train_mutation_map.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "fd00779f-6743-4f90-b3ca-677c9460efa0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import wolf\n",
    "from workflow import dig_workflow\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f68a34d6-c5b1-4773-ba45-bd7d27ce0d24",
   "metadata": {},
   "source": [
    "# Reading hg19, DIG formatted MAF file "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7e155c08-09aa-47ef-946d-814f3e0a5d89",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('TCGA_WGS_UVM.validated.maf', sep='\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e91231e2-4204-4b6f-8398-ccf36fa6f7fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         TCGA-RZ-AB0B-01A\n",
       "1         TCGA-RZ-AB0B-01A\n",
       "2         TCGA-RZ-AB0B-01A\n",
       "3         TCGA-RZ-AB0B-01A\n",
       "4         TCGA-RZ-AB0B-01A\n",
       "                ...       \n",
       "159664    TCGA-YZ-A985-01A\n",
       "159665    TCGA-YZ-A985-01A\n",
       "159666    TCGA-YZ-A985-01A\n",
       "159667    TCGA-YZ-A985-01A\n",
       "159668    TCGA-YZ-A985-01A\n",
       "Name: Tumor_Sample_Barcode, Length: 159669, dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Tumor_Sample_Barcode']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fa13d2c-2175-4d3a-a2c0-0463caf0990b",
   "metadata": {},
   "source": [
    "# Mutation burden across samples within the cohort"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d1d8eb7f-9350-4db2-b6e3-c3bfb09b23a6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TCGA-YZ-A985-01A: 26385 number of mutations\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(15, 5))\n",
    "col_sids = 'Tumor_Sample_Barcode'\n",
    "ax.hist(df[col_sids], bins=df[col_sids].unique().shape[0])\n",
    "ax.tick_params(axis='x', labelrotation=90)\n",
    "ids, cts = np.unique(df[col_sids], return_counts=True)\n",
    "idx_max = np.argsort(cts)[-1]\n",
    "print(ids[idx_max] + ': ' + str(cts[idx_max]) + ' number of mutations')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e488eb3-3161-4b61-bea5-607bf31c899e",
   "metadata": {},
   "source": [
    "# Excluding outlier from training cohort"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5d386452-77ea-4ea2-aca7-897fafb21a6c",
   "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>Hugo_Symbol</th>\n",
       "      <th>Entrez_Gene_Id</th>\n",
       "      <th>Center</th>\n",
       "      <th>NCBI_Build</th>\n",
       "      <th>Chromosome</th>\n",
       "      <th>Start_position</th>\n",
       "      <th>End_position</th>\n",
       "      <th>Strand</th>\n",
       "      <th>Variant_Classification</th>\n",
       "      <th>Variant_Type</th>\n",
       "      <th>...</th>\n",
       "      <th>dbSNP_VP</th>\n",
       "      <th>dbSNP_WGT</th>\n",
       "      <th>dbSNP_WTD</th>\n",
       "      <th>dbSNP_dbSNPBuildID</th>\n",
       "      <th>dbSNP_ID</th>\n",
       "      <th>dbSNP_FILTER</th>\n",
       "      <th>HGNC_Entrez_Gene_ID(supplied_by_NCBI)</th>\n",
       "      <th>dbSNP_RSPOS</th>\n",
       "      <th>dbSNP_VLD</th>\n",
       "      <th>SOMATIC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TTLL10</td>\n",
       "      <td>254173.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chr1</td>\n",
       "      <td>1171734</td>\n",
       "      <td>1171734</td>\n",
       "      <td>+</td>\n",
       "      <td>5'Flank</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>254173.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LRRC47</td>\n",
       "      <td>57470.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chr1</td>\n",
       "      <td>3801184</td>\n",
       "      <td>3801184</td>\n",
       "      <td>+</td>\n",
       "      <td>5'Flank</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>57470.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chr1</td>\n",
       "      <td>4881419</td>\n",
       "      <td>4881419</td>\n",
       "      <td>+</td>\n",
       "      <td>IGR</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CAMTA1</td>\n",
       "      <td>23261.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chr1</td>\n",
       "      <td>7487512</td>\n",
       "      <td>7487512</td>\n",
       "      <td>+</td>\n",
       "      <td>Intron</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23261.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PER3</td>\n",
       "      <td>8863.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chr1</td>\n",
       "      <td>7798053</td>\n",
       "      <td>7798053</td>\n",
       "      <td>+</td>\n",
       "      <td>Intron</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>0x050000080005000002000100</td>\n",
       "      <td>1</td>\n",
       "      <td>false</td>\n",
       "      <td>151</td>\n",
       "      <td>rs1377387790</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8863.0</td>\n",
       "      <td>7798053</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133279</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chrX</td>\n",
       "      <td>147314488</td>\n",
       "      <td>147314488</td>\n",
       "      <td>+</td>\n",
       "      <td>IGR</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>0x050000000005000002000100</td>\n",
       "      <td>1</td>\n",
       "      <td>false</td>\n",
       "      <td>151</td>\n",
       "      <td>rs1278307797</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>147314488</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133280</th>\n",
       "      <td>AFF2</td>\n",
       "      <td>2334.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chrX</td>\n",
       "      <td>148547207</td>\n",
       "      <td>148547207</td>\n",
       "      <td>+</td>\n",
       "      <td>Intron</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>0x050100080005000002000100</td>\n",
       "      <td>1</td>\n",
       "      <td>false</td>\n",
       "      <td>92</td>\n",
       "      <td>rs2016745</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2334.0</td>\n",
       "      <td>148547207</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133281</th>\n",
       "      <td>AF274853.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chrX</td>\n",
       "      <td>151362837</td>\n",
       "      <td>151362837</td>\n",
       "      <td>+</td>\n",
       "      <td>RNA</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>0x050000000005000002000100</td>\n",
       "      <td>1</td>\n",
       "      <td>false</td>\n",
       "      <td>151</td>\n",
       "      <td>rs1336421296</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>151362837</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133282</th>\n",
       "      <td>PASD1</td>\n",
       "      <td>139135.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chrX</td>\n",
       "      <td>151665924</td>\n",
       "      <td>151665924</td>\n",
       "      <td>+</td>\n",
       "      <td>Intron</td>\n",
       "      <td>SNP</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>139135.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133283</th>\n",
       "      <td>PNMA5</td>\n",
       "      <td>114824.0</td>\n",
       "      <td>UNKNOWN</td>\n",
       "      <td>hg38</td>\n",
       "      <td>chrX</td>\n",
       "      <td>152993008</td>\n",
       "      <td>152993008</td>\n",
       "      <td>+</td>\n",
       "      <td>Intron</td>\n",
       "      <td>DEL</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>114824.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>false</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>133284 rows × 193 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Hugo_Symbol  Entrez_Gene_Id   Center NCBI_Build Chromosome  \\\n",
       "0           TTLL10        254173.0  UNKNOWN       hg38       chr1   \n",
       "1           LRRC47         57470.0  UNKNOWN       hg38       chr1   \n",
       "2          Unknown             NaN  UNKNOWN       hg38       chr1   \n",
       "3           CAMTA1         23261.0  UNKNOWN       hg38       chr1   \n",
       "4             PER3          8863.0  UNKNOWN       hg38       chr1   \n",
       "...            ...             ...      ...        ...        ...   \n",
       "133279     Unknown             NaN  UNKNOWN       hg38       chrX   \n",
       "133280        AFF2          2334.0  UNKNOWN       hg38       chrX   \n",
       "133281  AF274853.1             NaN  UNKNOWN       hg38       chrX   \n",
       "133282       PASD1        139135.0  UNKNOWN       hg38       chrX   \n",
       "133283       PNMA5        114824.0  UNKNOWN       hg38       chrX   \n",
       "\n",
       "        Start_position  End_position Strand Variant_Classification  \\\n",
       "0              1171734       1171734      +                5'Flank   \n",
       "1              3801184       3801184      +                5'Flank   \n",
       "2              4881419       4881419      +                    IGR   \n",
       "3              7487512       7487512      +                 Intron   \n",
       "4              7798053       7798053      +                 Intron   \n",
       "...                ...           ...    ...                    ...   \n",
       "133279       147314488     147314488      +                    IGR   \n",
       "133280       148547207     148547207      +                 Intron   \n",
       "133281       151362837     151362837      +                    RNA   \n",
       "133282       151665924     151665924      +                 Intron   \n",
       "133283       152993008     152993008      +                 Intron   \n",
       "\n",
       "       Variant_Type  ...                    dbSNP_VP dbSNP_WGT dbSNP_WTD  \\\n",
       "0               SNP  ...                         NaN       NaN     false   \n",
       "1               SNP  ...                         NaN       NaN     false   \n",
       "2               SNP  ...                         NaN       NaN     false   \n",
       "3               SNP  ...                         NaN       NaN     false   \n",
       "4               SNP  ...  0x050000080005000002000100         1     false   \n",
       "...             ...  ...                         ...       ...       ...   \n",
       "133279          SNP  ...  0x050000000005000002000100         1     false   \n",
       "133280          SNP  ...  0x050100080005000002000100         1     false   \n",
       "133281          SNP  ...  0x050000000005000002000100         1     false   \n",
       "133282          SNP  ...                         NaN       NaN     false   \n",
       "133283          DEL  ...                         NaN       NaN     false   \n",
       "\n",
       "       dbSNP_dbSNPBuildID      dbSNP_ID dbSNP_FILTER  \\\n",
       "0                     NaN           NaN          NaN   \n",
       "1                     NaN           NaN          NaN   \n",
       "2                     NaN           NaN          NaN   \n",
       "3                     NaN           NaN          NaN   \n",
       "4                     151  rs1377387790          NaN   \n",
       "...                   ...           ...          ...   \n",
       "133279                151  rs1278307797          NaN   \n",
       "133280                 92     rs2016745          NaN   \n",
       "133281                151  rs1336421296          NaN   \n",
       "133282                NaN           NaN          NaN   \n",
       "133283                NaN           NaN          NaN   \n",
       "\n",
       "       HGNC_Entrez_Gene_ID(supplied_by_NCBI) dbSNP_RSPOS dbSNP_VLD SOMATIC  \n",
       "0                                   254173.0         NaN     false    True  \n",
       "1                                    57470.0         NaN     false    True  \n",
       "2                                        NaN         NaN     false    True  \n",
       "3                                    23261.0         NaN     false    True  \n",
       "4                                     8863.0     7798053     false    True  \n",
       "...                                      ...         ...       ...     ...  \n",
       "133279                                   NaN   147314488     false    True  \n",
       "133280                                2334.0   148547207     false    True  \n",
       "133281                                   NaN   151362837     false    True  \n",
       "133282                              139135.0         NaN     false    True  \n",
       "133283                              114824.0         NaN     false    True  \n",
       "\n",
       "[133284 rows x 193 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.loc[~df[col_sids].isin([ids[idx_max]])]\n",
    "df.to_csv('TCGA_WGS_UVM.maf', sep='\\t')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fd599f8e-93e2-477a-9d10-4cbb3af41344",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh8AAAGdCAYAAACyzRGfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAi+0lEQVR4nO3de1TUdeL/8dcoMoICCYrKBmrlWl7CS+pSfU2TI5KpXXYr1y1W28qiyGhNqbWyG1bnmG0Ztm1pe7rYZdPcTMu81olMVFK3IjFUytAtc0YwJ5T3748O82sEBOozbxh8Ps6Zc5rP5z3zeb/7MPFsLozLGGMEAABgSaumngAAADi5EB8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwKqypJ3C8qqoq7d27V1FRUXK5XE09HQAA0ADGGB06dEgJCQlq1erEz200u/jYu3evEhMTm3oaAADgFygtLdWpp556wjHNLj6ioqIk/TT56OjoJp4NAABoCK/Xq8TERP/v8RNpdvFR/VJLdHQ08QEAQIhpyFsmeMMpAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYFVYU08ALVP3GcuaegqNtmv2mKaeAgCcFHjmAwAAWEV8AAAAq4gPAABgFfEBAACsIj4AAIBVxAcAALCK+AAAAFYRHwAAwCriAwAAWEV8AAAAq4gPAABgFfEBAACsIj4AAIBVxAcAALCK+AAAAFYRHwAAwKpGx8f69es1duxYJSQkyOVyacmSJTXGfPbZZxo3bpxiYmLUrl07DR48WHv27HFivgAAIMQ1Oj4qKiqUnJysefPm1bp/586dOv/883XmmWdq7dq12rp1q2bOnKm2bdv+6skCAIDQF9bYG6Snpys9Pb3O/XfddZcuuugiPfLII/5tp59++i+bHQAAaHEcfc9HVVWVli1bpt/+9rdKS0tTfHy8hg4dWutLM9V8Pp+8Xm/ABQAAtFyOxsf+/ftVXl6u2bNna/To0Xr33Xd16aWX6rLLLtO6detqvU1ubq5iYmL8l8TERCenBAAAmhnHn/mQpPHjx+u2225T//79NWPGDF188cWaP39+rbfJycmRx+PxX0pLS52cEgAAaGYa/Z6PE+nYsaPCwsLUu3fvgO1nnXWWPvjgg1pv43a75Xa7nZwGAABoxhx95iM8PFyDBw9WUVFRwPYvvvhC3bp1c/JQAAAgRDX6mY/y8nIVFxf7r5eUlKiwsFCxsbFKSkrStGnTdOWVV2rYsGEaMWKEVqxYof/85z9au3atk/MGAAAhqtHxUVBQoBEjRvivZ2dnS5IyMjK0cOFCXXrppZo/f75yc3OVlZWlXr166d///rfOP/9852YNAABCVqPjY/jw4TLGnHDM5MmTNXny5F88KQAA0HLx3S4AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVY2Oj/Xr12vs2LFKSEiQy+XSkiVL6hw7ZcoUuVwuzZ0791dMEQAAtCSNjo+KigolJydr3rx5Jxy3ePFiffTRR0pISPjFkwMAAC1PWGNvkJ6ervT09BOO+frrr3XLLbfonXfe0ZgxY37x5AAAQMvT6PioT1VVla6++mpNmzZNffr0qXe8z+eTz+fzX/d6vU5PCQAANCOOv+H04YcfVlhYmLKysho0Pjc3VzExMf5LYmKi01MCAADNiKPxsWnTJj3++ONauHChXC5Xg26Tk5Mjj8fjv5SWljo5JQAA0Mw4Gh/vv/++9u/fr6SkJIWFhSksLEy7d+/W7bffru7du9d6G7fbrejo6IALAABouRx9z8fVV1+t1NTUgG1paWm6+uqrNWnSJCcPBQAAQlSj46O8vFzFxcX+6yUlJSosLFRsbKySkpIUFxcXML5Nmzbq0qWLevXq9etnCwAAQl6j46OgoEAjRozwX8/OzpYkZWRkaOHChY5NDAAAtEyNjo/hw4fLGNPg8bt27WrsIQAAQAvGd7sAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAqkbHx/r16zV27FglJCTI5XJpyZIl/n2VlZWaPn26+vXrp3bt2ikhIUHXXHON9u7d6+ScAQBACGt0fFRUVCg5OVnz5s2rse/w4cPavHmzZs6cqc2bN+uNN95QUVGRxo0b58hkAQBA6Atr7A3S09OVnp5e676YmBitXLkyYNuTTz6pIUOGaM+ePUpKSvplswQAAC1Go+OjsTwej1wul0455ZRa9/t8Pvl8Pv91r9cb7CkBAIAmFNQ3nB45ckTTp0/XhAkTFB0dXeuY3NxcxcTE+C+JiYnBnBIAAGhiQYuPyspKXXHFFTLGKC8vr85xOTk58ng8/ktpaWmwpgQAAJqBoLzsUh0eu3fv1urVq+t81kOS3G633G53MKYBAACaIcfjozo8duzYoTVr1iguLs7pQwAAgBDW6PgoLy9XcXGx/3pJSYkKCwsVGxurrl276ve//702b96st956S8eOHVNZWZkkKTY2VuHh4c7NHAAAhKRGx0dBQYFGjBjhv56dnS1JysjI0L333qulS5dKkvr37x9wuzVr1mj48OG/fKYAAKBFaHR8DB8+XMaYOvefaB8AAADf7QIAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABY1ej4WL9+vcaOHauEhAS5XC4tWbIkYL8xRnfffbe6du2qiIgIpaamaseOHU7NFwAAhLhGx0dFRYWSk5M1b968Wvc/8sgj+vvf/6758+drw4YNateundLS0nTkyJFfPVkAABD6whp7g/T0dKWnp9e6zxijuXPn6m9/+5vGjx8vSfrXv/6lzp07a8mSJbrqqqt+3WwBAEDIc/Q9HyUlJSorK1Nqaqp/W0xMjIYOHar8/Pxab+Pz+eT1egMuAACg5Wr0Mx8nUlZWJknq3LlzwPbOnTv79x0vNzdXs2bNcnIawC/Sfcaypp5Co+2aPaappwAAjdbkn3bJycmRx+PxX0pLS5t6SgAAIIgcjY8uXbpIkvbt2xewfd++ff59x3O73YqOjg64AACAlsvR+OjRo4e6dOmiVatW+bd5vV5t2LBBKSkpTh4KAACEqEa/56O8vFzFxcX+6yUlJSosLFRsbKySkpI0depUPfDAA+rZs6d69OihmTNnKiEhQZdccomT8wYAACGq0fFRUFCgESNG+K9nZ2dLkjIyMrRw4ULdcccdqqio0PXXX6+DBw/q/PPP14oVK9S2bVvnZg0AAEKWyxhjmnoSP+f1ehUTEyOPx8P7P0JYKH5yJBTxaRcAzUVjfn83+addAADAyYX4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGBVWFNPAPXj6+lRl1D82dg1e0xTTwFAE+OZDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFjleHwcO3ZMM2fOVI8ePRQREaHTTz9d999/v4wxTh8KAACEoDCn7/Dhhx9WXl6enn/+efXp00cFBQWaNGmSYmJilJWV5fThAABAiHE8Pj788EONHz9eY8aMkSR1795dL7/8sj7++GOnDwUAAEKQ4y+7nHvuuVq1apW++OILSdInn3yiDz74QOnp6U4fCgAAhCDHn/mYMWOGvF6vzjzzTLVu3VrHjh3Tgw8+qIkTJ9Y63ufzyefz+a97vV6npwQAAJoRx+Pj1Vdf1YsvvqiXXnpJffr0UWFhoaZOnaqEhARlZGTUGJ+bm6tZs2Y5PQ0AzVT3Gcuaegq/yK7ZY5p6CkCL4TIOfwwlMTFRM2bMUGZmpn/bAw88oBdeeEGff/55jfG1PfORmJgoj8ej6OhoJ6cWskL1P9ZAS0J8ACfm9XoVExPToN/fjj/zcfjwYbVqFfhWktatW6uqqqrW8W63W2632+lpAACAZsrx+Bg7dqwefPBBJSUlqU+fPtqyZYvmzJmjyZMnO30oAAAQghyPjyeeeEIzZ87UTTfdpP379yshIUE33HCD7r77bqcPBQAAQpDj8REVFaW5c+dq7ty5Tt81AABoAfhuFwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMCqsKaegG3dZyxr6ikAAHBS45kPAABgFfEBAACsIj4AAIBVxAcAALCK+AAAAFYRHwAAwCriAwAAWEV8AAAAq4gPAABgFfEBAACsIj4AAIBVxAcAALCK+AAAAFYRHwAAwCriAwAAWEV8AAAAq4ISH19//bX+9Kc/KS4uThEREerXr58KCgqCcSgAABBiwpy+w++//17nnXeeRowYoeXLl6tTp07asWOHOnTo4PShAABACHI8Ph5++GElJiZqwYIF/m09evRw+jAAACBEOf6yy9KlS3XOOefoD3/4g+Lj4zVgwAA988wzdY73+Xzyer0BFwAA0HI5Hh9ffvml8vLy1LNnT73zzju68cYblZWVpeeff77W8bm5uYqJifFfEhMTnZ4SAABoRlzGGOPkHYaHh+ucc87Rhx9+6N+WlZWljRs3Kj8/v8Z4n88nn8/nv+71epWYmCiPx6Po6GgnpyZJ6j5jmeP3CaDl2zV7TFNPAWjWvF6vYmJiGvT72/FnPrp27arevXsHbDvrrLO0Z8+eWse73W5FR0cHXAAAQMvleHycd955KioqCtj2xRdfqFu3bk4fCgAAhCDH4+O2227TRx99pIceekjFxcV66aWX9I9//EOZmZlOHwoAAIQgx+Nj8ODBWrx4sV5++WX17dtX999/v+bOnauJEyc6fSgAABCCHP87H5J08cUX6+KLLw7GXQMAgBDHd7sAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVhEfAADAKuIDAABYRXwAAACriA8AAGAV8QEAAKwiPgAAgFXEBwAAsIr4AAAAVgU9PmbPni2Xy6WpU6cG+1AAACAEBDU+Nm7cqKefflpnn312MA8DAABCSNDio7y8XBMnTtQzzzyjDh06BOswAAAgxAQtPjIzMzVmzBilpqaecJzP55PX6w24AACAlissGHe6aNEibd68WRs3bqx3bG5urmbNmhWMaQCAY7rPWNbUUzgp7Jo9pqmnAAscf+ajtLRUt956q1588UW1bdu23vE5OTnyeDz+S2lpqdNTAgAAzYjjz3xs2rRJ+/fv18CBA/3bjh07pvXr1+vJJ5+Uz+dT69at/fvcbrfcbrfT0wAAAM2U4/ExcuRIbdu2LWDbpEmTdOaZZ2r69OkB4QEAAE4+jsdHVFSU+vbtG7CtXbt2iouLq7EdAACcfPgLpwAAwKqgfNrleGvXrrVxGAAAEAJ45gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWOR4fubm5Gjx4sKKiohQfH69LLrlERUVFTh8GAACEKMfjY926dcrMzNRHH32klStXqrKyUqNGjVJFRYXThwIAACEozOk7XLFiRcD1hQsXKj4+Xps2bdKwYcOcPhwAAAgxjsfH8TwejyQpNja21v0+n08+n89/3ev1BntKAACgCQU1PqqqqjR16lSdd9556tu3b61jcnNzNWvWrGBOAwCAoOk+Y1lTT6HRds0e06THD+qnXTIzM7V9+3YtWrSozjE5OTnyeDz+S2lpaTCnBAAAmljQnvm4+eab9dZbb2n9+vU69dRT6xzndrvldruDNQ0AANDMOB4fxhjdcsstWrx4sdauXasePXo4fQgAABDCHI+PzMxMvfTSS3rzzTcVFRWlsrIySVJMTIwiIiKcPhwAAAgxjr/nIy8vTx6PR8OHD1fXrl39l1deecXpQwEAgBAUlJddAAAA6sJ3uwAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWER8AAMAq4gMAAFhFfAAAAKuIDwAAYBXxAQAArCI+AACAVcQHAACwivgAAABWBS0+5s2bp+7du6tt27YaOnSoPv7442AdCgAAhJCgxMcrr7yi7Oxs3XPPPdq8ebOSk5OVlpam/fv3B+NwAAAghAQlPubMmaPrrrtOkyZNUu/evTV//nxFRkbqueeeC8bhAABACAlz+g5//PFHbdq0STk5Of5trVq1UmpqqvLz82uM9/l88vl8/usej0eS5PV6nZ6aJKnKdzgo9wsA+PWC9d/+YArF3yvB+PdcfZ/GmHrHOh4f3377rY4dO6bOnTsHbO/cubM+//zzGuNzc3M1a9asGtsTExOdnhoAoJmLmdvUMzg5BPPf86FDhxQTE3PCMY7HR2Pl5OQoOzvbf72qqkoHDhxQXFycXC5XUI/t9XqVmJio0tJSRUdHB/VYzQVrZs0tFWtmzS1RKK3XGKNDhw4pISGh3rGOx0fHjh3VunVr7du3L2D7vn371KVLlxrj3W633G53wLZTTjnF6WmdUHR0dLM/qU5jzScH1nxyYM0tX6ist75nPKo5/obT8PBwDRo0SKtWrfJvq6qq0qpVq5SSkuL04QAAQIgJyssu2dnZysjI0DnnnKMhQ4Zo7ty5qqio0KRJk4JxOAAAEEKCEh9XXnml/ve//+nuu+9WWVmZ+vfvrxUrVtR4E2pTc7vduueee2q87NOSseaTA2s+ObDmlq+lrtdlGvKZGAAAAIfw3S4AAMAq4gMAAFhFfAAAAKuIDwAAYFXIx8f69es1duxYJSQkyOVyacmSJf59lZWVmj59uvr166d27dopISFB11xzjfbu3RtwH927d5fL5Qq4zJ49O2DM1q1b9X//939q27atEhMT9cgjj9hYXq1OtGZJ+vOf/1xjPaNHjw4Yc+DAAU2cOFHR0dE65ZRTdO2116q8vDxgTCit+fj1Vl8effRR/5hQO8+5ubkaPHiwoqKiFB8fr0suuURFRUUBY44cOaLMzEzFxcWpffv2uvzyy2v8gb89e/ZozJgxioyMVHx8vKZNm6ajR48GjFm7dq0GDhwot9utM844QwsXLgz28mqob70HDhzQLbfcol69eikiIkJJSUnKysryfx9Utdp+DhYtWhQwpjmsV2rYOR4+fHiN9UyZMiVgTKicY6n+Ne/atavOx/Nrr73mHxdK5zkvL09nn322/w+FpaSkaPny5f79Lelx3GAmxL399tvmrrvuMm+88YaRZBYvXuzfd/DgQZOammpeeeUV8/nnn5v8/HwzZMgQM2jQoID76Natm7nvvvvMN99847+Ul5f793s8HtO5c2czceJEs337dvPyyy+biIgI8/TTT9taZoATrdkYYzIyMszo0aMD1nPgwIGAMaNHjzbJycnmo48+Mu+//74544wzzIQJE/z7Q23NP1/rN998Y5577jnjcrnMzp07/WNC7TynpaWZBQsWmO3bt5vCwkJz0UUXmaSkpIA5T5kyxSQmJppVq1aZgoIC87vf/c6ce+65/v1Hjx41ffv2NampqWbLli3m7bffNh07djQ5OTn+MV9++aWJjIw02dnZ5tNPPzVPPPGEad26tVmxYkWzWu+2bdvMZZddZpYuXWqKi4vNqlWrTM+ePc3ll18ecD+SzIIFCwLO8w8//NDs1mtMw87xBRdcYK677rqA9Xg8Hv/+UDrHxtS/5qNHj9Z4PM+aNcu0b9/eHDp0yH8/oXSely5dapYtW2a++OILU1RUZO68807Tpk0bs337dmNMy3ocN1TIx8fP1fZL6Xgff/yxkWR2797t39atWzfz2GOP1Xmbp556ynTo0MH4fD7/tunTp5tevXr92in/anXFx/jx4+u8zaeffmokmY0bN/q3LV++3LhcLvP1118bY0JvzccbP368ufDCCwO2hfJ5NsaY/fv3G0lm3bp1xpif4rpNmzbmtdde84/57LPPjCSTn59vjPkp2lq1amXKysr8Y/Ly8kx0dLR/nXfccYfp06dPwLGuvPJKk5aWFuwlndDx663Nq6++asLDw01lZaV/W30/H811vcbUvuYLLrjA3HrrrXXeJpTPsTENO8/9+/c3kydPDtgWyufZGGM6dOhg/vnPf7b4x3FdQv5ll8byeDxyuVw1vj9m9uzZiouL04ABA/Too48GPJ2Vn5+vYcOGKTw83L8tLS1NRUVF+v77721NvVHWrl2r+Ph49erVSzfeeKO+++47/778/HydcsopOuecc/zbUlNT1apVK23YsME/JtTWXG3fvn1atmyZrr322hr7Qvk8V7+8EBsbK0natGmTKisrlZqa6h9z5plnKikpSfn5+ZJ+WlO/fv0C/sBfWlqavF6v/vvf//rH/Pw+qsdU30dTOX69dY2Jjo5WWFjg30vMzMxUx44dNWTIED333HMBX/HdXNcr1b3mF198UR07dlTfvn2Vk5Ojw4f//1e4h/I5luo/z5s2bVJhYWGtj+dQPM/Hjh3TokWLVFFRoZSUlBb/OK5Lk3+rrU1HjhzR9OnTNWHChIAv6MnKytLAgQMVGxurDz/8UDk5Ofrmm280Z84cSVJZWZl69OgRcF/VPwRlZWXq0KGDvUU0wOjRo3XZZZepR48e2rlzp+68806lp6crPz9frVu3VllZmeLj4wNuExYWptjYWJWVlUkKvTX/3PPPP6+oqChddtllAdtD+TxXVVVp6tSpOu+889S3b1//nMLDw2uEdOfOnQPO4/F/WfjnazrRGK/Xqx9++EERERHBWNIJ1bbe43377be6//77df311wdsv++++3ThhRcqMjJS7777rm666SaVl5crKytLUvNcr1T3mv/4xz+qW7duSkhI0NatWzV9+nQVFRXpjTfekBS651hq2Hl+9tlnddZZZ+ncc88N2B5q53nbtm1KSUnRkSNH1L59ey1evFi9e/dWYWFhi30cn8hJEx+VlZW64oorZIxRXl5ewL7s7Gz/P5999tkKDw/XDTfcoNzc3JD8k7ZXXXWV/5/79euns88+W6effrrWrl2rkSNHNuHM7Hjuuec0ceJEtW3bNmB7KJ/nzMxMbd++XR988EFTT8WK+tbr9Xo1ZswY9e7dW/fee2/AvpkzZ/r/ecCAAaqoqNCjjz7q/6XUXNW15p/HVb9+/dS1a1eNHDlSO3fu1Omnn257mo6q7zz/8MMPeumllwLOabVQO8+9evVSYWGhPB6PXn/9dWVkZGjdunVNPa0mc1K87FIdHrt379bKlSvr/VrioUOH6ujRo9q1a5ckqUuXLjXeeVx9vUuXLkGZs5NOO+00dezYUcXFxZJ+mvP+/fsDxhw9elQHDhzwrydU1/z++++rqKhIf/nLX+odGyrn+eabb9Zbb72lNWvW6NRTT/Vv79Kli3788UcdPHgwYPy+ffsadR7rGhMdHd0k/7dU13qrHTp0SKNHj1ZUVJQWL16sNm3anPD+hg4dqq+++ko+n09S81uvVP+af27o0KGSFPB4DrVzLDVsza+//roOHz6sa665pt77a+7nOTw8XGeccYYGDRqk3NxcJScn6/HHH2+xj+P6tPj4qA6PHTt26L333lNcXFy9tyksLFSrVq38L02kpKRo/fr1qqys9I9ZuXKlevXq1axffqj21Vdf6bvvvlPXrl0l/bSegwcPatOmTf4xq1evVlVVlf8/bKG65meffVaDBg1ScnJyvWOb+3k2xujmm2/W4sWLtXr16hovCQ0aNEht2rTRqlWr/NuKioq0Z88epaSkSPppTdu2bQuIzeoA7927t3/Mz++jekz1fdhS33qln57xGDVqlMLDw7V06dIaz27VprCwUB06dPA/u9Vc1is1bM3HKywslKSAx3OonGOpcWt+9tlnNW7cOHXq1Kne+23O57k2VVVV8vl8Le5x3GBN+GZXRxw6dMhs2bLFbNmyxUgyc+bMMVu2bDG7d+82P/74oxk3bpw59dRTTWFhYcBHsqrfIfzhhx+axx57zBQWFpqdO3eaF154wXTq1Mlcc801/mMcPHjQdO7c2Vx99dVm+/btZtGiRSYyMrLJPoJ5ojUfOnTI/PWvfzX5+fmmpKTEvPfee2bgwIGmZ8+e5siRI/77GD16tBkwYIDZsGGD+eCDD0zPnj0DPmobSmuu5vF4TGRkpMnLy6tx+1A8zzfeeKOJiYkxa9euDfjZPXz4sH/MlClTTFJSklm9erUpKCgwKSkpJiUlxb+/+iN6o0aNMoWFhWbFihWmU6dOtX5Eb9q0aeazzz4z8+bNa5KP6NW3Xo/HY4YOHWr69etniouLA8YcPXrUGPPTRxqfeeYZs23bNrNjxw7z1FNPmcjISHP33Xc3u/U2ZM3FxcXmvvvuMwUFBaakpMS8+eab5rTTTjPDhg3z30coneOGrLnajh07jMvlMsuXL69xH6F2nmfMmGHWrVtnSkpKzNatW82MGTOMy+Uy7777rjGmZT2OGyrk42PNmjVGUo1LRkaGKSkpqXWfJLNmzRpjjDGbNm0yQ4cONTExMaZt27bmrLPOMg899FDAL2pjjPnkk0/M+eefb9xut/nNb35jZs+e3QSr/cmJ1nz48GEzatQo06lTJ9OmTRvTrVs3c9111wV8RMsYY7777jszYcIE0759exMdHW0mTZoU8Bl6Y0JnzdWefvppExERYQ4ePFjj9qF4nuv62V2wYIF/zA8//GBuuukm06FDBxMZGWkuvfRS88033wTcz65du0x6erqJiIgwHTt2NLfffnvAR1ON+enfb//+/U14eLg57bTTAo5hS33rretnQJIpKSkxxvz0kfH+/fub9u3bm3bt2pnk5GQzf/58c+zYsYBjNYf1GlP/mvfs2WOGDRtmYmNjjdvtNmeccYaZNm1awN/5MCZ0zrExDfu5NsaYnJwck5iYWOPcGRN653ny5MmmW7duJjw83HTq1MmMHDnSHx7GtKzHcUO5jPnZZ5MAAACCrMW/5wMAADQvxAcAALCK+AAAAFYRHwAAwCriAwAAWEV8AAAAq4gPAABgFfEBAACsIj4AAIBVxAcAALCK+AAAAFYRHwAAwKr/B6IWZv5yXLOxAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ids, cts = np.unique(df[col_sids], return_counts=True)\n",
    "plt.hist(cts)\n",
    "plt.show()\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(15, 5))\n",
    "ind_sorted = np.argsort(cts)\n",
    "plt.bar(ids[ind_sorted], cts[ind_sorted])\n",
    "ax.tick_params(axis='x', labelrotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "251e81f9-2d2a-4923-9ad4-a67253f7766c",
   "metadata": {},
   "source": [
    "# Running workflow for mutation map training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fa0b0e9c-df8f-4222-9e89-cd00757af562",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[20241009-16:02:43] [prefect] Starting Slurm controller ...\n",
      "[20241009-16:02:43] [prefect] Waiting up to 60 seconds for Slurm controller to start ...\n",
      "[20241009-16:02:56] [prefect] Started Slurm controller.\n",
      "[20241009-16:02:56] [prefect] Workflow results disk low on space (26 GB remaining)\n",
      "[20241009-16:03:02] [prefect] Enqueued workflow uvm_tcga_dig\n",
      "[20241009-16:03:03] [uvm_tcga_dig:DIG_convert_maf] Hashing file TCGA_WGS_UVM.maf; 100/190 MiB completed\n",
      "[20241009-16:03:04] [uvm_tcga_dig:DIG_convert_maf] Localizing inputs...\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_tracks] Job avoidance disabled for this task; overwriting output.\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_tracks] Localizing inputs...\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_ref_fasta] Job avoidance disabled for this task; overwriting output.\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_ref_fasta] Localizing inputs...\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_tracks] Disk name is canine-a88d8673c93a0607e747fcb58805de1e\n",
      "[20241009-16:03:04] [uvm_tcga_dig:DIG_convert_maf] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DIG_convert_maf__2024-10-09--16-03-04_mblxxey_s2nsq0q_cvilttyadla1u\n",
      "[20241009-16:03:04] [uvm_tcga_dig:Localize_ref_fasta] Disk name is canine-5714e9942090819a76b081ec3fd090aa\n",
      "[20241009-16:03:04] [uvm_tcga_dig:DIG_convert_maf] 1 job submitted.\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_tracks] Creating new persistent disk canine-a88d8673c93a0607e747fcb58805de1e\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_tracks] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/Localize_tracks__2024-10-09--16-03-04_mg4xwxy_tbhx1ki_urlzoh5t1wzrw\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_ref_fasta] Creating new persistent disk canine-5714e9942090819a76b081ec3fd090aa\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_ref_fasta] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/Localize_ref_fasta__2024-10-09--16-03-04_mg4xwxy_tbhx1ki_ykhip0lgm0pxq\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_tracks] 1 job submitted.\n",
      "[20241009-16:03:05] [uvm_tcga_dig:Localize_ref_fasta] 1 job submitted.\n",
      "[20241009-16:07:07] [uvm_tcga_dig:Localize_ref_fasta] Finished with status COMPLETED\n",
      "[20241009-16:08:06] [uvm_tcga_dig:DIG_convert_maf] Finished with status COMPLETED\n",
      "[20241009-16:08:06] [uvm_tcga_dig:DIG_annotate_maf] Localizing inputs...\n",
      "[20241009-16:08:06] [uvm_tcga_dig:DIG_annotate_maf] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DIG_annotate_maf__2024-10-09--16-08-06_exnnusq_s2nsq0q_o35wynerh5vvo\n",
      "[20241009-16:08:06] [uvm_tcga_dig:DIG_annotate_maf] 1 job submitted.\n",
      "[20241009-16:09:37] [uvm_tcga_dig:DIG_annotate_maf] Finished with status COMPLETED\n",
      "[20241009-16:13:39] [uvm_tcga_dig:Localize_tracks] Finished with status COMPLETED\n",
      "[20241009-16:13:39] [uvm_tcga_dig:DIG_unzip_h5] Localizing inputs...\n",
      "[20241009-16:13:39] [uvm_tcga_dig:DIG_unzip_h5] Creating new persistent disk canine-scratch-tracks-0\n",
      "[20241009-16:13:39] [uvm_tcga_dig:DIG_unzip_h5] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DIG_unzip_h5__2024-10-09--16-13-39_iui3eoy_s2nsq0q_3jhzsqsvabgqw\n",
      "[20241009-16:13:40] [uvm_tcga_dig:DIG_unzip_h5] 1 job submitted.\n",
      "[20241009-17:44:47] [uvm_tcga_dig:DIG_unzip_h5] Finished with status COMPLETED\n",
      "[20241009-17:44:47] [uvm_tcga_dig:DIG_add_objectives] Localizing inputs...\n",
      "[20241009-17:44:48] [uvm_tcga_dig:DIG_add_objectives] Creating new persistent disk canine-scratch-tracks-with-objectives-tcga-wgs-uvm-0\n",
      "[20241009-17:44:48] [uvm_tcga_dig:DIG_add_objectives] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DIG_add_objectives__2024-10-09--17-44-47_vave3cy_s2nsq0q_for2pvrvuagyw\n",
      "[20241009-17:44:48] [uvm_tcga_dig:DIG_add_objectives] 1 job submitted.\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DIG_add_objectives] Finished with status COMPLETED\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DeleteDisk] Job avoidance disabled for this task; overwriting output.\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DeleteDisk] Localizing inputs...\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DeleteDisk] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DeleteDisk__2024-10-09--18-59-49_ask2hla_tbhx1ki_zdxl4tedwdsyu\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DIG_run_kfold_training] Localizing inputs...\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DIG_run_kfold_training] Task staged in /mnt/nfs/workspace/uvm_tcga_dig/DIG_run_kfold_training__2024-10-09--18-59-49_wlngc2q_exikiti_jyycnxxv4capq\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DeleteDisk] 1 job submitted.\n",
      "[20241009-18:59:49] [uvm_tcga_dig:DIG_run_kfold_training] 1 job submitted.\n",
      "[20241009-19:01:50] [uvm_tcga_dig:DeleteDisk] Finished with status COMPLETED\n"
     ]
    }
   ],
   "source": [
    "with wolf.Workflow(workflow = dig_workflow) as w:\n",
    "    w.run(\n",
    "        maf_file = 'DLBCL_WGS_UVM_hg38.maf', # MAF from characterization pipeline        \n",
    "        genewise_interval_set_name = [\n",
    "            \"promoters\",\n",
    "            \"3-prime_UTRs\",\n",
    "            \"5-prime_UTRs\"\n",
    "        ], # the first 3 interval sets must remain these, and in this order, additional sets can be added \n",
    "        genewise_interval_set_bed = [\n",
    "            \"gs://getzlab-workflows-reference_files-oa/hg19/dig/gc19_pc.prom.bed\",\n",
    "            \"gs://getzlab-workflows-reference_files-oa/hg19/dig/gc19_pc.3utr.bed\",\n",
    "            \"gs://getzlab-workflows-reference_files-oa/hg19/dig/gc19_pc.5utr.bed\"\n",
    "        ], # must be consistent with list above\n",
    "        # mutation_map = \"https://cb.csail.mit.edu/DIG/downloads/mutation_maps/Kidney-RCC_SNV_MNV_INDEL.Pretrained.h5\", # mutation map trained for the analyzed cohort\n",
    "        ref_build = \"hg38\", # reference genome for MAF file\n",
    "        RUN_NAME = \"dlbcl_wgs\"\n",
    "    )"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}