyr/nnGZmtZJOIdir8H1Xf6mZLc+gXKlkvJndLKmMIE3VMuDfZpZTaWoL0uyRdJKklyW9IGmOpLaSiiX9XdJM914cqV+0Mc1sKtBOUvanJEkSZlZJsOf6nEzL0hwKUvkd1WZ2DvAScCpwPrDUJd54H/hBlH6xWAd0cMcTXRyiZ5IvelaxiiBax+nu806X9L1MCxUPhWz2LHHvKwmiTXQliE0EsAA4ClgboV8sOgEb3HE+mz2hdCew+b/yZk/uELrGK4J4Qke58mBXjtQvIm7zz0Yz25lMIbMZl6DwF8DzmZalORTyzB/OZOACSTMJVm7uAY6P47zXJNUSRLK+NqR+oqS61Z4LzWx9UqXNLJdJOpZgtechgogbp4esfk0xs/syJVy8+IdcnoKlkM0eT4Hjld9TsHjl9xQsXvk9BYtXfk/B4pU/SUiqkDTVPeF8XVI8y6RZj6QBkn6caTlSgV/qTAKSOgD/BL7nok3vBfQzs0UZFq0eSS3MbFem5cgm/MyfHM4GJpjZZgAz21Kn+JKGu9xks1y6JiS9J+kJSe9KGhCtXx2SOkp6XtI0SWNc3d/drHy4pEmu7k2X32yBpO+6uumS7iXIfBNpnGvdedMkDZJ0vsuZNlXS2c65707X97+dg99UST2ifZacwcz8aw9fBGHVh7rjS4DZwB8InNyeJ3CLaE+Q4gkCn6FSgifI90frFzL+H4Fj3fE9BGmeOgOvA68C3VzbJ0APgrRPc1zd9JBzI43zKlDm6gSMB3qFlE8C7gS6AK+4+hOAByN9lkz/LxJ5efeG5LAa6AZgZk9ImgvcBvQBjgSmhfVfbmZVkuqc5aL1q6M/cLckA9oC88zsDUmfArvMbJXr95WZfQEgKdTHaGG0cYBbgQclVQO3AHcBN0sqdsd19AIWu+MF7rxInyVn8MqfHF4Cnpb0lJltouHv+hkw38wuAJDU0tWHO8tF61fHMgKzaqFrL5Z0BLAX0ErSQWa2DGgvaT/gawK/mzp2RRsHaGlmVypIAXslMNrMhks6jiBb5pPu3M8JvqDQDMe/bMQrfxIws/WSbgOek7SLYFvj3a5+inOW2wlMBe6Icn6sfr8DHpLUjkCRryEwYS4jSO00VtJQAnfq24ABQGhuhWjjDAd+K6k30IogQeBtkr5F8MvwyxAZ17j7grlANXBFM/5UWYVf7ckjFEcyP08DfrXHU7D4md9TsPiZ31OweOX3FCxe+T0Fi1d+T8Hild9TsHjl9xQs/w9wAz6GZOkvzQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_nonDE_auc, psix_DE_auc], positions=[1,5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_nonDE_auc, kw_DE_auc], positions=[1.75,5.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_nonDE_auc, gearyc_DE_auc], positions=[2.5,6.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75])\n",
    "ax.set_xticklabels(['non DE', 'DE'], size=8)\n",
    "ax.set_xlabel('Gene expression', size=8)\n",
    "ax.set_title('Simulated data', size=8)\n",
    "\n",
    "ax.set_yticks([0.85, 0.9, 0.95])\n",
    "ax.set_yticklabels(['0.85', '0.90', '0.95'], size=8)\n",
    "ax.set_ylabel('Precision-recall AUC', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0, 8])\n",
    "\n",
    "# plt.savefig('../plots/single_lineage_auc_DEvnonDE.png', bbox_inches='tight', res=20000, dpi =2000)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:05<00:00, 37.08it/s]\n",
      "100%|██████████| 200/200 [00:05<00:00, 35.21it/s]\n",
      "100%|██████████| 200/200 [00:05<00:00, 34.67it/s]\n"
     ]
    }
   ],
   "source": [
    "midDE = ATres.loc[(ATres.loc[idx].FDR < 0.05) & (ATres.loc[idx].meanLogFC >= 0.25) & (ATres.loc[idx].meanLogFC < 0.75)].index\n",
    "bigDE = ATres.loc[(ATres.loc[idx].FDR < 0.05) & (ATres.loc[idx].meanLogFC >= 0.75)].index\n",
    "nDE = ATres.loc[(ATres.loc[idx].FDR > 0.05) | (ATres.loc[idx].meanLogFC < 0.25)].index\n",
    "\n",
    "\n",
    "psix_bigDE_auc, kw_bigDE_auc, gearyc_bigDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in bigDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in bigDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in bigDE]])\n",
    "\n",
    "psix_midDE_auc, kw_midDE_auc, gearyc_midDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in midDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in midDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in midDE]])\n",
    "\n",
    "psix_nDE_auc, kw_nDE_auc, gearyc_nDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in nDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in nDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in nDE]])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_nDE_auc, psix_midDE_auc, psix_bigDE_auc], positions=[1,5, 9], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_nDE_auc, kw_midDE_auc, kw_bigDE_auc], positions=[1.75,5.75, 9.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_nDE_auc, gearyc_midDE_auc, gearyc_bigDE_auc], positions=[2.5,6.5, 10.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75, 9.75])\n",
    "ax.set_xticklabels(['nonDE', 'midDE', 'bigDE'], size=8)\n",
    "ax.set_xlabel('Gene expression', size=8)\n",
    "\n",
    "ax.set_yticks([0.6, 0.7, 0.8, 0.9, 1])\n",
    "ax.set_yticklabels(['0.6', '0.7', '0.8', '0.9', '1.0'], size=8)\n",
    "ax.set_ylabel('Precision-recall AUC', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0, 12])\n",
    "\n",
    "plt.savefig('../plots/single_lineage_auc_bigDE_v_midDE_v_nonDE.png', bbox_inches='tight', res=20000, dpi =2000)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.special import logit, expit\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import recall_score, precision_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import auc\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "platonic_psi = pd.read_csv('sim_output/psi_platonic.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "platonic_psi.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "\n",
    "platonic_psi = expit(platonic_psi)\n",
    "\n",
    "\n",
    "def get_random_samples(psix_results, kw, gearyc):\n",
    "    \n",
    "    pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results.index\n",
    "    neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results.index\n",
    "\n",
    "    psix_auc = []\n",
    "    kw_auc = []\n",
    "    gearyc_auc = []\n",
    "\n",
    "    for i in tqdm(range(200)):\n",
    "\n",
    "        pos_sample = np.random.choice(pos, int(len(pos)*0.2), replace=False)\n",
    "        neg_sample = np.random.choice(neg, int(len(neg)*0.2), replace=False)\n",
    "\n",
    "        labels = [1]*len(pos_sample) + [0]*len(neg_sample)\n",
    "        order = list(pos_sample) + list(neg_sample)\n",
    "\n",
    "\n",
    "        y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "\n",
    "        rec = precision_score(y_, psix_results.loc[order].qvals<=0.05)\n",
    "        rec_kw = precision_score(y_, kw.loc[order].qvals<=0.05)\n",
    "        rec_c = precision_score(y_, gearyc.loc[order].qvals<=0.05)\n",
    "\n",
    "        psix_auc.append(rec)\n",
    "        kw_auc.append(rec_kw)\n",
    "        gearyc_auc.append(rec_c)\n",
    "        \n",
    "    return psix_auc, kw_auc, gearyc_auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:06<00:00, 30.65it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.97it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.69it/s]\n"
     ]
    }
   ],
   "source": [
    "psix_auc, kw_auc, gearyc_auc = get_random_samples(psix_results_1, kw_1, gearyc_1)\n",
    "\n",
    "psix_DE_auc, kw_DE_auc, gearyc_DE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in DE]])\n",
    "\n",
    "psix_nonDE_auc, kw_nonDE_auc, gearyc_nonDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in nonDE]])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_auc, psix_nonDE_auc, psix_DE_auc], positions=[1,5, 9], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_auc, kw_nonDE_auc, kw_DE_auc], positions=[1.75,5.75, 9.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_auc, gearyc_nonDE_auc, gearyc_DE_auc], positions=[2.5,6.5, 10.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75, 9.75])\n",
    "ax.set_xticklabels(['all', 'non DE', 'DE'], size=8)\n",
    "ax.set_xlabel('Gene expression', size=8)\n",
    "\n",
    "# ax.set_yticks([0.85, 0.9, 0.95])\n",
    "# ax.set_yticklabels(['0.85', '0.90', '0.95'], size=8)\n",
    "ax.set_ylabel('Precision', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0, 12])\n",
    "\n",
    "plt.savefig('../plots/single_lineage_precision_DEvnonDE.png', bbox_inches='tight', res=20000, dpi =2000)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:06<00:00, 29.97it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.59it/s]\n",
      "100%|██████████| 200/200 [00:06<00:00, 32.56it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.special import logit, expit\n",
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import recall_score, precision_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import auc\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "platonic_psi = pd.read_csv('sim_output/psi_platonic.tab.gz', sep='\\t', \n",
    "                               names = ['cell_'+str(i+1) for i in range(1000)])\n",
    "platonic_psi.index = ['gene' + str(i+1) for i in range(5000)]\n",
    "\n",
    "platonic_psi = expit(platonic_psi)\n",
    "\n",
    "\n",
    "def get_random_samples(psix_results, kw, gearyc):\n",
    "    \n",
    "    pos = pd.Index(['gene' + str(i) for i in range(1, 2501)]) & psix_results.index\n",
    "    neg = pd.Index(['gene' + str(i) for i in range(2501, 5001)]) & psix_results.index\n",
    "\n",
    "    psix_auc = []\n",
    "    kw_auc = []\n",
    "    gearyc_auc = []\n",
    "\n",
    "    for i in tqdm(range(200)):\n",
    "\n",
    "        pos_sample = np.random.choice(pos, int(len(pos)*0.2), replace=False)\n",
    "        neg_sample = np.random.choice(neg, int(len(neg)*0.2), replace=False)\n",
    "\n",
    "        labels = [1]*len(pos_sample) + [0]*len(neg_sample)\n",
    "        order = list(pos_sample) + list(neg_sample)\n",
    "\n",
    "\n",
    "        y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[order])\n",
    "\n",
    "\n",
    "        rec = recall_score(y_, psix_results.loc[order].qvals<=0.05)\n",
    "        rec_kw = recall_score(y_, kw.loc[order].qvals<=0.05)\n",
    "        rec_c = recall_score(y_, gearyc.loc[order].qvals<=0.05)\n",
    "\n",
    "        psix_auc.append(rec)\n",
    "        kw_auc.append(rec_kw)\n",
    "        gearyc_auc.append(rec_c)\n",
    "        \n",
    "    return psix_auc, kw_auc, gearyc_auc\n",
    "\n",
    "psix_auc, kw_auc, gearyc_auc = get_random_samples(psix_results_1, kw_1, gearyc_1)\n",
    "\n",
    "psix_DE_auc, kw_DE_auc, gearyc_DE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in DE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in DE]])\n",
    "\n",
    "psix_nonDE_auc, kw_nonDE_auc, gearyc_nonDE_auc = get_random_samples(psix_results_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        kw_1.loc[['gene'+str(x) for x in nonDE]], \n",
    "                                                        gearyc_1.loc[['gene'+str(x) for x in nonDE]])\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "# gs = GridSpec(1,7)\n",
    "# gs.update(wspace=0.05, hspace=0.10)\n",
    "ax = fig.add_subplot(111)\n",
    "# ax_2 = fig.add_subplot(gs[0,4:7])# option 1, specify props dictionaries\n",
    "\n",
    "ax.boxplot([psix_auc, psix_nonDE_auc, psix_DE_auc], positions=[1,5, 9], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='skyblue', color='navy'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='navy'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([kw_auc, kw_nonDE_auc, kw_DE_auc], positions=[1.75,5.75, 9.75], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='darkorange', color='darkred'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkred'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.boxplot([gearyc_auc, gearyc_nonDE_auc, gearyc_DE_auc], positions=[2.5,6.5, 10.5], notch=True, patch_artist=True,\n",
    "            boxprops=dict(facecolor='limegreen', color='darkgreen'),\n",
    "            capprops=dict(color='black'),\n",
    "            whiskerprops=dict(color='black'),\n",
    "            flierprops=dict(color='black', markeredgecolor='black'),\n",
    "            medianprops=dict(color='darkgreen'), showfliers=False\n",
    "            )\n",
    "\n",
    "ax.set_xticks([1.75, 5.75, 9.75])\n",
    "ax.set_xticklabels(['all', 'non DE', 'DE'], size=8)\n",
    "ax.set_xlabel('Gene expression', size=8)\n",
    "\n",
    "# ax.set_yticks([0.85, 0.9, 0.95])\n",
    "# ax.set_yticklabels(['0.85', '0.90', '0.95'], size=8)\n",
    "ax.set_ylabel('Recall', size=8)\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0, 12])\n",
    "\n",
    "plt.savefig('../plots/single_lineage_recall_DEvnonDE.png', bbox_inches='tight', res=20000, dpi =2000)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8134328358208955\n",
      "0.936426116838488\n",
      "0.781433607520564\n",
      "0.9554597701149425\n"
     ]
    }
   ],
   "source": [
    "psix_de = psix_results_1.index.intersection(pd.Index(['gene'+str(x) for x in DE]))\n",
    "psix_nonde = psix_results_1.index.intersection(pd.Index(['gene'+str(x) for x in nonDE]))\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[psix_de])\n",
    "psix_rec_DE = recall_score(y_, psix_results_1.loc[psix_de].qvals<=0.05)\n",
    "psix_pre_DE = precision_score(y_, psix_results_1.loc[psix_de].qvals<=0.05)\n",
    "\n",
    "print(psix_rec_DE)\n",
    "print(psix_pre_DE)\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[psix_nonde])\n",
    "psix_rec_nonDE = recall_score(y_, psix_results_1.loc[psix_nonde].qvals<=0.05)\n",
    "psix_pre_nonDE = precision_score(y_, psix_results_1.loc[psix_nonde].qvals<=0.05)\n",
    "\n",
    "\n",
    "print(psix_rec_nonDE)\n",
    "print(psix_pre_nonDE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8208955223880597\n",
      "0.7670850767085077\n",
      "0.7931844888366627\n",
      "0.8777633289986996\n"
     ]
    }
   ],
   "source": [
    "kw_de = kw_1.index.intersection(pd.Index(['gene'+str(x) for x in DE]))\n",
    "kw_nonde = kw_1.index.intersection(pd.Index(['gene'+str(x) for x in nonDE]))\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[kw_de])\n",
    "kw_rec_DE = recall_score(y_, kw_1.loc[kw_de].qvals<=0.05)\n",
    "kw_pre_DE = precision_score(y_, kw_1.loc[kw_de].qvals<=0.05)\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[kw_nonde])\n",
    "kw_rec_nonDE = recall_score(y_, kw_1.loc[kw_nonde].qvals<=0.05)\n",
    "kw_pre_nonDE = precision_score(y_, kw_1.loc[kw_nonde].qvals<=0.05)\n",
    "\n",
    "\n",
    "print(kw_rec_DE)\n",
    "print(kw_pre_DE)\n",
    "print(kw_rec_nonDE)\n",
    "print(kw_pre_nonDE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8835820895522388\n",
      "0.7072879330943848\n",
      "0.8108108108108109\n",
      "0.9090909090909091\n"
     ]
    }
   ],
   "source": [
    "gearyc_de = gearyc_1.index.intersection(pd.Index(['gene'+str(x) for x in DE]))\n",
    "gearyc_nonde = gearyc_1.index.intersection(pd.Index(['gene'+str(x) for x in nonDE]))\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[gearyc_de])\n",
    "gearyc_rec_DE = recall_score(y_, gearyc_1.loc[gearyc_de].qvals<=0.05)\n",
    "gearyc_pre_DE = precision_score(y_, gearyc_1.loc[gearyc_de].qvals<=0.05)\n",
    "\n",
    "y_ = list(((platonic_psi.max(axis=1) - platonic_psi.min(axis=1)) >= 0.2).astype(int).loc[gearyc_nonde])\n",
    "gearyc_rec_nonDE = recall_score(y_, gearyc_1.loc[gearyc_nonde].qvals<=0.05)\n",
    "gearyc_pre_nonDE = precision_score(y_, gearyc_1.loc[gearyc_nonde].qvals<=0.05)\n",
    "\n",
    "\n",
    "print(gearyc_rec_DE)\n",
    "print(gearyc_pre_DE)\n",
    "print(gearyc_rec_nonDE)\n",
    "print(gearyc_pre_nonDE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Precision')"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter([psix_rec_nonDE], [psix_pre_nonDE], marker='v', c='navy')\n",
    "ax.scatter([psix_rec_DE], [psix_pre_DE], marker='^', c='navy')\n",
    "\n",
    "ax.scatter([kw_rec_nonDE], [kw_pre_nonDE], marker='v', c='darkorange')\n",
    "ax.scatter([kw_rec_DE], [kw_pre_DE], marker='^', c='darkorange')\n",
    "\n",
    "ax.scatter([gearyc_rec_nonDE], [gearyc_pre_nonDE], marker='v', c='forestgreen')\n",
    "ax.scatter([gearyc_rec_DE], [gearyc_pre_DE], marker='^', c='forestgreen')\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.set_xlim([0.5, 1])\n",
    "ax.set_ylim([0.5, 1])\n",
    "\n",
    "\n",
    "ax.set_xlabel('Recall', size=8)\n",
    "ax.set_ylabel('Precision', size=8)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.special import logit, expit\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import precision_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "deltaPSI = expit(psi_platonic).max(axis=1) - expit(psi_platonic).min(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "bin_0 = deltaPSI.loc[(deltaPSI <= 0.01)].index\n",
    "bin_0_20 = deltaPSI.loc[(deltaPSI > 0.01) & (deltaPSI <= 0.2)].index\n",
    "bin_20_40 = deltaPSI.loc[(deltaPSI > 0.2) & (deltaPSI <= 0.4)].index\n",
    "bin_40_60 = deltaPSI.loc[(deltaPSI > 0.4) & (deltaPSI <= 0.6)].index\n",
    "bin_60_80 = deltaPSI.loc[(deltaPSI > 0.6) & (deltaPSI <= 0.8)].index\n",
    "bin_80 = deltaPSI.loc[(deltaPSI > 0.8)].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_recall(predict, exon_bin):\n",
    "    return recall_score(([True]*len(exon_bin)), predict.loc[['gene'+str(x) for x in exon_bin]])\n",
    "\n",
    "\n",
    "def get_precision(predict, deltaPSI, cutoff):\n",
    "    delta_pos = ['gene'+str(x) for x in deltaPSI.loc[deltaPSI >= cutoff].index]\n",
    "    delta_neg = ['gene'+str(x) for x in deltaPSI.loc[deltaPSI < 0.01].index]\n",
    "    \n",
    "    delta_y = [True]*len(delta_pos) + [False]*len(delta_neg)\n",
    "    \n",
    "    print(len(delta_y))\n",
    "    \n",
    "    return precision_score(delta_y, predict.loc[delta_pos + delta_neg])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "plt.plot([ 0.4, 0.6, 0.8, 1.0],\n",
    "     [#get_recall((psix_results_1.qvals <= 0.05), bin_0_20),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_20_40),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_40_60),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_60_80),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_80)],\n",
    "        c='navy', linewidth=3)\n",
    "\n",
    "plt.scatter([ 0.4, 0.6, 0.8, 1.0],\n",
    "     [#get_recall((psix_results_1.qvals <= 0.05), bin_0_20),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_20_40),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_40_60),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_60_80),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_80)],\n",
    "            s=30, c='navy', linewidth=3)\n",
    "\n",
    "plt.xlabel(r'Simulated $\\Delta \\Psi$', fontsize=8)\n",
    "plt.ylabel('Recall', fontsize=8)\n",
    "ax.set_xticks([ 0.4, 0.6, 0.8, 1.0])\n",
    "ax.set_xticklabels(['(0.2-0.4]', '(0.4-0.6]', '(0.6-0.8]', '(0.8-1.0]'])\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "plt.ylim([-0.05, 1.05])\n",
    "\n",
    "ax.tick_params(axis='both', which='major', labelsize=8)\n",
    "plt.savefig('/mnt/lareaulab/cfbuenabadn/plots_pdf/delta_psi_psix_recall.pdf', bbox_inches='tight', res=20000, dpi =2000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.94679186228482"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(deltaPSI >= 0.2, (psix_results_1.loc[['gene'+str(x) for x in deltaPSI.index]].qvals <= 0.05))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_score, recall_score, f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_stats(x, pos, neg):\n",
    "    \n",
    "    pos = x.index.intersection(pos)\n",
    "    neg = x.index.intersection(neg)\n",
    "          \n",
    "    labels = [1]*len(pos) + [0]*len(neg)\n",
    "    pred = list(x.loc[pos]) + list(x.loc[neg])\n",
    "    \n",
    "    stats_dict = {}\n",
    "    \n",
    "    stats_dict.update({'precision':precision_score(labels, pred)})\n",
    "    stats_dict.update({'recall':recall_score(labels, pred)})\n",
    "    stats_dict.update({'f1_score':f1_score(labels, pred)})\n",
    "          \n",
    "    return stats_dict\n",
    "          \n",
    "    \n",
    "def get_stats_DE(x, pos, neg, isDE):\n",
    "    \n",
    "    pos_DE = pos.intersection(isDE.loc[isDE].index)\n",
    "    neg_DE = neg.intersection(isDE.loc[isDE].index)\n",
    "    \n",
    "    pos_nonDE = pos.intersection(isDE.loc[~isDE].index)\n",
    "    neg_nonDE = neg.intersection(isDE.loc[~isDE].index)\n",
    "    \n",
    "    stats_dict = {}\n",
    "    \n",
    "    stats_dict.update({'all':get_stats(x, pos, neg)})\n",
    "    stats_dict.update({'DE':get_stats(x, pos_DE, neg_DE)})\n",
    "    stats_dict.update({'nonDE':get_stats(x, pos_nonDE, neg_nonDE)})\n",
    "    \n",
    "    return stats_dict\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ATres.index = ['gene'+str(x) for x in ATres.index]\n",
    "# nonDS = pd.Index(['gene' + str(x) for x in nonDS])\n",
    "# DS = pd.Index(['gene' + str(x) for x in DS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "psix_stats =[]\n",
    "kw_stats = []\n",
    "gearyc_stats = []\n",
    "for i in tqdm(range(200), position=0, leave=True):\n",
    "    psix_stats.append(get_stats_DE(psix_results_1.qvals <= 0.05, \n",
    "                              pd.Index(np.random.choice(weyn_pos.intersection(psix_results_1.index), int(len(weyn_pos.intersection(psix_results_1.index))*0.2), replace=False)), \n",
    "                              pd.Index(np.random.choice(weyn_neg.intersection(psix_results_1.index), int(len(weyn_neg.intersection(psix_results_1.index))*0.2), replace=False)), \n",
    "                              ATres.significant))\n",
    "    \n",
    "    kw_stats.append(get_stats_DE(kw_1.qvals <= 0.05, \n",
    "                              pd.Index(np.random.choice(weyn_pos.intersection(kw_1.index), int(len(weyn_pos.intersection(kw_1.index))*0.2), replace=False)), \n",
    "                              pd.Index(np.random.choice(weyn_neg.intersection(kw_1.index), int(len(weyn_neg.intersection(kw_1.index))*0.2), replace=False)), \n",
    "                              ATres.significant))\n",
    "    \n",
    "    gearyc_stats.append(get_stats_DE(gearyc_1.qvals <= 0.05, \n",
    "                              pd.Index(np.random.choice(weyn_pos.intersection(gearyc_1.index), int(len(weyn_pos.intersection(gearyc_1.index))*0.2), replace=False)), \n",
    "                              pd.Index(np.random.choice(weyn_neg.intersection(gearyc_1.index), int(len(weyn_neg.intersection(gearyc_1.index))*0.2), replace=False)), \n",
    "                              ATres.significant))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_psix_stats = get_stats_DE(psix_results_1.qvals <= 0.05, DS, nonDS, ATres.significant)\n",
    "all_kw_stats = get_stats_DE(kw_1.qvals <= 0.05, DS, nonDS, ATres.significant)\n",
    "all_gearyc_stats = get_stats_DE(gearyc_1.qvals <= 0.05, DS, nonDS, ATres.significant)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.scatter([all_psix_stats['DE']['recall']], [all_psix_stats['DE']['precision']], c='navy', marker='^')\n",
    "ax.scatter([all_psix_stats['nonDE']['recall']], [all_psix_stats['nonDE']['precision']], c='navy', marker='v')\n",
    "\n",
    "\n",
    "ax.scatter([all_kw_stats['DE']['recall']], [all_kw_stats['DE']['precision']], c='darkorange', marker='^')\n",
    "ax.scatter([all_kw_stats['nonDE']['recall']], [all_kw_stats['nonDE']['precision']], c='darkorange', marker='v')\n",
    "\n",
    "ax.scatter([all_gearyc_stats['DE']['recall']], [all_gearyc_stats['DE']['precision']], c='forestgreen', marker='^')\n",
    "ax.scatter([all_gearyc_stats['nonDE']['recall']], [all_gearyc_stats['nonDE']['precision']], c='forestgreen', marker='v')\n",
    "ax.set_xlim([-0.05, 1.01])\n",
    "ax.set_ylim([-0.05, 1.01])\n",
    "\n",
    "\n",
    "plt.xlabel('Recall', fontsize=12)\n",
    "plt.ylabel('Precision', fontsize=12)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.tick_params(axis='both', which='major', labelsize=8)\n",
    "\n",
    "plt.savefig('../plots/pre_rec_scatter.png', dpi=1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.010298867659163628"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.std([x['nonDE']['precision'] for x in psix_stats])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1756"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(nonDS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# deltaPSI = expit(psi_platonic).max(axis=1) - expit(psi_platonic).min(axis=1)\n",
    "\n",
    "bin_0 = deltaPSI.loc[(deltaPSI <= 0.01)].index\n",
    "bin_0_20 = deltaPSI.loc[(deltaPSI > 0.01) & (deltaPSI <= 0.2)].index\n",
    "bin_20_40 = deltaPSI.loc[(deltaPSI > 0.2) & (deltaPSI <= 0.4)].index\n",
    "bin_40_60 = deltaPSI.loc[(deltaPSI > 0.4) & (deltaPSI <= 0.6)].index\n",
    "bin_60_80 = deltaPSI.loc[(deltaPSI > 0.6) & (deltaPSI <= 0.8)].index\n",
    "bin_80 = deltaPSI.loc[(deltaPSI > 0.8)].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_recall(predict, exon_bin):\n",
    "    return recall_score(([True]*len(exon_bin)), predict.loc[[str(x) for x in exon_bin]])\n",
    "\n",
    "\n",
    "def get_precision(predict, deltaPSI, cutoff):\n",
    "    delta_pos = [str(x) for x in deltaPSI.loc[deltaPSI >= cutoff].index]\n",
    "    delta_neg = [str(x) for x in deltaPSI.loc[deltaPSI < 0.01].index]\n",
    "    \n",
    "    delta_y = [True]*len(delta_pos) + [False]*len(delta_neg)\n",
    "    \n",
    "    print(len(delta_y))\n",
    "    \n",
    "    return precision_score(delta_y, predict.loc[delta_pos + delta_neg])\n",
    "\n",
    "def plot_recall(x, y, color, label):\n",
    "    plt.plot(x, y, c=color, linewidth=3)\n",
    "    plt.scatter(x, y, s=80, facecolors='white', edgecolors=color, linewidth=3, label=label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>psix_score</th>\n",
       "      <th>pvals</th>\n",
       "      <th>qvals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>gene1128</th>\n",
       "      <td>1.515963</td>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.001716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene1880</th>\n",
       "      <td>1.364254</td>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.001716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene2332</th>\n",
       "      <td>1.325777</td>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.001716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene705</th>\n",
       "      <td>1.319631</td>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.001716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene1878</th>\n",
       "      <td>1.233293</td>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.001716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene4900</th>\n",
       "      <td>-0.075527</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene3637</th>\n",
       "      <td>-0.081095</td>\n",
       "      <td>0.991004</td>\n",
       "      <td>0.998666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene3293</th>\n",
       "      <td>-0.084103</td>\n",
       "      <td>0.994003</td>\n",
       "      <td>0.999739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene2372</th>\n",
       "      <td>-0.099092</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gene2807</th>\n",
       "      <td>-0.100722</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3486 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          psix_score     pvals     qvals\n",
       "gene1128    1.515963  0.000500  0.001716\n",
       "gene1880    1.364254  0.000500  0.001716\n",
       "gene2332    1.325777  0.000500  0.001716\n",
       "gene705     1.319631  0.000500  0.001716\n",
       "gene1878    1.233293  0.000500  0.001716\n",
       "...              ...       ...       ...\n",
       "gene4900   -0.075527  1.000000  1.000000\n",
       "gene3637   -0.081095  0.991004  0.998666\n",
       "gene3293   -0.084103  0.994003  0.999739\n",
       "gene2372   -0.099092  1.000000  1.000000\n",
       "gene2807   -0.100722  1.000000  1.000000\n",
       "\n",
       "[3486 rows x 3 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "psix_results_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "\"None of [Index(['3328', '2860', '3276', '3527', '4876', '3972', '4621', '3331', '3483',\\n       '2868',\\n       ...\\n       '2851', '4789', '3013', '4659', '3031', '3828', '4900', '3637', '3293',\\n       '2807'],\\n      dtype='object', length=1756)] are in the [index]\"",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-34-efb5fc3caf8e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m plot_recall([0, 0.2, 0.4, 0.6, 0.8, 1.0],\n\u001b[0;32m---> 19\u001b[0;31m      [get_recall((psix_results_1.qvals <= 0.05), bin_0),\n\u001b[0m\u001b[1;32m     20\u001b[0m       \u001b[0mget_recall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpsix_results_1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mqvals\u001b[0m \u001b[0;34m<=\u001b[0m \u001b[0;36m0.05\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbin_0_20\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m       \u001b[0mget_recall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpsix_results_1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mqvals\u001b[0m \u001b[0;34m<=\u001b[0m \u001b[0;36m0.05\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbin_20_40\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m<ipython-input-32-9708cb5b6c66>\u001b[0m in \u001b[0;36mget_recall\u001b[0;34m(predict, exon_bin)\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mget_recall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexon_bin\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mrecall_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mexon_bin\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpredict\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mexon_bin\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mget_precision\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdeltaPSI\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcutoff\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/pandas/core/indexing.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1766\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1767\u001b[0m             \u001b[0mmaybe_callable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_if_callable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1768\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_getitem_axis\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmaybe_callable\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1769\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1770\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_is_scalar_access\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkey\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTuple\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/pandas/core/indexing.py\u001b[0m in \u001b[0;36m_getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1952\u001b[0m                     \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Cannot index with multidimensional key\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1953\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1954\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_getitem_iterable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1955\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1956\u001b[0m             \u001b[0;31m# nested tuple slicing\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/pandas/core/indexing.py\u001b[0m in \u001b[0;36m_getitem_iterable\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1593\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1594\u001b[0m             \u001b[0;31m# A collection of keys\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1595\u001b[0;31m             \u001b[0mkeyarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_listlike_indexer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraise_missing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1596\u001b[0m             return self.obj._reindex_with_indexers(\n\u001b[1;32m   1597\u001b[0m                 \u001b[0;34m{\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mkeyarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mallow_dups\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/pandas/core/indexing.py\u001b[0m in \u001b[0;36m_get_listlike_indexer\u001b[0;34m(self, key, axis, raise_missing)\u001b[0m\n\u001b[1;32m   1551\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1552\u001b[0m         self._validate_read_indexer(\n\u001b[0;32m-> 1553\u001b[0;31m             \u001b[0mkeyarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mo\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_axis_number\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mraise_missing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mraise_missing\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1554\u001b[0m         )\n\u001b[1;32m   1555\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mkeyarr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/pandas/core/indexing.py\u001b[0m in \u001b[0;36m_validate_read_indexer\u001b[0;34m(self, key, indexer, axis, raise_missing)\u001b[0m\n\u001b[1;32m   1638\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mmissing\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1639\u001b[0m                 \u001b[0maxis_name\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_axis_name\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1640\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"None of [{key}] are in the [{axis_name}]\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1641\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1642\u001b[0m             \u001b[0;31m# We (temporarily) allow for some missing keys with .loc, except in\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: \"None of [Index(['3328', '2860', '3276', '3527', '4876', '3972', '4621', '3331', '3483',\\n       '2868',\\n       ...\\n       '2851', '4789', '3013', '4659', '3031', '3828', '4900', '3637', '3293',\\n       '2807'],\\n      dtype='object', length=1756)] are in the [index]\""
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "# plot_recall([0, 0.2, 0.4, 0.6, 0.8, 1.0],\n",
    "#      [get_recall((kw_1.qvals <= 0.05), bin_0),\n",
    "#       get_recall((kw_1.qvals <= 0.05), bin_0_20),\n",
    "#       get_recall((kw_1.qvals <= 0.05), bin_20_40),\n",
    "#       get_recall((kw_1.qvals <= 0.05), bin_40_60),\n",
    "#       get_recall((kw_1.qvals <= 0.05), bin_60_80),\n",
    "#       get_recall((kw_1.qvals <= 0.05), bin_80)], 'darkorange', 'Kruskal-Wallis')\n",
    "\n",
    "# plot_recall([0, 0.2, 0.4, 0.6, 0.8, 1.0],\n",
    "#      [get_recall((gearyc_1.qvals <= 0.05), bin_0),\n",
    "#       get_recall((gearyc_1.qvals <= 0.05), bin_0_20),\n",
    "#       get_recall((gearyc_1.qvals <= 0.05), bin_20_40),\n",
    "#       get_recall((gearyc_1.qvals <= 0.05), bin_40_60),\n",
    "#       get_recall((gearyc_1.qvals <= 0.05), bin_60_80),\n",
    "#       get_recall((gearyc_1.qvals <= 0.05), bin_80)], 'forestgreen', \"Geary's C\")\n",
    "\n",
    "plot_recall([0, 0.2, 0.4, 0.6, 0.8, 1.0],\n",
    "     [get_recall((psix_results_1.qvals <= 0.05), bin_0),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_0_20),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_20_40),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_40_60),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_60_80),\n",
    "      get_recall((psix_results_1.qvals <= 0.05), bin_80)], 'steelblue', 'Psix')\n",
    "\n",
    "plt.xlabel(r'Simulated $\\Delta \\Psi$', fontsize=14)\n",
    "plt.ylabel('Recall', fontsize=14)\n",
    "ax.set_xticklabels(['0', '0', '(0-0.2]', '(0.2-0.4]', '(0.4-0.6]', '(0.6-0.8]', '(0.8-1.0]'])\n",
    "plt.title('Simulated exons', fontsize=14)\n",
    "\n",
    "plt.legend(frameon=False, fontsize=12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([3328, 2860, 3276, 3527, 4876, 3972, 4621, 3331, 3483, 2868,\n",
       "            ...\n",
       "            2851, 4789, 3013, 4659, 3031, 3828, 4900, 3637, 3293, 2807],\n",
       "           dtype='int64', length=1756)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bin_0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x7f2e900fd690>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e900e17d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e90094d90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e90052410>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e90002a50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ffbf0d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ff71710>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ffa3e10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ff60410>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ff11a90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fec7d90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fe87350>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fe3c8d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fdefe50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fdb4410>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fd68990>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fd9ef10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fd604d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fd14a50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fccbfd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fc8e590>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fc42b10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e19c018d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e90027950>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ff22610>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e19628310>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e0d0339d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e0d063910>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e049175d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e04b392d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e04b4fbd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e902e6b90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fbe4710>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fb99d50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fb5c3d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fb12a10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fac8f50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fa896d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fa40d10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fa03390>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f9b99d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f971f10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f934690>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f8e9cd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f8ab350>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f8e1990>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f896fd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f85a650>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f80dc90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f7d3310>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f78b950>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f741f90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f702610>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f6b8c10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f67b1d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f62f750>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f5e5cd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f5a8290>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f5de810>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f593d90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f554350>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f50b8d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f4c0e50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f481410>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f438990>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f3eef10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f3b14d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f367a50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f39dfd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f360590>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f314b10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f2da0d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f28d650>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f23ebd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f204190>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f1bb710>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f16fc90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f133250>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f0e77d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e0d045450>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8fb29b50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f9cb8d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8f7ea650>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8efe9f50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8efad5d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8efe3c10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ef25290>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ef5b8d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ef0ff10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8eed3590>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ee8abd0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ee4d250>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ee03890>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8edb8ed0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ed7d550>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ed30b90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ecf5210>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ecaa850>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ece2e90>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ec25510>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ec59b50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e8ec1d1d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88b9b810>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88b50e50>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88b134d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88ac9b10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88a8e190>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88a437d0>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e889f9e10>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e889bb490>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x7f2e88972a90>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "34"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(bin_80)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gene3067    0.000000\n",
       "gene4170    0.000000\n",
       "gene4174    0.000000\n",
       "gene4577    0.000000\n",
       "gene4764    0.000000\n",
       "              ...   \n",
       "gene1580    0.200450\n",
       "gene1112    0.202196\n",
       "gene276     0.203011\n",
       "gene483     0.203090\n",
       "gene2332    0.203364\n",
       "Length: 1909, dtype: float64"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deltaPSI.loc[['gene'+str(x) for x in nonDE]].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 180x120 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6/2.4, 4/2.4))\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "ax.hist(deltaPSI.loc[deltaPSI>0], bins=20)\n",
    "ax.set_xlim([-0.05,1.05])\n",
    "ax.set_xlabel(r'Simulated $\\Delta \\Psi$', size=8)\n",
    "ax.set_ylabel('Counts', size=8)\n",
    "\n",
    "\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "\n",
    "ax.tick_params(axis='both', which='major', labelsize=8)\n",
    "\n",
    "plt.savefig('../plots/single_lineage_delta_PSI.png', bbox_inches='tight', res=20000, dpi =2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting splice junctions....\n",
      "Obtaining PSI tables...\n",
      "Reading TPM and transforming to mRNA counts...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:18<00:00, 52.68it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully processed RNA-seq data\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 7922.38it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:32<00:00, 108.27it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 7904.19it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:32<00:00, 107.65it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:01<00:00,  1.02s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6225.66it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:32<00:00, 107.06it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.51it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 7802.40it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:34<00:00, 101.33it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.41it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 7908.81it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:31<00:00, 111.42it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6202.85it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:34<00:00, 100.72it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:01<00:00,  1.19s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6201.63it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:33<00:00, 104.15it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6158.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:34<00:00, 102.15it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.41it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 5714.45it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:34<00:00, 102.46it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:01<00:00,  1.13s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6211.33it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:37<00:00, 91.97it/s]   "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.13it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n",
      "Computing cell-cell metric...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1000/1000 [00:00<00:00, 6225.09it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed cell-cell metric\n",
      "Computing Psix score in 3486 exons\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 3486/3486 [00:38<00:00, 91.04it/s]   "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully computed Psix score of exons.\n",
      "Estimating p-values. This might take a while...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 1/1 [00:00<00:00,  1.07it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Successfully estimated p-values\n"
     ]
    }
   ],
   "source": [
    "psix_single_lineage_1 = psix.Psix()\n",
    "psix_single_lineage_1.junctions2psi(\n",
    "        'processed_tables/splice_junction_counts_0.1.tab.gz',\n",
    "        '',\n",
    "        'processed_tables/tpm_0.1.tab.gz',\n",
    "        save_files_in='psix_output/psix_object_0.1/')\n",
    "\n",
    "psix_score_list = []\n",
    "\n",
    "for capture in [0.01, 0.05, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5]:\n",
    "\n",
    "    psix_single_lineage_1.run_psix(latent='processed_tables/pc2_rd.tab.gz',  n_jobs=25,\n",
    "                                    n_random_exons=3, n_neighbors=100, capture_efficiency = capture, pvals_bins=1)\n",
    "\n",
    "    psix_results_ =  psix_single_lineage_1.psix_results.psix_score\n",
    "    \n",
    "    psix_score_list.append(psix_results_)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "scores_df = pd.DataFrame()\n",
    "scores_df['0.5'] = psix_score_list[10]\n",
    "scores_df['0.45'] = psix_score_list[9]\n",
    "scores_df['0.4'] = psix_score_list[8]\n",
    "scores_df['0.35'] = psix_score_list[7]\n",
    "scores_df['0.3'] = psix_score_list[6]\n",
    "scores_df['0.25'] = psix_score_list[5]\n",
    "scores_df['0.2'] = psix_score_list[4]\n",
    "scores_df['0.15'] = psix_score_list[3]\n",
    "scores_df['0.1'] = psix_score_list[2]\n",
    "scores_df['0.05'] = psix_score_list[1]\n",
    "scores_df['0.01'] = psix_score_list[0]\n",
    "\n",
    "# scores_df.to_csv('psix_output/scores_different_capture_efficiency.tab.gz', sep='\\t', index=True, header=True)\n",
    "\n",
    "# psix_score_list[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 240x180 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "fig = plt.figure(figsize=(8/2.4, 6/2.4))\n",
    "\n",
    "sns.heatmap(scores_df.corr(), cbar_kws={'label': 'Pearson r'})\n",
    "plt.xlabel('Capture efficiency')\n",
    "plt.ylabel('Capture efficiency')\n",
    "\n",
    "plt.title('Psix score correlation')\n",
    "\n",
    "plt.savefig('../plots/scores_corr.png', bbox_inches='tight', res=20000, dpi =2000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}