diff --git a/04_plot_results.ipynb b/04_plot_stratified_results.ipynb
similarity index 98%
rename from 04_plot_results.ipynb
rename to 04_plot_stratified_results.ipynb
index 23fb325..fee0a0c 100644
--- a/04_plot_results.ipynb
+++ b/04_plot_stratified_results.ipynb
@@ -465,8 +465,8 @@
    ],
    "source": [
     "vogelstein_results_df = au.compare_results(vogelstein_df, metric='aupr', correction=True,\n",
-    "                                correction_method='fdr_bh', correction_alpha=0.001,\n",
-    "                                verbose=True)\n",
+    "                                           correction_method='fdr_bh', correction_alpha=0.001,\n",
+    "                                           verbose=True)\n",
     "vogelstein_results_df.sort_values(by='p_value').head(n=10)"
    ]
   },
@@ -665,7 +665,12 @@
    "source": [
     "The plot above is similar to a volcano plot used in differential expression analysis. The x-axis shows the difference between AUPR in the signal (true labels) case and in the negative control (shuffled labels) case, and the y-axis shows the negative log of the t-test p-value, after FDR adjustment.\n",
     "\n",
-    "Orange points are significant at a cutoff of $\\alpha = 0.001$ after FDR correction."
+    "Orange points are significant at a cutoff of $\\alpha = 0.001$ after FDR correction.\n",
+    "\n",
+    "Our interpretation of these results:\n",
+    "\n",
+    "* For the top 50 analysis, we mostly reproduced the results from BioBombe which also used this gene set (some of the less significant hits weren't found in BioBombe, but we should have better statistical power here so it makes sense that we see more results)\n",
+    "* For the Vogelstein analysis, it was surprising/interesting that we saw lots more significant hits than we did for the top 50 analysis! On some level it's not shocking (if a gene is mutated frequently that doesn't necessarily make it a driver, and conversely drivers aren't always frequently mutated across all samples) but seeing visual confirmation of this was neat."
    ]
   },
   {
@@ -736,6 +741,17 @@
    "source": [
     "We have usually used TTN as our negative control (not understood to be a cancer driver, but is a large gene that is frequently mutated as a passenger). So it's a bit weird that it has a fairly low p-value here (would be significant at $\\alpha = 0.05$). We'll have to think about why this is."
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# save significance testing results\n",
+    "top50_results_df.to_csv(os.path.join(cfg.results_dir, 'top50_stratified_pvals.tsv'), index=False, sep='\\t')\n",
+    "vogelstein_results_df.to_csv(os.path.join(cfg.results_dir, 'vogelstein_stratified_pvals.tsv'), index=False, sep='\\t')"
+   ]
   }
  ],
  "metadata": {
diff --git a/05_plot_cancer_type_results.ipynb b/05_plot_cancer_type_results.ipynb
new file mode 100644
index 0000000..fcd58c6
--- /dev/null
+++ b/05_plot_cancer_type_results.ipynb
@@ -0,0 +1,1539 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Single-cancer holdout analysis"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "from matplotlib_venn import venn2\n",
+    "import seaborn as sns\n",
+    "\n",
+    "import pancancer_evaluation.config as cfg\n",
+    "import pancancer_evaluation.utilities.analysis_utilities as au"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here, we want to look at experiments where a single cancer type is held out. We want to compare cross-validation results when models are trained only on data from a single cancer type with results when models are trained on data from all cancer types.\n",
+    "\n",
+    "In the following plots, each point is a gene/cancer type combination (rather than the stratified experiments, where each point is a gene and the test set is a combination of different cancer types).\n",
+    "\n",
+    "(All of these experiments use the Vogelstein genes, since this gene set seemed to capture more relevant signal in the stratified CV results)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# TODO could merge different seeds into the same directory in the future\n",
+    "pancancer_dir = os.path.join(cfg.results_dir, 'pancancer')\n",
+    "pancancer_dir2 = os.path.join(cfg.results_dir, 'vogelstein_seed1_results', 'pancancer')\n",
+    "single_cancer_dir = os.path.join(cfg.results_dir, 'single_cancer')\n",
+    "single_cancer_dir2 = os.path.join(cfg.results_dir, 'vogelstein_seed1_results', 'single_cancer')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 1 42]\n",
+      "(20940, 10)\n"
+     ]
+    },
+    {
+     "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>auroc</th>\n",
+       "      <th>aupr</th>\n",
+       "      <th>gene</th>\n",
+       "      <th>holdout_cancer_type</th>\n",
+       "      <th>signal</th>\n",
+       "      <th>seed</th>\n",
+       "      <th>data_type</th>\n",
+       "      <th>fold</th>\n",
+       "      <th>train_set</th>\n",
+       "      <th>identifier</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.99357</td>\n",
+       "      <td>0.95200</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>train</td>\n",
+       "      <td>0</td>\n",
+       "      <td>single_cancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.69079</td>\n",
+       "      <td>0.27286</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>test</td>\n",
+       "      <td>0</td>\n",
+       "      <td>single_cancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0.74039</td>\n",
+       "      <td>0.39101</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>cv</td>\n",
+       "      <td>0</td>\n",
+       "      <td>single_cancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0.99982</td>\n",
+       "      <td>0.99826</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>train</td>\n",
+       "      <td>1</td>\n",
+       "      <td>single_cancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0.75636</td>\n",
+       "      <td>0.62930</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>test</td>\n",
+       "      <td>1</td>\n",
+       "      <td>single_cancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     auroc     aupr    gene holdout_cancer_type  signal  seed data_type  fold  \\\n",
+       "0  0.99357  0.95200  MAP3K1                BRCA  signal    42     train     0   \n",
+       "1  0.69079  0.27286  MAP3K1                BRCA  signal    42      test     0   \n",
+       "2  0.74039  0.39101  MAP3K1                BRCA  signal    42        cv     0   \n",
+       "3  0.99982  0.99826  MAP3K1                BRCA  signal    42     train     1   \n",
+       "4  0.75636  0.62930  MAP3K1                BRCA  signal    42      test     1   \n",
+       "\n",
+       "       train_set   identifier  \n",
+       "0  single_cancer  MAP3K1_BRCA  \n",
+       "1  single_cancer  MAP3K1_BRCA  \n",
+       "2  single_cancer  MAP3K1_BRCA  \n",
+       "3  single_cancer  MAP3K1_BRCA  \n",
+       "4  single_cancer  MAP3K1_BRCA  "
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "single_cancer_df1 = au.load_prediction_results(single_cancer_dir, 'single_cancer')\n",
+    "single_cancer_df2 = au.load_prediction_results(single_cancer_dir2, 'single_cancer')\n",
+    "single_cancer_df = pd.concat((single_cancer_df1, single_cancer_df2))\n",
+    "print(np.unique(single_cancer_df.seed))\n",
+    "print(single_cancer_df.shape)\n",
+    "single_cancer_df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 1 42]\n",
+      "(20760, 10)\n"
+     ]
+    },
+    {
+     "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>auroc</th>\n",
+       "      <th>aupr</th>\n",
+       "      <th>gene</th>\n",
+       "      <th>holdout_cancer_type</th>\n",
+       "      <th>signal</th>\n",
+       "      <th>seed</th>\n",
+       "      <th>data_type</th>\n",
+       "      <th>fold</th>\n",
+       "      <th>train_set</th>\n",
+       "      <th>identifier</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.67257</td>\n",
+       "      <td>0.30731</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>train</td>\n",
+       "      <td>0</td>\n",
+       "      <td>pancancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.64775</td>\n",
+       "      <td>0.30118</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>test</td>\n",
+       "      <td>0</td>\n",
+       "      <td>pancancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0.64893</td>\n",
+       "      <td>0.21980</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>cv</td>\n",
+       "      <td>0</td>\n",
+       "      <td>pancancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0.99443</td>\n",
+       "      <td>0.91075</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>train</td>\n",
+       "      <td>1</td>\n",
+       "      <td>pancancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0.77054</td>\n",
+       "      <td>0.54482</td>\n",
+       "      <td>MAP3K1</td>\n",
+       "      <td>BRCA</td>\n",
+       "      <td>signal</td>\n",
+       "      <td>42</td>\n",
+       "      <td>test</td>\n",
+       "      <td>1</td>\n",
+       "      <td>pancancer</td>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     auroc     aupr    gene holdout_cancer_type  signal  seed data_type  fold  \\\n",
+       "0  0.67257  0.30731  MAP3K1                BRCA  signal    42     train     0   \n",
+       "1  0.64775  0.30118  MAP3K1                BRCA  signal    42      test     0   \n",
+       "2  0.64893  0.21980  MAP3K1                BRCA  signal    42        cv     0   \n",
+       "3  0.99443  0.91075  MAP3K1                BRCA  signal    42     train     1   \n",
+       "4  0.77054  0.54482  MAP3K1                BRCA  signal    42      test     1   \n",
+       "\n",
+       "   train_set   identifier  \n",
+       "0  pancancer  MAP3K1_BRCA  \n",
+       "1  pancancer  MAP3K1_BRCA  \n",
+       "2  pancancer  MAP3K1_BRCA  \n",
+       "3  pancancer  MAP3K1_BRCA  \n",
+       "4  pancancer  MAP3K1_BRCA  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pancancer_df1 = au.load_prediction_results(pancancer_dir, 'pancancer')\n",
+    "pancancer_df2 = au.load_prediction_results(pancancer_dir2, 'pancancer')\n",
+    "pancancer_df = pd.concat((pancancer_df1, pancancer_df2))\n",
+    "print(np.unique(pancancer_df.seed))\n",
+    "print(pancancer_df.shape)\n",
+    "pancancer_df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "First, we want to see which gene/cancer type combinations we can use to successfully build a classifier; i.e. which combinations perform significantly better than the negative control with shuffled labels.\n",
+    "\n",
+    "We expect many of these genes to be the same ones that we identified in the stratified analysis, but we're also interested in genes that are identified in this analysis but not the stratified analysis (i.e. genes where we can build a good classifier in a particular cancer type, but the classifier does not generalize well across cancer types)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "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>identifier</th>\n",
+       "      <th>delta_mean</th>\n",
+       "      <th>p_value</th>\n",
+       "      <th>corr_pval</th>\n",
+       "      <th>reject_null</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>137</th>\n",
+       "      <td>CTNNB1_UCEC</td>\n",
+       "      <td>0.761315</td>\n",
+       "      <td>1.444065e-16</td>\n",
+       "      <td>5.790699e-14</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>332</th>\n",
+       "      <td>RB1_BLCA</td>\n",
+       "      <td>0.657051</td>\n",
+       "      <td>1.333908e-15</td>\n",
+       "      <td>2.674486e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>331</th>\n",
+       "      <td>PTEN_UCEC</td>\n",
+       "      <td>0.762998</td>\n",
+       "      <td>4.466359e-15</td>\n",
+       "      <td>5.970033e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>79</th>\n",
+       "      <td>BRAF_THCA</td>\n",
+       "      <td>0.738155</td>\n",
+       "      <td>8.457498e-15</td>\n",
+       "      <td>8.478641e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>201</th>\n",
+       "      <td>IDH1_LGG</td>\n",
+       "      <td>0.439180</td>\n",
+       "      <td>1.168782e-14</td>\n",
+       "      <td>9.373631e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>288</th>\n",
+       "      <td>PBRM1_KIRC</td>\n",
+       "      <td>0.720956</td>\n",
+       "      <td>2.195482e-14</td>\n",
+       "      <td>1.467314e-12</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>56</th>\n",
+       "      <td>ATRX_LGG</td>\n",
+       "      <td>0.710357</td>\n",
+       "      <td>1.178548e-13</td>\n",
+       "      <td>6.751394e-12</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>158</th>\n",
+       "      <td>ERBB2_BRCA</td>\n",
+       "      <td>0.684424</td>\n",
+       "      <td>1.990134e-13</td>\n",
+       "      <td>8.867153e-12</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>APC_COAD</td>\n",
+       "      <td>0.655265</td>\n",
+       "      <td>1.913388e-13</td>\n",
+       "      <td>8.867153e-12</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>ARID1A_UCEC</td>\n",
+       "      <td>0.608521</td>\n",
+       "      <td>2.573409e-13</td>\n",
+       "      <td>1.031937e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      identifier  delta_mean       p_value     corr_pval  reject_null\n",
+       "137  CTNNB1_UCEC    0.761315  1.444065e-16  5.790699e-14         True\n",
+       "332     RB1_BLCA    0.657051  1.333908e-15  2.674486e-13         True\n",
+       "331    PTEN_UCEC    0.762998  4.466359e-15  5.970033e-13         True\n",
+       "79     BRAF_THCA    0.738155  8.457498e-15  8.478641e-13         True\n",
+       "201     IDH1_LGG    0.439180  1.168782e-14  9.373631e-13         True\n",
+       "288   PBRM1_KIRC    0.720956  2.195482e-14  1.467314e-12         True\n",
+       "56      ATRX_LGG    0.710357  1.178548e-13  6.751394e-12         True\n",
+       "158   ERBB2_BRCA    0.684424  1.990134e-13  8.867153e-12         True\n",
+       "8       APC_COAD    0.655265  1.913388e-13  8.867153e-12         True\n",
+       "24   ARID1A_UCEC    0.608521  2.573409e-13  1.031937e-11         True"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "single_cancer_comparison_df = au.compare_results(single_cancer_df,\n",
+    "                                                 identifier='identifier',\n",
+    "                                                 metric='aupr',\n",
+    "                                                 correction=True,\n",
+    "                                                 correction_alpha=0.001,\n",
+    "                                                 verbose=False)\n",
+    "single_cancer_comparison_df.sort_values(by='corr_pval').head(n=10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Train single cancer/test single cancer, Vogelstein et al. cancer genes')"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# volcano plot like in the previous analysis, but for gene/holdout cancer type combinations\n",
+    "# (instead of individual genes as before)\n",
+    "single_cancer_comparison_df['nlog10_p'] = -np.log(single_cancer_comparison_df.corr_pval)\n",
+    "\n",
+    "sns.set({'figure.figsize': (8, 6)})\n",
+    "sns.scatterplot(data=single_cancer_comparison_df, x='delta_mean', y='nlog10_p', hue='reject_null')\n",
+    "plt.xlabel('AUPRC(signal) - AUPRC(shuffled)')\n",
+    "plt.ylabel(r'$-\\log_{10}($adjusted p-value$)$')\n",
+    "plt.title('Train single cancer/test single cancer, Vogelstein et al. cancer genes')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# get genes that are significant in single cancer holdout analysis\n",
+    "sig_ids = single_cancer_comparison_df[\n",
+    "    single_cancer_comparison_df.reject_null\n",
+    "].identifier.values\n",
+    "sig_genes = list(set([id_str.split('_')[0] for id_str in sig_ids]))\n",
+    "\n",
+    "# then get overlap with genes that are significant in stratified analysis\n",
+    "vogelstein_results_df = pd.read_csv(os.path.join(cfg.results_dir, 'vogelstein_stratified_pvals.tsv'),\n",
+    "                                    sep='\\t')\n",
+    "sig_genes_stratified = vogelstein_results_df[\n",
+    "    vogelstein_results_df.reject_null\n",
+    "].identifier.values\n",
+    "\n",
+    "# then plot results in a venn diagram\n",
+    "def get_venn(g1, g2):\n",
+    "    s1, s2 = set(g1), set(g2)\n",
+    "    s_inter = list(s1 & s2)\n",
+    "    s1_only = list(s1 - s2)\n",
+    "    s2_only = list(s2 - s1)\n",
+    "    return ((s1_only, s2_only, s_inter),\n",
+    "            (len(s1_only), len(s2_only), len(s_inter)))\n",
+    "\n",
+    "venn_sets, venn_counts = au.get_venn(sig_genes, sig_genes_stratified)\n",
+    "v = venn2(subsets=venn_counts, set_labels=('Cancer type only', 'Stratified only', 'Both'))\n",
+    "v.get_label_by_id('A').set_y(-0.6)\n",
+    "v.get_label_by_id('B').set_y(-0.6)\n",
+    "v.get_label_by_id('A').set_x(-0.3)\n",
+    "v.get_label_by_id('B').set_x(0.3)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['JAK2']\n"
+     ]
+    },
+    {
+     "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>identifier</th>\n",
+       "      <th>delta_mean</th>\n",
+       "      <th>p_value</th>\n",
+       "      <th>corr_pval</th>\n",
+       "      <th>reject_null</th>\n",
+       "      <th>nlog10_p</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>KIT_TGCT</td>\n",
+       "      <td>0.677021</td>\n",
+       "      <td>2.050943e-10</td>\n",
+       "      <td>3.575775e-09</td>\n",
+       "      <td>True</td>\n",
+       "      <td>19.449084</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>CDH1_BRCA</td>\n",
+       "      <td>0.461297</td>\n",
+       "      <td>2.770618e-09</td>\n",
+       "      <td>3.107181e-08</td>\n",
+       "      <td>True</td>\n",
+       "      <td>17.286965</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>KDM6A_BLCA</td>\n",
+       "      <td>0.387436</td>\n",
+       "      <td>3.100787e-09</td>\n",
+       "      <td>3.272146e-08</td>\n",
+       "      <td>True</td>\n",
+       "      <td>17.235235</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>JAK1_UCEC</td>\n",
+       "      <td>0.470654</td>\n",
+       "      <td>1.324074e-08</td>\n",
+       "      <td>1.206371e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>15.930479</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FUBP1_LGG</td>\n",
+       "      <td>0.571169</td>\n",
+       "      <td>2.049619e-08</td>\n",
+       "      <td>1.677341e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>15.600886</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>CASP8_HNSC</td>\n",
+       "      <td>0.432279</td>\n",
+       "      <td>5.010243e-08</td>\n",
+       "      <td>3.652923e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>14.822568</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>STAG2_BLCA</td>\n",
+       "      <td>0.424916</td>\n",
+       "      <td>1.507026e-07</td>\n",
+       "      <td>9.906845e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.824870</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>STK11_LUAD</td>\n",
+       "      <td>0.391472</td>\n",
+       "      <td>1.663597e-07</td>\n",
+       "      <td>1.042348e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.774035</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>PPP2R1A_UCEC</td>\n",
+       "      <td>0.370095</td>\n",
+       "      <td>1.930102e-07</td>\n",
+       "      <td>1.172683e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.656216</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>SETD2_KIRC</td>\n",
+       "      <td>0.524931</td>\n",
+       "      <td>2.458635e-07</td>\n",
+       "      <td>1.449872e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.444035</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>RNF43_UCEC</td>\n",
+       "      <td>0.302333</td>\n",
+       "      <td>7.493966e-07</td>\n",
+       "      <td>3.954053e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>12.440769</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>MAP3K1_BRCA</td>\n",
+       "      <td>0.363941</td>\n",
+       "      <td>1.150532e-06</td>\n",
+       "      <td>5.840042e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>12.050773</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>SMARCA4_LUAD</td>\n",
+       "      <td>0.323078</td>\n",
+       "      <td>1.267754e-06</td>\n",
+       "      <td>6.354616e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>11.966329</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>KDM5C_KIRC</td>\n",
+       "      <td>0.448056</td>\n",
+       "      <td>1.296445e-06</td>\n",
+       "      <td>6.418205e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>11.956372</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>FGFR2_STAD</td>\n",
+       "      <td>0.369514</td>\n",
+       "      <td>1.585627e-06</td>\n",
+       "      <td>7.660678e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>11.779410</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      identifier  delta_mean       p_value     corr_pval  reject_null  \\\n",
+       "0       KIT_TGCT    0.677021  2.050943e-10  3.575775e-09         True   \n",
+       "1      CDH1_BRCA    0.461297  2.770618e-09  3.107181e-08         True   \n",
+       "2     KDM6A_BLCA    0.387436  3.100787e-09  3.272146e-08         True   \n",
+       "3      JAK1_UCEC    0.470654  1.324074e-08  1.206371e-07         True   \n",
+       "4      FUBP1_LGG    0.571169  2.049619e-08  1.677341e-07         True   \n",
+       "5     CASP8_HNSC    0.432279  5.010243e-08  3.652923e-07         True   \n",
+       "6     STAG2_BLCA    0.424916  1.507026e-07  9.906845e-07         True   \n",
+       "7     STK11_LUAD    0.391472  1.663597e-07  1.042348e-06         True   \n",
+       "8   PPP2R1A_UCEC    0.370095  1.930102e-07  1.172683e-06         True   \n",
+       "9     SETD2_KIRC    0.524931  2.458635e-07  1.449872e-06         True   \n",
+       "10    RNF43_UCEC    0.302333  7.493966e-07  3.954053e-06         True   \n",
+       "11   MAP3K1_BRCA    0.363941  1.150532e-06  5.840042e-06         True   \n",
+       "12  SMARCA4_LUAD    0.323078  1.267754e-06  6.354616e-06         True   \n",
+       "13    KDM5C_KIRC    0.448056  1.296445e-06  6.418205e-06         True   \n",
+       "14    FGFR2_STAD    0.369514  1.585627e-06  7.660678e-06         True   \n",
+       "\n",
+       "     nlog10_p  \n",
+       "0   19.449084  \n",
+       "1   17.286965  \n",
+       "2   17.235235  \n",
+       "3   15.930479  \n",
+       "4   15.600886  \n",
+       "5   14.822568  \n",
+       "6   13.824870  \n",
+       "7   13.774035  \n",
+       "8   13.656216  \n",
+       "9   13.444035  \n",
+       "10  12.440769  \n",
+       "11  12.050773  \n",
+       "12  11.966329  \n",
+       "13  11.956372  \n",
+       "14  11.779410  "
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# print genes (and cancer types in holdout analysis) that are not in the intersection\n",
+    "holdout_only_genes = venn_sets[0]\n",
+    "disjoint_identifiers_df = single_cancer_comparison_df[\n",
+    "    (single_cancer_comparison_df.reject_null) &\n",
+    "    (single_cancer_comparison_df.identifier.str.startswith(tuple(holdout_only_genes)))\n",
+    "].sort_values(by='corr_pval').reset_index(drop=True)\n",
+    "print(venn_sets[1])\n",
+    "disjoint_identifiers_df.head(n=15)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now we can do the same thing using pan-cancer data (i.e. train the model on all cancer types, and test on held-out data from a single cancer type). These results should be similar to the single-cancer results, but due to the additional data we should be better powered to identify certain effects (e.g. in relatively rare cancer types) and other effects (single cancer-specific effects) should be washed out due to the addition of the pan-cancer data.\n",
+    "\n",
+    "We'll attempt to identify these later."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "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>identifier</th>\n",
+       "      <th>delta_mean</th>\n",
+       "      <th>p_value</th>\n",
+       "      <th>corr_pval</th>\n",
+       "      <th>reject_null</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>122</th>\n",
+       "      <td>CTNNB1_UCEC</td>\n",
+       "      <td>0.747696</td>\n",
+       "      <td>2.283201e-16</td>\n",
+       "      <td>8.813158e-14</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>309</th>\n",
+       "      <td>PTEN_UCEC</td>\n",
+       "      <td>0.749178</td>\n",
+       "      <td>1.576221e-15</td>\n",
+       "      <td>3.042106e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>APC_READ</td>\n",
+       "      <td>0.752628</td>\n",
+       "      <td>7.547138e-15</td>\n",
+       "      <td>9.710652e-13</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>51</th>\n",
+       "      <td>ATRX_LGG</td>\n",
+       "      <td>0.735297</td>\n",
+       "      <td>2.395309e-13</td>\n",
+       "      <td>2.092726e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>107</th>\n",
+       "      <td>CIC_LGG</td>\n",
+       "      <td>0.678060</td>\n",
+       "      <td>2.710785e-13</td>\n",
+       "      <td>2.092726e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>146</th>\n",
+       "      <td>ERBB2_ESCA</td>\n",
+       "      <td>0.761216</td>\n",
+       "      <td>3.625853e-13</td>\n",
+       "      <td>2.332632e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>143</th>\n",
+       "      <td>ERBB2_BRCA</td>\n",
+       "      <td>0.649181</td>\n",
+       "      <td>5.474466e-13</td>\n",
+       "      <td>2.641430e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>71</th>\n",
+       "      <td>BRAF_THCA</td>\n",
+       "      <td>0.703306</td>\n",
+       "      <td>4.798131e-13</td>\n",
+       "      <td>2.641430e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>305</th>\n",
+       "      <td>PTEN_PRAD</td>\n",
+       "      <td>0.615811</td>\n",
+       "      <td>1.705229e-12</td>\n",
+       "      <td>7.313536e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>132</th>\n",
+       "      <td>EGFR_LGG</td>\n",
+       "      <td>0.717573</td>\n",
+       "      <td>2.567462e-12</td>\n",
+       "      <td>9.910403e-11</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      identifier  delta_mean       p_value     corr_pval  reject_null\n",
+       "122  CTNNB1_UCEC    0.747696  2.283201e-16  8.813158e-14         True\n",
+       "309    PTEN_UCEC    0.749178  1.576221e-15  3.042106e-13         True\n",
+       "11      APC_READ    0.752628  7.547138e-15  9.710652e-13         True\n",
+       "51      ATRX_LGG    0.735297  2.395309e-13  2.092726e-11         True\n",
+       "107      CIC_LGG    0.678060  2.710785e-13  2.092726e-11         True\n",
+       "146   ERBB2_ESCA    0.761216  3.625853e-13  2.332632e-11         True\n",
+       "143   ERBB2_BRCA    0.649181  5.474466e-13  2.641430e-11         True\n",
+       "71     BRAF_THCA    0.703306  4.798131e-13  2.641430e-11         True\n",
+       "305    PTEN_PRAD    0.615811  1.705229e-12  7.313536e-11         True\n",
+       "132     EGFR_LGG    0.717573  2.567462e-12  9.910403e-11         True"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pancancer_comparison_df = au.compare_results(pancancer_df,\n",
+    "                                             identifier='identifier',\n",
+    "                                             metric='aupr',\n",
+    "                                             correction=True,\n",
+    "                                             correction_alpha=0.001,\n",
+    "                                             verbose=False)\n",
+    "pancancer_comparison_df.sort_values(by='corr_pval').head(n=10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Train pan-cancer/test single cancer, Vogelstein et al. cancer genes')"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# TODO: Venn diagram of genes significant for stratified analysis vs. new genes\n",
+    "pancancer_comparison_df['nlog10_p'] = -np.log(pancancer_comparison_df.corr_pval)\n",
+    "\n",
+    "sns.set({'figure.figsize': (8, 6)})\n",
+    "sns.scatterplot(data=pancancer_comparison_df, x='delta_mean', y='nlog10_p', hue='reject_null')\n",
+    "plt.xlabel('AUPRC(signal) - AUPRC(shuffled)')\n",
+    "plt.ylabel(r'$-\\log_{10}($adjusted p-value$)$')\n",
+    "plt.title('Train pan-cancer/test single cancer, Vogelstein et al. cancer genes')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# get genes that are significant in single cancer holdout analysis\n",
+    "sig_ids = pancancer_comparison_df[\n",
+    "    pancancer_comparison_df.reject_null\n",
+    "].identifier.values\n",
+    "sig_genes = list(set([id_str.split('_')[0] for id_str in sig_ids]))\n",
+    "\n",
+    "# then get overlap with genes that are significant in stratified analysis\n",
+    "sig_genes_stratified = vogelstein_results_df[\n",
+    "    vogelstein_results_df.reject_null\n",
+    "].identifier.values\n",
+    "\n",
+    "# then plot results in a venn diagram\n",
+    "venn_sets, venn_counts = au.get_venn(sig_genes, sig_genes_stratified)\n",
+    "v = venn2(subsets=venn_counts, set_labels=('Cancer type only', 'Stratified only', 'Both')) \n",
+    "v.get_label_by_id('A').set_y(-0.6)\n",
+    "v.get_label_by_id('B').set_y(-0.6)\n",
+    "v.get_label_by_id('A').set_x(-0.3)\n",
+    "v.get_label_by_id('B').set_x(0.35)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[]\n"
+     ]
+    },
+    {
+     "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>identifier</th>\n",
+       "      <th>delta_mean</th>\n",
+       "      <th>p_value</th>\n",
+       "      <th>corr_pval</th>\n",
+       "      <th>reject_null</th>\n",
+       "      <th>nlog10_p</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>CDH1_BRCA</td>\n",
+       "      <td>0.452599</td>\n",
+       "      <td>1.969560e-10</td>\n",
+       "      <td>3.041001e-09</td>\n",
+       "      <td>True</td>\n",
+       "      <td>19.611079</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>NF1_OV</td>\n",
+       "      <td>0.505394</td>\n",
+       "      <td>5.210836e-09</td>\n",
+       "      <td>4.469739e-08</td>\n",
+       "      <td>True</td>\n",
+       "      <td>16.923351</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>RNF43_UCEC</td>\n",
+       "      <td>0.309720</td>\n",
+       "      <td>8.866017e-09</td>\n",
+       "      <td>6.984250e-08</td>\n",
+       "      <td>True</td>\n",
+       "      <td>16.477023</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>NF1_PCPG</td>\n",
+       "      <td>0.615230</td>\n",
+       "      <td>1.174347e-08</td>\n",
+       "      <td>9.065961e-08</td>\n",
+       "      <td>True</td>\n",
+       "      <td>16.216154</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>CASP8_HNSC</td>\n",
+       "      <td>0.472699</td>\n",
+       "      <td>2.282630e-08</td>\n",
+       "      <td>1.468492e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>15.733860</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>KDM6A_BLCA</td>\n",
+       "      <td>0.336795</td>\n",
+       "      <td>4.160516e-08</td>\n",
+       "      <td>2.549142e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>15.182339</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>H3F3A_BRCA</td>\n",
+       "      <td>0.212826</td>\n",
+       "      <td>1.305036e-07</td>\n",
+       "      <td>6.996445e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>14.172694</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>CARD11_BLCA</td>\n",
+       "      <td>0.384759</td>\n",
+       "      <td>1.846825e-07</td>\n",
+       "      <td>9.633440e-07</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.852855</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>NF1_LGG</td>\n",
+       "      <td>0.502823</td>\n",
+       "      <td>2.893103e-07</td>\n",
+       "      <td>1.431715e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.456638</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>STK11_LUAD</td>\n",
+       "      <td>0.349020</td>\n",
+       "      <td>3.328740e-07</td>\n",
+       "      <td>1.626448e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>13.329112</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>SETD2_KIRC</td>\n",
+       "      <td>0.467686</td>\n",
+       "      <td>6.470631e-07</td>\n",
+       "      <td>2.838254e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>12.772321</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>PPP2R1A_UCEC</td>\n",
+       "      <td>0.389092</td>\n",
+       "      <td>1.110789e-06</td>\n",
+       "      <td>4.610370e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>12.287202</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>SMARCA4_LUAD</td>\n",
+       "      <td>0.388497</td>\n",
+       "      <td>1.490245e-06</td>\n",
+       "      <td>5.992027e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>12.025081</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>TSC1_BLCA</td>\n",
+       "      <td>0.513599</td>\n",
+       "      <td>1.989432e-06</td>\n",
+       "      <td>7.628283e-06</td>\n",
+       "      <td>True</td>\n",
+       "      <td>11.783648</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>NF1_SKCM</td>\n",
+       "      <td>0.334425</td>\n",
+       "      <td>7.815366e-06</td>\n",
+       "      <td>2.668468e-05</td>\n",
+       "      <td>True</td>\n",
+       "      <td>10.531421</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      identifier  delta_mean       p_value     corr_pval  reject_null  \\\n",
+       "0      CDH1_BRCA    0.452599  1.969560e-10  3.041001e-09         True   \n",
+       "1         NF1_OV    0.505394  5.210836e-09  4.469739e-08         True   \n",
+       "2     RNF43_UCEC    0.309720  8.866017e-09  6.984250e-08         True   \n",
+       "3       NF1_PCPG    0.615230  1.174347e-08  9.065961e-08         True   \n",
+       "4     CASP8_HNSC    0.472699  2.282630e-08  1.468492e-07         True   \n",
+       "5     KDM6A_BLCA    0.336795  4.160516e-08  2.549142e-07         True   \n",
+       "6     H3F3A_BRCA    0.212826  1.305036e-07  6.996445e-07         True   \n",
+       "7    CARD11_BLCA    0.384759  1.846825e-07  9.633440e-07         True   \n",
+       "8        NF1_LGG    0.502823  2.893103e-07  1.431715e-06         True   \n",
+       "9     STK11_LUAD    0.349020  3.328740e-07  1.626448e-06         True   \n",
+       "10    SETD2_KIRC    0.467686  6.470631e-07  2.838254e-06         True   \n",
+       "11  PPP2R1A_UCEC    0.389092  1.110789e-06  4.610370e-06         True   \n",
+       "12  SMARCA4_LUAD    0.388497  1.490245e-06  5.992027e-06         True   \n",
+       "13     TSC1_BLCA    0.513599  1.989432e-06  7.628283e-06         True   \n",
+       "14      NF1_SKCM    0.334425  7.815366e-06  2.668468e-05         True   \n",
+       "\n",
+       "     nlog10_p  \n",
+       "0   19.611079  \n",
+       "1   16.923351  \n",
+       "2   16.477023  \n",
+       "3   16.216154  \n",
+       "4   15.733860  \n",
+       "5   15.182339  \n",
+       "6   14.172694  \n",
+       "7   13.852855  \n",
+       "8   13.456638  \n",
+       "9   13.329112  \n",
+       "10  12.772321  \n",
+       "11  12.287202  \n",
+       "12  12.025081  \n",
+       "13  11.783648  \n",
+       "14  10.531421  "
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# print genes (and cancer types in holdout analysis) that are not in the intersection\n",
+    "holdout_only_genes = venn_sets[0]\n",
+    "disjoint_identifiers_df = pancancer_comparison_df[\n",
+    "    (pancancer_comparison_df.reject_null) &\n",
+    "    (pancancer_comparison_df.identifier.str.startswith(tuple(holdout_only_genes)))\n",
+    "].sort_values(by='corr_pval').reset_index(drop=True)\n",
+    "print(venn_sets[1])\n",
+    "disjoint_identifiers_df.head(n=15)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now, we want to compare results when models are trained on single-cancer data to results for the same cancer type, trained on pan-cancer data. That is, we want to identify examples where pan-cancer models perform significantly better than single-cancer models (not just better than the control), and vice-versa - examples where single-cancer models perform significantly better will also be interesting."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "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>identifier</th>\n",
+       "      <th>delta_mean</th>\n",
+       "      <th>p_value</th>\n",
+       "      <th>corr_pval</th>\n",
+       "      <th>reject_null</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>324</th>\n",
+       "      <td>FBXW7_COAD</td>\n",
+       "      <td>0.289451</td>\n",
+       "      <td>0.000003</td>\n",
+       "      <td>0.001214</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>152</th>\n",
+       "      <td>PTEN_BLCA</td>\n",
+       "      <td>0.220198</td>\n",
+       "      <td>0.000008</td>\n",
+       "      <td>0.001378</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>191</th>\n",
+       "      <td>NF1_SKCM</td>\n",
+       "      <td>0.319458</td>\n",
+       "      <td>0.000009</td>\n",
+       "      <td>0.001378</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>146</th>\n",
+       "      <td>SPOP_PRAD</td>\n",
+       "      <td>-0.382461</td>\n",
+       "      <td>0.000050</td>\n",
+       "      <td>0.004330</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>231</th>\n",
+       "      <td>NF1_UCEC</td>\n",
+       "      <td>0.219967</td>\n",
+       "      <td>0.000045</td>\n",
+       "      <td>0.004330</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>177</th>\n",
+       "      <td>NF1_BLCA</td>\n",
+       "      <td>0.184491</td>\n",
+       "      <td>0.000288</td>\n",
+       "      <td>0.020974</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>SMAD4_HNSC</td>\n",
+       "      <td>0.277824</td>\n",
+       "      <td>0.000524</td>\n",
+       "      <td>0.026864</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>95</th>\n",
+       "      <td>FBXW7_UCS</td>\n",
+       "      <td>0.245030</td>\n",
+       "      <td>0.000553</td>\n",
+       "      <td>0.026864</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>340</th>\n",
+       "      <td>ARID1A_COAD</td>\n",
+       "      <td>0.267054</td>\n",
+       "      <td>0.000460</td>\n",
+       "      <td>0.026864</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>429</th>\n",
+       "      <td>PTEN_CESC</td>\n",
+       "      <td>0.258053</td>\n",
+       "      <td>0.000653</td>\n",
+       "      <td>0.028528</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>333</th>\n",
+       "      <td>KDM5C_KIRC</td>\n",
+       "      <td>-0.277978</td>\n",
+       "      <td>0.000838</td>\n",
+       "      <td>0.033290</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>141</th>\n",
+       "      <td>ARID2_LUAD</td>\n",
+       "      <td>0.132200</td>\n",
+       "      <td>0.001325</td>\n",
+       "      <td>0.048257</td>\n",
+       "      <td>True</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>21</th>\n",
+       "      <td>FBXW7_UCEC</td>\n",
+       "      <td>0.244338</td>\n",
+       "      <td>0.002641</td>\n",
+       "      <td>0.080350</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>256</th>\n",
+       "      <td>JAK2_UCEC</td>\n",
+       "      <td>0.319374</td>\n",
+       "      <td>0.002758</td>\n",
+       "      <td>0.080350</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>323</th>\n",
+       "      <td>CDKN2A_SARC</td>\n",
+       "      <td>0.172874</td>\n",
+       "      <td>0.002538</td>\n",
+       "      <td>0.080350</td>\n",
+       "      <td>False</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      identifier  delta_mean   p_value  corr_pval  reject_null\n",
+       "324   FBXW7_COAD    0.289451  0.000003   0.001214         True\n",
+       "152    PTEN_BLCA    0.220198  0.000008   0.001378         True\n",
+       "191     NF1_SKCM    0.319458  0.000009   0.001378         True\n",
+       "146    SPOP_PRAD   -0.382461  0.000050   0.004330         True\n",
+       "231     NF1_UCEC    0.219967  0.000045   0.004330         True\n",
+       "177     NF1_BLCA    0.184491  0.000288   0.020974         True\n",
+       "24    SMAD4_HNSC    0.277824  0.000524   0.026864         True\n",
+       "95     FBXW7_UCS    0.245030  0.000553   0.026864         True\n",
+       "340  ARID1A_COAD    0.267054  0.000460   0.026864         True\n",
+       "429    PTEN_CESC    0.258053  0.000653   0.028528         True\n",
+       "333   KDM5C_KIRC   -0.277978  0.000838   0.033290         True\n",
+       "141   ARID2_LUAD    0.132200  0.001325   0.048257         True\n",
+       "21    FBXW7_UCEC    0.244338  0.002641   0.080350        False\n",
+       "256    JAK2_UCEC    0.319374  0.002758   0.080350        False\n",
+       "323  CDKN2A_SARC    0.172874  0.002538   0.080350        False"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiment_comparison_df = au.compare_results(single_cancer_df,\n",
+    "                                              pancancer_df=pancancer_df,\n",
+    "                                              identifier='identifier',\n",
+    "                                              metric='aupr',\n",
+    "                                              correction=True,\n",
+    "                                              correction_alpha=0.05,\n",
+    "                                              verbose=False)\n",
+    "experiment_comparison_df.sort_values(by='corr_pval').head(n=15)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Comparison of pan-cancer and single-cancer results, Vogelstein genes')"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiment_comparison_df['nlog10_p'] = -np.log(experiment_comparison_df.corr_pval)\n",
+    "\n",
+    "sns.set({'figure.figsize': (8, 6)})\n",
+    "sns.scatterplot(data=experiment_comparison_df, x='delta_mean', y='nlog10_p', hue='reject_null')\n",
+    "plt.xlabel('AUPRC(pancancer) - AUPRC(single cancer)')\n",
+    "plt.ylabel(r'$-\\log_{10}($adjusted p-value$)$')\n",
+    "plt.title('Comparison of pan-cancer and single-cancer results, Vogelstein genes')"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python [conda env:pancancer-evaluation]",
+   "language": "python",
+   "name": "conda-env-pancancer-evaluation-py"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/environment.yml b/environment.yml
index 350ff1f..09dcb90 100644
--- a/environment.yml
+++ b/environment.yml
@@ -7,6 +7,7 @@ dependencies:
   - ipython=7.16.1
   - jupyter_core=4.6.3
   - matplotlib=3.3.0
+  - matplotlib-venn=0.11.5
   - numpy=1.16.5
   - pandas=0.24.2
   - pytest=5.4.3
diff --git a/pancancer_evaluation/utilities/analysis_utilities.py b/pancancer_evaluation/utilities/analysis_utilities.py
index cc78914..66bd043 100644
--- a/pancancer_evaluation/utilities/analysis_utilities.py
+++ b/pancancer_evaluation/utilities/analysis_utilities.py
@@ -3,7 +3,6 @@
 
 import numpy as np
 import pandas as pd
-# TODO should this be a paired t-test?
 from scipy.stats import ttest_ind
 
 def load_prediction_results(results_dir, train_set_descriptor):
@@ -34,19 +33,66 @@ def load_prediction_results(results_dir, train_set_descriptor):
             results_df = pd.concat((results_df, gene_results_df))
     return results_df
 
-def compare_results(results_df,
+def compare_results(single_cancer_df,
+                    pancancer_df=None,
                     identifier='gene',
                     metric='auroc',
                     correction=False,
                     correction_method='fdr_bh',
                     correction_alpha=0.05,
                     verbose=False):
-    """which gene/cancer type combinations beat the negative control baseline?
+    """Compare cross-validation results between two experimental conditions.
 
-    TODO better documentation
+    Main uses for this are comparing an experiment against its negative control
+    (shuffled labels), and for comparing two experimental conditions against
+    one another.
+
+    Note that this currently uses an unpaired t-test to compare results.
+    TODO this could probably use a paired t-test, but need to verify that
+    CV folds are actually the same between runs
+
+    Arguments
+    ---------
+    single_cancer_df (pd.DataFrame): either a single dataframe to compare against
+                                     its negative control, or the single-cancer
+                                     dataframe
+    pancancer_df (pd.DataFrame): if provided, a second dataframe to compare against
+                                 single_cancer_df
+    identifier (str): column to use as the sample identifier
+    metric (str): column to use as the evaluation metric
+    correction (bool): whether or not to use a multiple testing correction
+    correction_method (str): which method to use for multiple testing correction
+                             (from options in statsmodels.stats.multitest)
+    correction_alpha (float): significance cutoff to use
+    verbose (bool): if True, print verbose output to stderr
+
+    Returns
+    -------
+    results_df (pd.DataFrame): identifiers and results of statistical test
     """
+    if pancancer_df is None:
+        results_df = compare_control(single_cancer_df, identifier, metric, verbose)
+    else:
+        results_df = compare_experiment(single_cancer_df, pancancer_df,
+                                        identifier, metric, verbose)
+    if correction:
+        from statsmodels.stats.multitest import multipletests
+        corr = multipletests(results_df['p_value'],
+                             alpha=correction_alpha,
+                             method=correction_method)
+        results_df = results_df.assign(corr_pval=corr[1], reject_null=corr[0])
+
+    return results_df
+
+
+def compare_control(results_df,
+                    identifier='gene',
+                    metric='auroc',
+                    verbose=False):
+
     results = []
     unique_identifiers = np.unique(results_df[identifier].values)
+
     for id_str in unique_identifiers:
 
         conditions = ((results_df[identifier] == id_str) &
@@ -65,18 +111,81 @@ def compare_results(results_df,
                       file=sys.stderr)
             continue
 
+        if (signal_results.size == 0) or (shuffled_results.size == 0):
+            if verbose:
+                print('size 0 results array for {}, skipping'.format(id_str),
+                      file=sys.stderr)
+            continue
+
         delta_mean = np.mean(signal_results) - np.mean(shuffled_results)
         p_value = ttest_ind(signal_results, shuffled_results)[1]
         results.append([id_str, delta_mean, p_value])
 
-    results_df =  pd.DataFrame(results, columns=['identifier', 'delta_mean', 'p_value'])
+    return pd.DataFrame(results, columns=['identifier', 'delta_mean', 'p_value'])
 
-    if correction:
-        from statsmodels.stats.multitest import multipletests
-        corr = multipletests(results_df['p_value'],
-                             alpha=correction_alpha,
-                             method=correction_method)
-        results_df = results_df.assign(corr_pval=corr[1], reject_null=corr[0])
 
-    return results_df
+def compare_experiment(single_cancer_df,
+                       pancancer_df,
+                       identifier='gene',
+                       metric='auroc',
+                       verbose=False):
+
+    results = []
+    single_cancer_ids = np.unique(single_cancer_df[identifier].values)
+    pancancer_ids = np.unique(pancancer_df[identifier].values)
+    unique_identifiers = list(set(single_cancer_ids).intersection(pancancer_ids))
+
+    for id_str in unique_identifiers:
+
+        conditions = ((single_cancer_df[identifier] == id_str) &
+                      (single_cancer_df.data_type == 'test') &
+                      (single_cancer_df.signal == 'signal'))
+        single_cancer_results = single_cancer_df[conditions][metric].values
+
+        conditions = ((pancancer_df[identifier] == id_str) &
+                      (pancancer_df.data_type == 'test') &
+                      (pancancer_df.signal == 'signal'))
+        pancancer_results = pancancer_df[conditions][metric].values
+
+        if single_cancer_results.shape != pancancer_results.shape:
+            if verbose:
+                print('shapes unequal for {}, skipping'.format(id_str),
+                      file=sys.stderr)
+            continue
+
+        if (single_cancer_results.size == 0) or (pancancer_results.size == 0):
+            if verbose:
+                print('size 0 results array for {}, skipping'.format(id_str),
+                      file=sys.stderr)
+            continue
+
+        delta_mean = np.mean(pancancer_results) - np.mean(single_cancer_results)
+        p_value = ttest_ind(single_cancer_results, pancancer_results)[1]
+        results.append([id_str, delta_mean, p_value])
+
+    return pd.DataFrame(results, columns=['identifier', 'delta_mean', 'p_value'])
+
+
+def get_venn(g1, g2):
+    """Given 2 sets, calculate the intersection/disjoint union.
+
+    Output is formatted to work with matplotlib_venn.
+
+    Arguments
+    ---------
+    g1 (list): list of genes, or any strings
+    g2 (list): second list of genes, or any strings
+
+    Returns
+    -------
+    venn_sets (tuple): objects only in g1, objects only in g2, objects in both
+                       (in that order)
+    venn_counts (tuple): lengths of above sets
+    """
+    s1, s2 = set(g1), set(g2)
+    s_inter = list(s1 & s2)
+    s1_only = list(s1 - s2)
+    s2_only = list(s2 - s1)
+    return ((s1_only, s2_only, s_inter),
+            (len(s1_only), len(s2_only), len(s_inter)))