{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project PTMs onto ESRP1 knockdown data from MATS (Yang et al, 2016)\n",
    "\n",
    "Here is an example of running PTM-POSE on MATS analysis of RNA sequencing data from ESRP1 knockdown experiments performed by Yang et al, 2016\n",
    "\n",
    "First, let's focus on skipped exon events.\n",
    "\n",
    "## Phase 1: Load the data and initialize PTM-POSE\n",
    " To identify differentially included PTMs as a result of ESRP1 knockdown, we need three layers of information for each splice event: \n",
    "1. Chromosome\n",
    "2. DNA strand\n",
    "2. Start and end coordinates of the event (either hg19 or hg38)\n",
    "\n",
    "Optionally, we can also provide:\n",
    "1. Gene name\n",
    "2. Event ID\n",
    "3. Delta PSI for the event\n",
    "4. Significance of the event\n",
    "\n",
    "With PTM-POSE, we need to indicate where to find this information within the splice data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>geneSymbol</th>\n",
       "      <th>chr</th>\n",
       "      <th>strand</th>\n",
       "      <th>exonStart_0base</th>\n",
       "      <th>exonEnd</th>\n",
       "      <th>meanDeltaPSI</th>\n",
       "      <th>FDR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SPAG9</td>\n",
       "      <td>chr17</td>\n",
       "      <td>-</td>\n",
       "      <td>49053223</td>\n",
       "      <td>49053262</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>chr16</td>\n",
       "      <td>-</td>\n",
       "      <td>24950684</td>\n",
       "      <td>24950918</td>\n",
       "      <td>0.413</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ITGA6</td>\n",
       "      <td>chr2</td>\n",
       "      <td>+</td>\n",
       "      <td>173366499</td>\n",
       "      <td>173366629</td>\n",
       "      <td>-0.361</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KRAS</td>\n",
       "      <td>chr12</td>\n",
       "      <td>-</td>\n",
       "      <td>25368370</td>\n",
       "      <td>25368494</td>\n",
       "      <td>-0.068</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>TCIRG1</td>\n",
       "      <td>chr11</td>\n",
       "      <td>+</td>\n",
       "      <td>67817953</td>\n",
       "      <td>67818131</td>\n",
       "      <td>0.368</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  geneSymbol    chr strand exonStart_0base    exonEnd meanDeltaPSI FDR\n",
       "0      SPAG9  chr17      -        49053223   49053262        0.227   0\n",
       "1   ARHGAP17  chr16      -        24950684   24950918        0.413   0\n",
       "2      ITGA6   chr2      +       173366499  173366629       -0.361   0\n",
       "3       KRAS  chr12      -        25368370   25368494       -0.068   0\n",
       "4     TCIRG1  chr11      +        67817953   67818131        0.368   0"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "SE_data = pd.read_excel('../../ESRP1_data/Yang2016/esrp1_knockdown_data_Yang2016.xlsx', sheet_name='rMATS ESRP KD', header = 2).iloc[0:179]\n",
    "\n",
    "\n",
    "# required column information\n",
    "chromosome_col = 'chr'\n",
    "strand_col = 'strand'\n",
    "region_start_col = 'exonStart_0base'\n",
    "region_end_col = 'exonEnd'\n",
    "\n",
    "# optional column information (None if nothing is provided and will not be appended to the output)\n",
    "gene_col = 'geneSymbol'\n",
    "event_id_col = None   #not in the data\n",
    "dPSI_col = 'meanDeltaPSI'\n",
    "sig_col = 'FDR'\n",
    "\n",
    "#look at the data\n",
    "SE_data[[gene_col, chromosome_col, strand_col, region_start_col, region_end_col, dPSI_col, sig_col]].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The strand can either be provided use '+' and '-' or using 1 and -1 to indicate the forward and reverse strand, the code will convert strand to integer format (-1 or 1) when running.\n",
    "\n",
    "If this is the first time running PTM-POSE, you will need to download ptm_coordinates. If you set save = True, the coordinates will be saved for the future so you do not need to redownload them, but you can also set save = False to avoid saving the coordinates (will take ~60MB of space)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ptm_pose import pose_config\n",
    "pose_config.ptm_coordinates = pose_config.download_ptm_coordinates(save = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Phase 2: Project PTMs onto differentially included regions\n",
    "\n",
    " We can then use the project module of PTM-POSE to identify PTMs that can be found in these regions. This dataset uses the hg19 genome build, so we need to specify this using the 'coordinate_type' parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Translator file not found. Downloading mapping information between UniProt and Gene Names from pybiomart\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Projecting PTMs onto splice events using hg19 coordinates.: 100%|██████████| 179/179 [00:03<00:00, 48.82it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PTMs projection successful (475 identified).\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "from ptm_pose import project\n",
    "\n",
    "splice_data, spliced_ptms = project.project_ptms_onto_splice_events(SE_data, chromosome_col = chromosome_col, strand_col = strand_col, region_start_col = region_start_col, region_end_col = region_end_col, gene_col = gene_col, event_id_col = event_id_col, dPSI_col = dPSI_col, sig_col = sig_col, coordinate_type = 'hg19')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this, there are two outputs:\n",
    "1. The original splice dataframe with additional PTM information added"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>geneSymbol</th>\n",
       "      <th>chr</th>\n",
       "      <th>strand</th>\n",
       "      <th>exonStart_0base</th>\n",
       "      <th>exonEnd</th>\n",
       "      <th>meanDeltaPSI</th>\n",
       "      <th>FDR</th>\n",
       "      <th>PTMs</th>\n",
       "      <th>Number of PTMs Affected</th>\n",
       "      <th>Number of Unique PTM Sites by Position</th>\n",
       "      <th>Event Length</th>\n",
       "      <th>PTM Density (PTMs/bp)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SPAG9</td>\n",
       "      <td>17</td>\n",
       "      <td>-</td>\n",
       "      <td>49053223</td>\n",
       "      <td>49053262</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>16</td>\n",
       "      <td>-</td>\n",
       "      <td>24950684</td>\n",
       "      <td>24950918</td>\n",
       "      <td>0.413</td>\n",
       "      <td>0</td>\n",
       "      <td>Q68EM7_S575.0 (Phosphorylation)/Q68EM7_S570.0 ...</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>234</td>\n",
       "      <td>0.004274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ITGA6</td>\n",
       "      <td>2</td>\n",
       "      <td>+</td>\n",
       "      <td>173366499</td>\n",
       "      <td>173366629</td>\n",
       "      <td>-0.361</td>\n",
       "      <td>0</td>\n",
       "      <td>P23229_Ynan (Phosphorylation)/P23229_Tnan (Pho...</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>130</td>\n",
       "      <td>0.030769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KRAS</td>\n",
       "      <td>12</td>\n",
       "      <td>-</td>\n",
       "      <td>25368370</td>\n",
       "      <td>25368494</td>\n",
       "      <td>-0.068</td>\n",
       "      <td>0</td>\n",
       "      <td>P01116_C186 (Methylation)/P01116_C180 (Palmito...</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>124</td>\n",
       "      <td>0.016129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>TCIRG1</td>\n",
       "      <td>11</td>\n",
       "      <td>+</td>\n",
       "      <td>67817953</td>\n",
       "      <td>67818131</td>\n",
       "      <td>0.368</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>178</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  geneSymbol chr strand exonStart_0base    exonEnd meanDeltaPSI FDR  \\\n",
       "0      SPAG9  17      -        49053223   49053262        0.227   0   \n",
       "1   ARHGAP17  16      -        24950684   24950918        0.413   0   \n",
       "2      ITGA6   2      +       173366499  173366629       -0.361   0   \n",
       "3       KRAS  12      -        25368370   25368494       -0.068   0   \n",
       "4     TCIRG1  11      +        67817953   67818131        0.368   0   \n",
       "\n",
       "                                                PTMs  Number of PTMs Affected  \\\n",
       "0                                                NaN                        0   \n",
       "1  Q68EM7_S575.0 (Phosphorylation)/Q68EM7_S570.0 ...                        6   \n",
       "2  P23229_Ynan (Phosphorylation)/P23229_Tnan (Pho...                        7   \n",
       "3  P01116_C186 (Methylation)/P01116_C180 (Palmito...                        3   \n",
       "4                                                NaN                        0   \n",
       "\n",
       "   Number of Unique PTM Sites by Position Event Length PTM Density (PTMs/bp)  \n",
       "0                                       0           39                   0.0  \n",
       "1                                       1          234              0.004274  \n",
       "2                                       4          130              0.030769  \n",
       "3                                       2          124              0.016129  \n",
       "4                                       0          178                   0.0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "splice_data[[gene_col, chromosome_col, strand_col, region_start_col, region_end_col, dPSI_col, sig_col] + ['PTMs', 'Number of PTMs Affected', 'Number of Unique PTM Sites by Position', 'Event Length', 'PTM Density (PTMs/bp)']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2. New dataframe that has each PTM and additional information about the PTM in its own row"
   ]
  },
  {
   "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>dPSI</th>\n",
       "      <th>Significance</th>\n",
       "      <th>Gene</th>\n",
       "      <th>Source of PTM</th>\n",
       "      <th>UniProtKB Accession</th>\n",
       "      <th>Residue</th>\n",
       "      <th>PTM Position in Canonical Isoform</th>\n",
       "      <th>Gene Location (hg19)</th>\n",
       "      <th>Modification</th>\n",
       "      <th>Modification Class</th>\n",
       "      <th>Proximity to Region Start (bp)</th>\n",
       "      <th>Proximity to Region End (bp)</th>\n",
       "      <th>Proximity to Splice Boundary (bp)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.413</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>Q68EM7-1_S575</td>\n",
       "      <td>Q68EM7</td>\n",
       "      <td>S</td>\n",
       "      <td>575.0</td>\n",
       "      <td>24950686.0</td>\n",
       "      <td>Phosphoserine</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>2.0</td>\n",
       "      <td>232.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.413</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>Q68EM7-1_S570</td>\n",
       "      <td>Q68EM7</td>\n",
       "      <td>S</td>\n",
       "      <td>570.0</td>\n",
       "      <td>24950701.0</td>\n",
       "      <td>Phosphoserine</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>17.0</td>\n",
       "      <td>217.0</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.413</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>Q68EM7-1_S560</td>\n",
       "      <td>Q68EM7</td>\n",
       "      <td>S</td>\n",
       "      <td>560.0</td>\n",
       "      <td>24950731.0</td>\n",
       "      <td>Phosphoserine</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>47.0</td>\n",
       "      <td>187.0</td>\n",
       "      <td>47.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.413</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>Q68EM7-1_S553</td>\n",
       "      <td>Q68EM7</td>\n",
       "      <td>S</td>\n",
       "      <td>553.0</td>\n",
       "      <td>24950752.0</td>\n",
       "      <td>Phosphoserine</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>68.0</td>\n",
       "      <td>166.0</td>\n",
       "      <td>68.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.413</td>\n",
       "      <td>0.0</td>\n",
       "      <td>ARHGAP17</td>\n",
       "      <td>Q68EM7-1_S547</td>\n",
       "      <td>Q68EM7</td>\n",
       "      <td>S</td>\n",
       "      <td>547.0</td>\n",
       "      <td>24950770.0</td>\n",
       "      <td>Phosphoserine</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>86.0</td>\n",
       "      <td>148.0</td>\n",
       "      <td>86.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    dPSI  Significance      Gene  Source of PTM UniProtKB Accession Residue  \\\n",
       "0  0.413           0.0  ARHGAP17  Q68EM7-1_S575              Q68EM7       S   \n",
       "1  0.413           0.0  ARHGAP17  Q68EM7-1_S570              Q68EM7       S   \n",
       "2  0.413           0.0  ARHGAP17  Q68EM7-1_S560              Q68EM7       S   \n",
       "3  0.413           0.0  ARHGAP17  Q68EM7-1_S553              Q68EM7       S   \n",
       "4  0.413           0.0  ARHGAP17  Q68EM7-1_S547              Q68EM7       S   \n",
       "\n",
       "   PTM Position in Canonical Isoform  Gene Location (hg19)   Modification  \\\n",
       "0                              575.0            24950686.0  Phosphoserine   \n",
       "1                              570.0            24950701.0  Phosphoserine   \n",
       "2                              560.0            24950731.0  Phosphoserine   \n",
       "3                              553.0            24950752.0  Phosphoserine   \n",
       "4                              547.0            24950770.0  Phosphoserine   \n",
       "\n",
       "  Modification Class  Proximity to Region Start (bp)  \\\n",
       "0    Phosphorylation                             2.0   \n",
       "1    Phosphorylation                            17.0   \n",
       "2    Phosphorylation                            47.0   \n",
       "3    Phosphorylation                            68.0   \n",
       "4    Phosphorylation                            86.0   \n",
       "\n",
       "   Proximity to Region End (bp)  Proximity to Splice Boundary (bp)  \n",
       "0                         232.0                                2.0  \n",
       "1                         217.0                               17.0  \n",
       "2                         187.0                               47.0  \n",
       "3                         166.0                               68.0  \n",
       "4                         148.0                               86.0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spliced_ptms.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For MATS data, there is also a built in function for running PTM-POSE on MATS data, including all events: "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Phase 3: Identify PTMs with altered flanking sequences as a result of splice events\n",
    "\n",
    "In addition to differential inclusion of PTMs, some PTMs may experience altered flanking sequences. We can use the project module of PTM-POSE to identify PTMs for which this happens. You will need to provide the same layers of information, plus the genomic coordinates of the regions flanking the spliced region."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Sam\\miniconda3\\envs\\testing_pose\\Lib\\site-packages\\Bio\\pairwise2.py:278: BiopythonDeprecationWarning: Bio.pairwise2 has been deprecated, and we intend to remove it in a future release of Biopython. As an alternative, please consider using Bio.Align.PairwiseAligner as a replacement, and contact the Biopython developers if you still need the Bio.pairwise2 module.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "from ptm_pose import flanking_sequences\n",
    "\n",
    "first_flank_start_col = 'firstFlankingES'\n",
    "first_flank_end_col='firstFlankingEE'\n",
    "second_flank_start_col = 'secondFlankingES'\n",
    "second_flank_end_col = 'secondFlankingEE'\n",
    "\n",
    "flanks = flanking_sequences.get_flanking_changes_from_splice_data(SE_data, chromosome_col = chromosome_col, strand_col = strand_col, first_flank_start_col = first_flank_start_col, first_flank_end_col=first_flank_end_col, second_flank_start_col = second_flank_start_col, second_flank_end_col = second_flank_end_col , spliced_region_start_col = region_start_col, spliced_region_end_col = region_end_col, dPSI_col=dPSI_col, sig_col = sig_col, event_id_col = event_id_col, coordinate_type = 'hg19')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Event ID</th>\n",
       "      <th>Source of PTM</th>\n",
       "      <th>Residue</th>\n",
       "      <th>PTM Position in Canonical Isoform</th>\n",
       "      <th>Inclusion Sequence</th>\n",
       "      <th>Exclusion Sequence</th>\n",
       "      <th>Region</th>\n",
       "      <th>Translation Success</th>\n",
       "      <th>Matched</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>P01116-2_T148;P01116-1_T148</td>\n",
       "      <td>T</td>\n",
       "      <td>148</td>\n",
       "      <td>ETSAKtRQESG</td>\n",
       "      <td>ETSAKtRQGC*</td>\n",
       "      <td>Second</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>P01116-1_K147;P01116-2_K147</td>\n",
       "      <td>K</td>\n",
       "      <td>147</td>\n",
       "      <td>IETSAkTRQES</td>\n",
       "      <td>IETSAkTRQGC</td>\n",
       "      <td>Second</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>Q9UPQ0-1_S746</td>\n",
       "      <td>S</td>\n",
       "      <td>746</td>\n",
       "      <td>LPNLNsQGVAW</td>\n",
       "      <td>LPNLNsQGGFS</td>\n",
       "      <td>First</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8</td>\n",
       "      <td>Q9UPQ0-10_S750;Q9UPQ0-6_S596;Q9UPQ0-1_S750</td>\n",
       "      <td>S</td>\n",
       "      <td>750</td>\n",
       "      <td>PSQVDsPSSEK</td>\n",
       "      <td>ILKVDsPSSEK</td>\n",
       "      <td>Second</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11</td>\n",
       "      <td>P62847-1_K129</td>\n",
       "      <td>K</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NVGAGkKSVSW</td>\n",
       "      <td>NVGAGkKAEGV</td>\n",
       "      <td>First</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Event ID                               Source of PTM Residue  \\\n",
       "0         3                 P01116-2_T148;P01116-1_T148       T   \n",
       "1         3                 P01116-1_K147;P01116-2_K147       K   \n",
       "0         8                               Q9UPQ0-1_S746       S   \n",
       "1         8  Q9UPQ0-10_S750;Q9UPQ0-6_S596;Q9UPQ0-1_S750       S   \n",
       "0        11                               P62847-1_K129       K   \n",
       "\n",
       "  PTM Position in Canonical Isoform Inclusion Sequence Exclusion Sequence  \\\n",
       "0                               148        ETSAKtRQESG        ETSAKtRQGC*   \n",
       "1                               147        IETSAkTRQES        IETSAkTRQGC   \n",
       "0                               746        LPNLNsQGVAW        LPNLNsQGGFS   \n",
       "1                               750        PSQVDsPSSEK        ILKVDsPSSEK   \n",
       "0                               NaN        NVGAGkKSVSW        NVGAGkKAEGV   \n",
       "\n",
       "   Region  Translation Success  Matched  \n",
       "0  Second                 True    False  \n",
       "1  Second                 True    False  \n",
       "0   First                 True    False  \n",
       "1  Second                 True    False  \n",
       "0   First                 True    False  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "flanks.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also do additional comparisons, such as comparing sequence identity and looking for matching elm motifs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "flanks = flanking_sequences.compare_flanking_sequences(flanks)\n",
    "flanks = flanking_sequences.compare_inclusion_motifs(flanks)\n",
    "flanks[['Source of PTM','Sequence Identity', 'Altered Positions','Residue Changes', 'Altered Flank Side', 'Motif only in Inclusion', 'Motif only in Exclusion']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Phase 4: Annotate PTMs with functional information\n",
    "\n",
    "Once we have PTMs impacted by splicing, we can also annotate them with additional information. This can be done using the annotate module of PTM-POSE, and can be used with outputs from either the project module (differentially included PTMs) or the flanking_sequence module (PTMs with altered flanking sequences).\n",
    "\n",
    "Currently, there are functions for appending information from:\n",
    "1. PhosphoSitePlus (function, biological process, disease association, interactions, and kinase-substrate), \n",
    "2. PTMsigDB (iKiP db, perturbations)\n",
    "3. RegPhos (kinase-substrate), \n",
    "4. PTMcode (inter and intraprotein interactions)\n",
    "5. PTMInt (interactions)\n",
    "6. DEPOD (Phosphatase-substrate)\n",
    "7. ELM (interactions, motifs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PhosphoSitePlus regulatory_site information added:\n",
      "\t ->6 PTMs in dataset found associated with a molecular function \n",
      "\t ->7 PTMs in dataset found associated with a biological process\n",
      "\t ->2 PTMs in dataset found associated with a protein interaction\n",
      "PhosphoSitePlus disease associations added: 1 PTM sites in dataset found associated with a disease in PhosphoSitePlus\n",
      "PhosphoSitePlus kinase-substrate interactions added: 6 phosphorylation sites in dataset found associated with a kinase in PhosphoSitePlus\n",
      "ELM interaction instances added: 1 PTMs in dataset found associated with at least one known ELM instance\n",
      "PTMInt data added: 2 PTMs in dataset found with PTMInt interaction information\n",
      "PTMcode interprotein interactions added: 27 PTMs in dataset found with PTMcode interprotein interaction information\n",
      "DEPOD Phosphatase substrates added: 0 PTMs in dataset found with Phosphatase substrate information\n",
      "RegPhos kinase-substrate data added: 3 PTMs in dataset found with kinase-substrate information\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Sam\\miniconda3\\envs\\testing_pose\\Lib\\site-packages\\ptm_pose\\annotate.py:558: DtypeWarning: Columns (4) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  regphos = pd.read_csv('http://140.138.144.141/~RegPhos/download/RegPhos_Phos_human.txt', sep = '\\t')\n"
     ]
    }
   ],
   "source": [
    "from ptm_pose import annotate\n",
    "\n",
    "#where to find PhosphoSitePlus data\n",
    "psp_regulatory_file = '/PhosphoSitePlus/Regulatory_sites.gz'\n",
    "psp_disease_file = '/PhosphoSitePlus/Disease-associated_sites.gz'\n",
    "psp_kinase_file = '/Database_Information/PhosphoSitePlus/Kinase_Substrate_Dataset.gz'\n",
    "\n",
    "#where to find ELM data\n",
    "\n",
    "#PhosphoSitePlus data (due to licencsing issues, must be downloaded manually from PhosphoSitePlus and the file path provided)\n",
    "spliced_ptms = annotate.add_PSP_regulatory_site_data(spliced_ptms, '/PhosphoSitePlus/Regulatory_sites.gz')\n",
    "spliced_ptms = annotate.add_PSP_disease_association(spliced_ptms, '/PhosphoSitePlus/Disease-associated_sites.gz')\n",
    "spliced_ptms = annotate.add_PSP_kinase_substrate_data(spliced_ptms, '/Database_Information/PhosphoSitePlus/Kinase_Substrate_Dataset.gz')\n",
    "\n",
    "#ELM interactions (will be faster if file is downloaded manually from ELM and the file path provided)\n",
    "spliced_ptms = annotate.add_ELM_interactions(spliced_ptms)\n",
    "\n",
    "#PTMint interactions\n",
    "spliced_ptms = annotate.add_PTMint_data(spliced_ptms)\n",
    "\n",
    "#PTMcode interactions (will be faster/more reliable if file is downloaded manually from PTMcode and the file path provided)\n",
    "ptm_code_interprotein = '/PTMcode2_associations_between_proteins.txt.gz'\n",
    "\n",
    "#DEPOD phosphatase data\n",
    "spliced_ptms = annotate.add_DEPOD_phosphatase_data(spliced_ptms)\n",
    "\n",
    "#RegPhos data\n",
    "spliced_ptms = annotate.add_RegPhos_data(spliced_ptms)\n",
    "\n",
    "#annotate ptms\n",
    "spliced_ptms = annotate.annotate_ptms(spliced_ptms)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Phase 5: Analyze Results\n",
    "\n",
    "Once we have all of this information, we can start to assess how PTMs are impacted by splicing. Let's first get an idea for how many PTMs have different annotations associated with them from the various sources"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from ptm_pose import analyze\n",
    "\n",
    "analyze.show_available_annotations(spliced_ptms, figsize = (5,5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several ptms that have previously been annotated with specific functions, let's take a look at those:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>Gene</th>\n",
       "      <th>UniProtKB Accession</th>\n",
       "      <th>Residue</th>\n",
       "      <th>PTM Position in Canonical Isoform</th>\n",
       "      <th>Modification Class</th>\n",
       "      <th>PSP:ON_PROCESS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>CEACAM1</td>\n",
       "      <td>P13688</td>\n",
       "      <td>S</td>\n",
       "      <td>461.0</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>apoptosis, altered</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184</th>\n",
       "      <td>YAP1</td>\n",
       "      <td>P46937</td>\n",
       "      <td>K</td>\n",
       "      <td>342.0</td>\n",
       "      <td>Ubiquitination</td>\n",
       "      <td>carcinogenesis, altered</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>217</th>\n",
       "      <td>TSC2</td>\n",
       "      <td>P49815</td>\n",
       "      <td>S</td>\n",
       "      <td>981.0</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>carcinogenesis, inhibited; cell growth, inhibi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395</th>\n",
       "      <td>SPHK2</td>\n",
       "      <td>Q9NRA0</td>\n",
       "      <td>S</td>\n",
       "      <td>387.0</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>cell motility, altered</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>407</th>\n",
       "      <td>SPHK2</td>\n",
       "      <td>Q9NRA0</td>\n",
       "      <td>T</td>\n",
       "      <td>614.0</td>\n",
       "      <td>Phosphorylation</td>\n",
       "      <td>cell motility, altered</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Gene UniProtKB Accession Residue  PTM Position in Canonical Isoform  \\\n",
       "145  CEACAM1              P13688       S                              461.0   \n",
       "184     YAP1              P46937       K                              342.0   \n",
       "217     TSC2              P49815       S                              981.0   \n",
       "395    SPHK2              Q9NRA0       S                              387.0   \n",
       "407    SPHK2              Q9NRA0       T                              614.0   \n",
       "\n",
       "    Modification Class                                     PSP:ON_PROCESS  \n",
       "145    Phosphorylation                                 apoptosis, altered  \n",
       "184     Ubiquitination                            carcinogenesis, altered  \n",
       "217    Phosphorylation  carcinogenesis, inhibited; cell growth, inhibi...  \n",
       "395    Phosphorylation                             cell motility, altered  \n",
       "407    Phosphorylation                             cell motility, altered  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "annotations, annotation_counts = analyze.get_ptm_annotations(spliced_ptms, annotation_type = 'Process', database = 'PhosphoSitePlus')\n",
    "annotations.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>PSP:ON_PROCESS</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cell motility, altered</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>cell growth, induced</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>apoptosis, altered</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>carcinogenesis, altered</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>carcinogenesis, inhibited</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>cell growth, inhibited</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>autophagy, inhibited</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>signaling pathway regulation</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>cytoskeletal reorganization</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>cell adhesion, inhibited</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 PSP:ON_PROCESS  count\n",
       "0        cell motility, altered      3\n",
       "1          cell growth, induced      2\n",
       "2            apoptosis, altered      1\n",
       "3       carcinogenesis, altered      1\n",
       "4     carcinogenesis, inhibited      1\n",
       "5        cell growth, inhibited      1\n",
       "6          autophagy, inhibited      1\n",
       "7  signaling pathway regulation      1\n",
       "8   cytoskeletal reorganization      1\n",
       "9      cell adhesion, inhibited      1"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "annotation_counts"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pose",
   "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": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}