Structure analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "from scipy import stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def accessmap(row):\n",
    "    if ((float(row['ASA']) ==0) & (float(row['BSA'])==0)):\n",
    "        val = 'B'\n",
    "    elif ((float(row['ASA']) >0) & (float(row['BSA'])==0)):\n",
    "        val = 'A'\n",
    "    elif ((float(row['ASA']) >0) & (float(row['BSA'])>0)):\n",
    "        val = 'I'\n",
    "    return val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8482, 26)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#################################################################################### Accessibility statistics\n",
    "Scop3P_structure=pd.read_csv('Structure_analysis_PISA.txt',sep='\\t')\n",
    "Scop3P_structure.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "accdataSTY=pd.read_csv('PISA_ASA_allSTY.txt',sep='\\t')\n",
    "accdataSTY['PDBchain']=accdataSTY['PDBID']+accdataSTY['CHAINID']\n",
    "accdataSTY=accdataSTY[accdataSTY['RESPOS'].isin(['SER','THR','TYR','SEP','TPO','PTR'])]\n",
    "accdataSTY.PDBPOS=accdataSTY.PDBPOS.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "after fragment size site filtering and amino acid filtering 8432\n",
      "after fragment and length PDB IDs 3517\n",
      "after fragment and length UP IDs 3059\n",
      "\n",
      " RESIDUES and Ssec\n",
      "SER    4879\n",
      "THR    2035\n",
      "TYR    1442\n",
      "TPO      28\n",
      "SEP      25\n",
      "PTR      23\n",
      "Name: RESIDUE, dtype: int64\n",
      "C    4622\n",
      "H    2218\n",
      "E    1090\n",
      "-     502\n",
      "Name: Ssec, dtype: int64\n",
      "\n",
      "\n",
      "SER    0.578629\n",
      "THR    0.241343\n",
      "TYR    0.171015\n",
      "TPO    0.003321\n",
      "SEP    0.002965\n",
      "PTR    0.002728\n",
      "Name: RESIDUE, dtype: float64\n",
      "C    0.548150\n",
      "H    0.263046\n",
      "E    0.129269\n",
      "-    0.059535\n",
      "Name: Ssec, dtype: float64\n",
      "\n",
      "SER/SEP in Ssec C    0.607463\n",
      "H    0.244086\n",
      "E    0.093801\n",
      "-    0.054649\n",
      "Name: Ssec, dtype: float64\n",
      "\n",
      "THR/TPO in Ssec C    0.563742\n",
      "H    0.235095\n",
      "E    0.147358\n",
      "-    0.053805\n",
      "Name: Ssec, dtype: float64\n",
      "\n",
      "TYR/PTR in Ssec H    0.365870\n",
      "C    0.327645\n",
      "E    0.222526\n",
      "-    0.083959\n",
      "Name: Ssec, dtype: float64\n",
      "\n",
      "After ASA == - filter (6893, 27)\n",
      "Total number of sites in accessible region 5701\n",
      "Total number of sites in interface region 1076\n",
      "Total number of sites in buried region 116\n",
      "%%%%%%%%%%%%%%%%%%%%%%%% Accessible %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
      "SER    3301\n",
      "THR    1416\n",
      "TYR     928\n",
      "TPO      26\n",
      "PTR      17\n",
      "SEP      13\n",
      "Name: RESIDUE, dtype: int64\n",
      "SER    0.579021\n",
      "THR    0.248377\n",
      "TYR    0.162778\n",
      "TPO    0.004561\n",
      "PTR    0.002982\n",
      "SEP    0.002280\n",
      "Name: RESIDUE, dtype: float64\n",
      "%%%%%%%%%%%%%%%%%%%%%%%  Buried  %%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
      "SER    74\n",
      "THR    31\n",
      "TYR    11\n",
      "Name: RESIDUE, dtype: int64\n",
      "SER    0.637931\n",
      "THR    0.267241\n",
      "TYR    0.094828\n",
      "Name: RESIDUE, dtype: float64\n",
      "%%%%%%%%%%%%%%%%%%%%%%%%  Interface  %%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
      "SER    581\n",
      "TYR    245\n",
      "THR    230\n",
      "SEP     12\n",
      "PTR      6\n",
      "TPO      2\n",
      "Name: RESIDUE, dtype: int64\n",
      "SER    0.539963\n",
      "TYR    0.227695\n",
      "THR    0.213755\n",
      "SEP    0.011152\n",
      "PTR    0.005576\n",
      "TPO    0.001859\n",
      "Name: RESIDUE, dtype: float64\n",
      "%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/pathmanaban/.local/lib/python3.8/site-packages/pandas/core/generic.py:5159: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self[name] = value\n",
      "/home/pathmanaban/.local/lib/python3.8/site-packages/pandas/core/frame.py:3065: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self[k1] = value[k2]\n",
      "/home/pathmanaban/.local/lib/python3.8/site-packages/pandas/core/frame.py:3065: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self[k1] = value[k2]\n",
      "/home/pathmanaban/.local/lib/python3.8/site-packages/pandas/core/frame.py:3065: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self[k1] = value[k2]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Access, interface and buried 11 data points 4873 886 116\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAboAAAFECAYAAABRb/vsAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAAml0lEQVR4nO3deXTU5b3H8c/MhAHCFpITcFhks2AqZY0sQkAiyi5XXEBkkeUiClgFEZRCCoo2kKIH5Zha1Hs8F+FSlVB2sAq4VCtWWUTLUghChiQkrAlZyDz3D8ocMCEJycAkj+/XOZzDzPNbvr/vTH6f/J5Z4jDGGAEAYClnsAsAAOB6IugAAFYj6AAAViPoAABWI+gAAFYj6AAAViPoUKmNHz9eq1atkiR9+OGHevjhh6+67MiRI/WXv/zlRpVWpDlz5mjJkiVXHW/VqpWSk5NvYEXFi42N1RdffBHQbc6cOVOvvPJKQLcJFCck2AWg4hs5cqR+/PFHff7553K73cEu5wpLly4NdgnXZN68ecEuoVKZOXOm6tevr6effjrYpZRaZazZdlzRoVhHjx7Vjh075HA49Le//S3Y5VQIFy5cCHYJAK4BQYdiJSUlqW3btrrvvvuUlJR0xZjX69XkyZPVpUsXde7c+YqrlZUrV6pfv35q3769+vfvr++//16SlJqaqilTpqhLly6KjY3Vu+++619n165dGjJkiDp06KA77rhDL7/8siQpNzdXzzzzjDp37qzo6Gjdf//9OnHihKTC05HGGM2bN08dO3ZU37599fe///2qx/b++++rX79+uv322zVu3DgdO3asyOWOHj2qVq1a6S9/+YvuvPNOjR49utj1jTF66aWX1LVrV3Xo0EGDBg3Svn37JBWetlu6dKm6d++u7t276/33379ivz8/tp9PzR48eFBjxoxRp06d1KdPH61fv94/tm3bNvXv31/t27dXTEyM3nrrrSKP7ciRIxo1apQ6d+6szp07a9q0aTpz5swVy+zevVv9+/fX7bffrueee065ubmSpMzMTD322GOKjo5Wp06dNHz4cPl8Pn9tI0eOVHR0tAYMGHDVX5KKmm6+NH37f//3f1qzZo3eeusttW/fXhMnTpRU/HPo53JycvSHP/xBvXr1UseOHfXwww8rJydHkvS3v/1NAwYMUHR0tEaOHKmDBw8WquGSyx+3r776Sj169NDbb7+trl27qnv37vrggw8k6ao1I8gMUIzevXub//3f/zW7d+82v/71r016eroxxpgLFy6YQYMGmfnz55usrCyTk5Njvv76a2OMMevXrzfdu3c3O3fuND6fzxw+fNgcPXrUFBQUmPvuu8+89tprJjc31xw5csTExsaa7du3G2OMeeihh8yqVauMMcacO3fOfPvtt8YYY5YvX24ee+wxk52dbS5cuGB2795tzp49a4wxZsSIEWblypXGGGM++OADExUVZd555x2Tl5dn1q1bZzp06GBOnjxZaNktW7aY3r17mwMHDpj8/HyzZMkSM3To0CJ78NNPP5mWLVua6dOnm6ysLHP+/Pli19++fbu57777zOnTp43P5zMHDhwwqampxhhjZsyYYRYtWmSMMWbbtm2ma9eu5l//+pfJysoyU6dONS1btjSHDx8uVO+l4xs2bJgxxpisrCzTo0cP8/7775v8/Hzz/fffm06dOpn9+/cbY4zp1q2b//E4deqU2bNnT5HHdvjwYfPZZ5+Z3Nxck5GRYYYPH25efPFF/3ivXr3MgAEDTEpKijl58qQZOnSov/6EhAQze/Zsk5eXZ/Ly8szXX39tfD6fycvLM7179zZvvPGGyc3NNV988YVp166dOXjwYKEeXH5Ml1zeg8uXNcaU+Bz6ud///vdmxIgR5vjx4+bChQvmm2++Mbm5uebf//63adu2rfnss89MXl6eefPNN03v3r1Nbm5uoRp+XseXX35poqKizKuvvmry8vLM1q1bTZs2bcypU6eKrBnBxxUdrmrHjh1KSUlRv3791Lp1azVu3Fhr166VdPHqKy0tTc8++6xCQ0NVtWpVRUdHS7p4pTN+/Hi1adNGDodDTZo0UcOGDbV7925lZmZq8uTJcrvdaty4sR566CH/lUhISIiOHDmizMxM1ahRQ+3atfPff+rUKSUnJ8vlcql169aqWbNmkTWHh4dr9OjRqlKlivr3769mzZpp69athZZbsWKFJkyYoBYtWigkJEQTJ07UDz/8cNWrOkmaMmWKQkNDVa1atWLXDwkJUVZWlv7973/LGKMWLVqoXr16hba3YcMGDRkyRC1btlRoaKgmT55c6sdm69atatiwoe6//36FhITo17/+tfr06aONGzf6e3bgwAGdO3dOderU0W233Vbkdpo0aaJu3brJ7XYrPDxcY8aM0ddff33FMo888og8Ho/CwsL0+OOPa926df59pKenKyUlRVWqVFF0dLQcDod27typ7OxsTZgwQW63W127dlWvXr3865VHSc+hy/l8Pn3wwQeaNWuW6tevL5fLpQ4dOsjtdmv9+vXq2bOnunXrpipVqmjcuHHKycnRt99+W6o6QkJCNGnSJFWpUkU9e/ZUaGioDh06VO7jw/XBm1FwVUlJSerWrZvCw8MlSQMHDtSqVav06KOPyuv1qkGDBgoJKfwU8nq9uvnmmwvdf+zYMaWlpfkDUZIKCgr8t+fPn6/FixerX79+atSokSZPnqxevXpp8ODBOn78uKZOnaozZ87o3nvv1dNPP60qVaoU2kf9+vXlcDj8txs0aKC0tLRCy6WkpOill15SfHy8/z5jjFJTU9WwYcMi+3HTTTeVav2uXbvqkUce0bx583Ts2DHdc889mjFjRqFwTktLU+vWrf23r7bfohw7dky7du0q1Mt7771XkrR48WK98cYb+uMf/6hWrVpp2rRpat++faHtnDhxQvPnz9eOHTuUlZUlY4xq1659xTIej8f//8v7OW7cOL3++usaO3asJGno0KGaMGGC0tLSdNNNN8npdF6xXmpqaqmPr7jjLu45dLmTJ08qNzdXjRs3LjSWlpamBg0a+G87nU55PJ5S1xgWFnbFc7969erKzs6+lkPBDUTQoUg5OTnasGGDfD6funXrJknKy8vTmTNn9OOPP8rj8cjr9erChQuFws7j8ejIkSOFtunxeNSoUSNt3ry5yH02bdpUixYtks/n0+bNm/Xkk0/qq6++8l/tTJ48WUePHtWECRPUrFkzPfjgg4W2kZqaKmOMP+y8Xq9iY2OLrGXixIn+YCiNywO0pPVHjRqlUaNGKSMjQ0899ZSWLl2qp5566opl6tWrJ6/X67+dkpJyxXj16tV1/vx5/+1Lr0te2v/tt9+ud955p8j9t2nTRm+88Yby8/O1bNkyPfXUU9q2bVuh5RYtWiSHw6E1a9YoLCxMH330UaF3hv68xktXpzVr1tTMmTM1c+ZM7du3T6NHj9ZvfvMb1atXT8ePH5fP5/OHndfrVdOmTQvtv3r16v7XzCQpPT39ivHLe37puIt7Dl2ubt26qlq1qn766SfdeuutV4zVq1fP/7qpdPGXFK/Xq/r16/vrurz36enp/rGS/LxmBB9TlyjSRx99JJfLpXXr1ikpKUlJSUlav369oqOjlZSUpDZt2igyMlJ//OMflZ2drdzcXH3zzTeSpAceeEBvv/229uzZI2OMkpOTdezYMbVp00Y1atTQm2++qZycHBUUFGjfvn3atWuXJGn16tXKzMyU0+n0X1U4nU59+eWX+te//qWCggLVrFlTISEhV1wtXC4zM1Pvvvuu8vPztWHDBh08eFA9e/YstNywYcP05ptvav/+/ZKks2fPasOGDaXuT3Hr79q1Szt37lR+fr6qV68ut9tdZL19+/bVqlWrdODAAZ0/f16vv/76FeNRUVHasmWLzp8/r+Tk5CverHLnnXfq8OHDSkpKUn5+vvLz87Vr1y4dPHhQeXl5+utf/6qzZ8+qSpUqqlGjxlX7lZWVpdDQUNWqVUupqalFflzjvffe0/Hjx3Xq1CklJiaqf//+kqRPPvlEycnJMsaoVq1acrlccjgcatOmjapVq6alS5cqPz9fX331lT7++GP/epe79dZbtX//fv3www/Kzc3Va6+9dsV4RESEjh496r9d0nPock6nU/fff79efvllpaamqqCgQN9++63y8vLUr18/bdu2TX//+9+Vn5+vt99+W26323/Ve+utt2rt2rUqKCjQ9u3bC03nFufnNSP4CDoUadWqVRoyZIgaNGigyMhI/79HHnlEa9askTFGiYmJSk5OVq9evdSjRw//ib5fv36aOHGipk2bpg4dOmjSpEk6ffq0XC6XEhMT9eOPP+quu+5Sly5d9Lvf/U7nzp2TJH366acaMGCA2rdvr/nz5+uVV15RtWrVdOLECT355JPq2LGj+vfvr06dOmnw4MFF1t2mTRslJyerS5cuevXVV7V48WLVrVu30HJ33323xo8fr6lTp6pDhw4aOHCgtm/fXur+FLd+VlaWfve736lTp07q1auXwsLCNG7cuELb6Nmzp0aPHq3Ro0fr7rvvVpcuXa4Yv/Ra4x133KEZM2Zo0KBB/rGaNWvqrbfe0vr16xUTE6Pu3bsrISFBeXl5ki7+0hAbG6sOHTpoxYoVWrhwYZHHMXnyZO3du1fR0dGaMGGC7rnnnkLLDBw4UGPHjlXv3r1188036/HHH5ckJScna8yYMWrfvr2GDh2qhx9+WF26dJHb7VZiYqK2b9+uLl26aO7cuVqwYIFatGhRaNvNmjXTpEmT9Oijj+qee+5Rx44drxh/4IEHdODAAUVHR+uJJ54o8Tn0czNmzFDLli31wAMPqFOnTkpISJDP51Pz5s21cOFCvfDCC+rSpYs++eQTJSYm+j8nOmvWLH3yySeKjo7WmjVr1Lt37yK3X5Sf14zgcxjDH14FANiLKzoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVKu03o5w8mSWfLzifjIiIqKmMjKI/t/NLQQ/owSX0gR5Iwe2B0+lQ3bo1rjpeaYPO5zNBC7pL+/+lowf04BL6QA+kitsDpi4BAFYj6AAAViPoAABWI+gAAFYj6AAAViPoAABWI+gAAFYj6AAAViPoAABWq7TfjFIeNeq4FOoOLdc2IiNrlWv97LxsZZ0uKNc2AAAl+0UGXag7VI65jqDWYOKMsnQ2qDUAwC8BU5cAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKuFlGahJ554QkePHpXT6VRoaKhmz56tqKgoHTp0SDNnztSpU6cUFham+Ph4NW3aVJLKPAYAQCCV6oouPj5ef/3rX5WUlKSxY8fq+eeflyTFxcVp+PDh2rRpk4YPH645c+b41ynrGAAAgVSqoKtVq5b//+fOnZPD4VBGRob27t2rgQMHSpIGDhyovXv3KjMzs8xjAAAEWqmmLiVp1qxZ+vzzz2WM0dKlS+X1elW/fn25XC5JksvlUr169eT1emWMKdNYeHh4qQuPiKh5LcdZIUVG1ip5oQqsstcfCPTgIvpAD6SK24NSB938+fMlSUlJSVqwYIF++9vfXreiSiMj45x8PlOmdSvKg5GefjbYJZRZZGStSl1/INCDi+gDPZCC2wOn01Hsxc81v+vyv/7rv/TVV1/ppptuUmpqqgoKCiRJBQUFSktLk8fjkcfjKdMYAACBVmLQZWVlyev1+m9//PHHqlOnjiIiIhQVFaW1a9dKktauXauoqCiFh4eXeQwAgEBzGGOKnf87ceKEnnjiCZ0/f15Op1N16tTRjBkzdNttt+ngwYOaOXOmzpw5o9q1ays+Pl7NmzeXpDKPlVZ5py4dcx1lWjdQTJyp1FMdTNXQg0voAz2QKvbUZYlBV1ERdMHFDzY9uIQ+0AOpYgcd34wCALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwGkEHALAaQQcAsBpBBwCwWkhJC5w8eVLPPvusjhw5IrfbrSZNmmjevHkKDw9Xq1at1LJlSzmdF/NywYIFatWqlSTp448/1oIFC1RQUKDbbrtNL7/8sqpXr17iGAAAgVTiFZ3D4dD48eO1adMmrVmzRo0bN1ZCQoJ/fMWKFVq9erVWr17tD7msrCzNnj1biYmJ2rJli2rUqKG33nqrxDEAAAKtxKALCwtT586d/bfbtWunlJSUYtfZvn27WrduraZNm0qShg0bpg0bNpQ4BgBAoJU4dXk5n8+n5cuXKzY21n/fyJEjVVBQoB49emjKlClyu93yer1q0KCBf5kGDRrI6/VKUrFj1yIiouY1r1PRREbWCnYJ5VLZ6w8EenARfaAHUsXtwTUF3QsvvKDQ0FCNGDFCkrR161Z5PB6dO3dO06dP15IlS/T0009fl0J/LiPjnHw+U6Z1K8qDkZ5+NtgllFlkZK1KXX8g0IOL6AM9kILbA6fTUezFT6nfdRkfH6/k5GS9+uqr/jefeDweSVLNmjX14IMP6p///Kf//sunN1NSUvzLFjcGAECglSroFi1apD179mjJkiVyu92SpNOnTysnJ0eSdOHCBW3atElRUVGSpJiYGO3evVuHDx+WdPENK/369StxDACAQCtx6nL//v3605/+pKZNm2rYsGGSpEaNGmn8+PGaM2eOHA6HLly4oPbt2+u3v/2tpItXePPmzdNjjz0mn8+nqKgozZo1q8QxAAACzWGMKdsLXUFW3tfoHHMdAa7o2pg4U6nn9HlNgh5cQh/ogWTJa3QAAFRGBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAaiUG3cmTJ/Xf//3f6tOnjwYNGqTJkycrMzNTkvTdd9/p3nvvVZ8+fTR27FhlZGT41yvrGAAAgVRi0DkcDo0fP16bNm3SmjVr1LhxYyUkJMjn82n69OmaM2eONm3apOjoaCUkJEhSmccAAAi0EoMuLCxMnTt39t9u166dUlJStGfPHlWtWlXR0dGSpGHDhmnjxo2SVOYxAAACLeRaFvb5fFq+fLliY2Pl9XrVoEED/1h4eLh8Pp9OnTpV5rGwsLBS1xIRUfNaSq+QIiNrBbuEcqns9QcCPbiIPtADqeL24JqC7oUXXlBoaKhGjBihLVu2XK+aSiUj45x8PlOmdSvKg5GefjbYJZRZZGStSl1/INCDi+gDPZCC2wOn01HsxU+pgy4+Pl7JyclKTEyU0+mUx+NRSkqKfzwzM1NOp1NhYWFlHgMAINBK9fGCRYsWac+ePVqyZIncbrckqXXr1srJydGOHTskSStWrFDfvn3LNQYAQKCVeEW3f/9+/elPf1LTpk01bNgwSVKjRo20ZMkSLViwQHFxccrNzVXDhg21cOFCSZLT6SzTGAAAgeYwxpTtha4gK+9rdI65jgBXdG1MnKnUc/q8JkEPLqEP9ECq2K/R8c0oAACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAqxF0AACrEXQAAKsRdAAAq5Uq6OLj4xUbG6tWrVpp3759/vtjY2PVt29fDR48WIMHD9ann37qH/vuu+907733qk+fPho7dqwyMjJKNQYAQCCVKujuuusuLVu2TA0bNiw0tnjxYq1evVqrV69WTEyMJMnn82n69OmaM2eONm3apOjoaCUkJJQ4BgBAoJUq6KKjo+XxeEq90T179qhq1aqKjo6WJA0bNkwbN24scQwAgEALKe8GnnnmGRlj1LFjR02dOlW1a9eW1+tVgwYN/MuEh4fL5/Pp1KlTxY6FhYWVtxwAAK5QrqBbtmyZPB6P8vLyNH/+fM2bN++GTUNGRNS8Ifu5niIjawW7hHKp7PUHAj24iD7QA6ni9qBcQXdpOtPtdmv48OF6/PHH/fenpKT4l8vMzJTT6VRYWFixY9ciI+OcfD5TproryoORnn422CWUWWRkrUpdfyDQg4voAz2QgtsDp9NR7MVPmT9ekJ2drbNnLx6UMUbr169XVFSUJKl169bKycnRjh07JEkrVqxQ3759SxwDACDQSnVF9+KLL2rz5s06ceKExowZo7CwMCUmJmrKlCkqKCiQz+dTixYtFBcXJ0lyOp1asGCB4uLilJubq4YNG2rhwoUljgEAEGgOY0zZ5v+CrLxTl465jgBXdG1MnKnUUx1M1dCDS+gDPZAsnboEAKAyKPfHC1A51ajjUqg7tFzbKO+berLzspV1uqBc2wCAkhB0v1Ch7tAKMX2bpV/2dA+A64+pSwCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1Qg6AIDVCDoAgNUIOgCA1UoMuvj4eMXGxqpVq1bat2+f//5Dhw5p6NCh6tOnj4YOHarDhw+XewwAgEArMejuuusuLVu2TA0bNrzi/ri4OA0fPlybNm3S8OHDNWfOnHKPAQAQaCUGXXR0tDwezxX3ZWRkaO/evRo4cKAkaeDAgdq7d68yMzPLPAYAwPUQUpaVvF6v6tevL5fLJUlyuVyqV6+evF6vjDFlGgsPD7+mGiIiapal9AolMrJWsEsIusreg8pef6DQB3ogVdwelCnoKoKMjHPy+UyZ1q0oD0Z6+tmg7ZselF9kZK1KXX+g0Ad6IAW3B06no9iLnzIFncfjUWpqqgoKCuRyuVRQUKC0tDR5PB4ZY8o0BgDA9VCmjxdEREQoKipKa9eulSStXbtWUVFRCg8PL/MYAADXg8MYU+z834svvqjNmzfrxIkTqlu3rsLCwrRu3TodPHhQM2fO1JkzZ1S7dm3Fx8erefPmklTmsWtR3qlLx1xHmdYNFBNngj51+UvvQXkxXXURfaAHUsWeuiwx6Coqgq586EH5cXK7iD7QA6liB12lfTMKUF416rgU6g4t1zbK+6ae7LxsZZ0uKNc2ABSPoMMvVqg7tEJc1Wbpl30lAFxvfNclAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqBB0AwGoEHQDAagQdAMBqIeXdQGxsrNxut6pWrSpJeuaZZxQTE6PvvvtOc+bMUW5urho2bKiFCxcqIiJCkoodAwAgkAJyRbd48WKtXr1aq1evVkxMjHw+n6ZPn645c+Zo06ZNio6OVkJCgiQVOwYAQKBdl6nLPXv2qGrVqoqOjpYkDRs2TBs3bixxDACAQCv31KV0cbrSGKOOHTtq6tSp8nq9atCggX88PDxcPp9Pp06dKnYsLCys1PuMiKgZiNKDKjKyVrBLCDp6YEcPbDiG8qIHFbcH5Q66ZcuWyePxKC8vT/Pnz9e8efN09913B6K2YmVknJPPZ8q0bkV5MNLTzwZt3/SAHgRKZGStSn8M5UUPgtsDp9NR7MVPuacuPR6PJMntdmv48OH65z//KY/Ho5SUFP8ymZmZcjqdCgsLK3YMAIBAK1fQZWdn6+zZiwlujNH69esVFRWl1q1bKycnRzt27JAkrVixQn379pWkYscAAAi0ck1dZmRkaMqUKSooKJDP51OLFi0UFxcnp9OpBQsWKC4u7oqPEEgqdgwAgEArV9A1btxYSUlJRY516NBBa9asueYxAAACiW9GAQBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWC1oQXfo0CENHTpUffr00dChQ3X48OFglQIAsFhIsHYcFxen4cOHa/DgwVq9erXmzJmjd999N1jlAL9INeq4FOoOLfd2IiNrlXnd7LxsZZ0uKHcNwNUEJegyMjK0d+9evfPOO5KkgQMH6oUXXlBmZqbCw8NLtQ2n01GuGprUaVKu9QOhvMdQXvSAHoS6Q9X01aZB278kHX7qsM47zwVt/6G1XKrurl7u7ZQn7M/nnVf22eCFfWXvQUk/Q0EJOq/Xq/r168vlckmSXC6X6tWrJ6/XW+qgq1u3RrlqOPzU4XKtHwgRETWDun96QA8kelARVHdXV/WIYFcRXNezB7wZBQBgtaAEncfjUWpqqgoKLl6mFhQUKC0tTR6PJxjlAAAsFpSgi4iIUFRUlNauXStJWrt2raKioko9bQkAQGk5jDEmGDs+ePCgZs6cqTNnzqh27dqKj49X8+bNg1EKAMBiQQs6AABuBN6MAgCwGkEHALAaQQcAsBpBBwCwGkEHALBa0L7UuSLasWOHEhIStH//frlcLjVv3lzPP/+8Dhw4oFmzZqlatWpXLL9x40bVr19fsbGxOnHihFwul0JDQxUTE6PZs2erRo3yfU3ZjdS+fXv//8+fPy+32+3/ira5c+cqOTlZycnJSkhIuGK9Vq1aafPmzWrSpIlee+01JSYm+te95ZZbNGPGjCu2XdEV14c5c+ZowYIFWrdunf8zn3l5eRo8eLBGjRqlmJgY3XXXXQoNvfglyXXr1tWwYcM0YcKEG38g5VCa50Jxj/Px48e1cOFCffbZZ8rNzdUtt9yiSZMmqVevXkE5nrK6/Oe6evXq6tGjh3bs2KGMjAxJUk5OjkJCQhQScvE0+thjj6levXr+c4XD4VCjRo309NNPV7pjv1wgzouX+he086KBMcaYs2fPmo4dO5o1a9aYCxcumPPnz5tPP/3U/PDDD+aDDz4ww4YNu+q6vXr1Mp9//rkxxpi0tDQzaNAgs2jRohtVesBdfjyXLF682EybNq3Qsi1btjSHDx8utEx+fr5ZtGiRiYmJuf4FXydF9eG5554zU6dO9d9+5ZVXzMiRI43P5zM//fSTadmypcnPzzfGGLNr1y7Ttm1b89lnn93QugOppOdCXl6eiY+PN926dTM+n8+cPHnS9OrVy8ycOdOkpaWZ8+fPmzVr1pj27dubDRs2BOMQyuzyYz9+/LgZMGCAWbhwoX98xIgRZuXKlVesc/m5oqCgwCxfvty0a9fOnD59+sYVHkCBOi8W1b8bianL/zh06JCki39JweVyqVq1aurevbtuvfXWa9pOZGSkunfvrh9++OF6lFlphISEaNCgQUpNTVVmZmawywmY5557Tv/4xz+0detW7du3T8uWLdP8+fPlcBT+9vTf/OY3uuWWW6x+LlSpUkX33Xef0tPTdfLkSf3P//yPQkNDNX/+fEVGRqpatWoaOHCgJk6cqPj4eJlK+rHd+vXrKyYmRvv37y/1Ok6nU4MHD1Z2dnal/XubgTovlqV/gUTQ/UezZs3kcrk0Y8YMbdu2TadPny7Tdo4fP65PP/1UN998c4ArrFzy8vKUlJSksLAw1a5dO9jlBEytWrU0d+5cxcXF6fnnn9ekSZPUuHHjIpf97rvvtH//fjVpEvw/BXS95OXl6cMPP5TH41F4eLi++OIL3XPPPXI6rzy19OvXTykpKf4TZ2Xj9Xq1fft2RUVFlXqdgoICffjhh6pSpYoaNmx4Hau7fgJ1XixL/wKJ1+j+o2bNmnrvvff05z//WbNnz9aJEyfUo0cPvfjii5KknTt3Kjo62r98WFiYPvroI//tSZMmSZKys7PVpUsXPfnkkzf2AG6AjRs3auvWraVaJisrS7Vr19bixYv9r2HYIjY2VklJSTp69KhGjRpVaLxLly7Ky8tTbm6uxo4dq969ewehyuvr0uNcpUoV/epXv9Lrr78uSTp58qQiIyMLLV+vXj3/eGUyadIkuVwu1apVSz179tTEiRNLXOfSueL8+fNyuVxasGCBIiIq59/gCcR58Vr7dz3YdQYqpxYtWugPf/iDpIvfxTl9+nS99NJL6t69u9q2bavly5dfdd0lS5bojjvu0D/+8Q9NmzZNJ0+etOpKRpL69u1b5JtRilomMzNTTz75pL7//nt17tz5RpZ5Q/zqV7+S2+0udOUiSV9++aUcDofeffddrVmzRvn5+XK73UGo8vop6rkgXXwDTnp6eqH709LS/OOVyaWf62tx6VyRlZWlWbNm6ZtvvlH//v2vU4XXXyDOi8HG1OVVtGjRQkOGDLnmOeVOnTppyJAhio+Pv06VVQ7h4eGaN2+eXnvtNf9J7pfE5XJpzJgxqlq1qt57771gl3PDdO3aVVu2bJHP57vi/g0bNsjj8ahZs2ZBquzGq1Gjhn7/+99r9erV2rt3b7DLCYiynheDjaD7j4MHD+rtt9/W8ePHJV2cU167dq3atm17zdsaPXq0vvjiC/3444+BLrNSad68uWJiYrR06dJglxI0EyZM0NKlS5WbmxvsUm6IRx99VGfPntWsWbOUnp6u3NxcrV27VomJiXr22WeLfNOOzcLCwvTggw9qyZIlwS6lTAJ5Xgwmgu4/atasqZ07d+rBBx9Uu3bt9NBDD6lly5aaOXOmpItvLGjfvv0V/3bt2lXktsLDwzV48OBK++QOpHHjxmnlypX+zx790tx5552qU6eOVq5cGexSboi6devqvffeU25urgYMGKDOnTvrnXfe0YIFCyr19F15jB49Wtu2bauUv/gG8rwYTPyZHgCA1biiAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWI2gAwBYjaADAFiNoAMAWO3/ASV3htBir+daAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#######################################################################   Accessibility statistics overall\n",
    "STRtemp=Scop3P_structure\n",
    "STRtemp=STRtemp[STRtemp['length']>=50]\n",
    "STRtemp=STRtemp[STRtemp['fragment']>=50]\n",
    "sns.set(style='darkgrid')\n",
    "\n",
    "Scop3P_allSTR_filt=STRtemp\n",
    "Scop3P_allSTR_filt['RESIDUE']= np.where(Scop3P_allSTR_filt['RESIDUE'] == '-', Scop3P_allSTR_filt['UP_RES'],\\\n",
    "                                        Scop3P_allSTR_filt['RESIDUE'])\n",
    "\n",
    "\n",
    "Scop3P_allSTR_filt['RESIDUE']= np.where(Scop3P_allSTR_filt['RESIDUE'] == 'X', Scop3P_allSTR_filt['PDB_RES'],\\\n",
    "                                        Scop3P_allSTR_filt['RESIDUE'])\n",
    "Scop3P_allSTR_filt.loc[(Scop3P_allSTR_filt.RESIDUE == 'S'),'RESIDUE']='SER'\n",
    "Scop3P_allSTR_filt.loc[(Scop3P_allSTR_filt.RESIDUE == 'T'),'RESIDUE']='THR'\n",
    "Scop3P_allSTR_filt.loc[(Scop3P_allSTR_filt.RESIDUE == 'Y'),'RESIDUE']='TYR'\n",
    "\n",
    "Scop3P_allSTR_filt=Scop3P_allSTR_filt[Scop3P_allSTR_filt['RESIDUE'].isin(['SER','SEP','THR',\\\n",
    "                                                                                        'TPO','TYR','PTR'])]\n",
    "\n",
    "print(\"\\nafter fragment size site filtering and amino acid filtering\",Scop3P_allSTR_filt.shape[0])\n",
    "print(\"after fragment and length PDB IDs\",Scop3P_allSTR_filt['PDBID'].nunique())\n",
    "print(\"after fragment and length UP IDs\",Scop3P_allSTR_filt['ACC_ID'].nunique())\n",
    "\n",
    "print (\"\\n RESIDUES and Ssec\")\n",
    "print (Scop3P_allSTR_filt.RESIDUE.value_counts())\n",
    "print (Scop3P_allSTR_filt.Ssec.value_counts())\n",
    "print ('\\n')\n",
    "print (Scop3P_allSTR_filt.RESIDUE.value_counts(normalize=True))\n",
    "print (Scop3P_allSTR_filt.Ssec.value_counts(normalize=True))\n",
    "print (\"\\nSER/SEP in Ssec\",Scop3P_allSTR_filt[Scop3P_allSTR_filt['RESIDUE'].isin(['SER','SEP'])].Ssec.value_counts(normalize=True))\n",
    "print (\"\\nTHR/TPO in Ssec\",Scop3P_allSTR_filt[Scop3P_allSTR_filt['RESIDUE'].isin(['THR','TPO'])].Ssec.value_counts(normalize=True))\n",
    "print (\"\\nTYR/PTR in Ssec\",Scop3P_allSTR_filt[Scop3P_allSTR_filt['RESIDUE'].isin(['TYR','PTR'])].Ssec.value_counts(normalize=True))\n",
    "\n",
    "Scop3P_allSTR_filt=Scop3P_allSTR_filt[(Scop3P_allSTR_filt['ASA']!='-') & (Scop3P_allSTR_filt['RESIDUE'].isin(['SER','SEP','THR',\\\n",
    "                                                                                        'TPO','TYR','PTR']))]\n",
    "\n",
    "Scop3P_allSTR_filt['Accessibility']=Scop3P_allSTR_filt.apply(accessmap,axis=1)\n",
    "print(\"\\nAfter ASA == - filter\",Scop3P_allSTR_filt.shape)\n",
    "\n",
    "\n",
    "access=Scop3P_allSTR_filt[Scop3P_allSTR_filt['Accessibility']=='A']\n",
    "buried=Scop3P_allSTR_filt[Scop3P_allSTR_filt['Accessibility']=='B']\n",
    "interface=Scop3P_allSTR_filt[Scop3P_allSTR_filt['Accessibility']=='I']\n",
    "\n",
    "\n",
    "print (\"Total number of sites in accessible region\", access.shape[0])\n",
    "print (\"Total number of sites in interface region\", interface.shape[0])\n",
    "print (\"Total number of sites in buried region\", buried.shape[0])\n",
    "\n",
    "print (\"%%%%%%%%%%%%%%%%%%%%%%%% Accessible %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\")\n",
    "\n",
    "plt.figure(figsize=(7,5))\n",
    "prob_A = access['RESIDUE'].value_counts(normalize=False)\n",
    "prob_A.plot(kind='bar',color='green',rot=0)\n",
    "plt.title('Accessible residues absolute count')\n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "print (access['RESIDUE'].value_counts(normalize=False))\n",
    "\n",
    "print (access['RESIDUE'].value_counts(normalize=True))\n",
    "\n",
    "\n",
    "print (\"%%%%%%%%%%%%%%%%%%%%%%%  Buried  %%%%%%%%%%%%%%%%%%%%%%%%%%%\")\n",
    "\n",
    "\n",
    "plt.figure(figsize=(7,5))\n",
    "prob_B = buried['RESIDUE'].value_counts(normalize=False)\n",
    "prob_B.plot(kind='bar',color='steelblue',width=0.3,rot=0)\n",
    "plt.title('Buried residues absolute count')\n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "print (buried['RESIDUE'].value_counts(normalize=False))\n",
    "\n",
    "print (buried['RESIDUE'].value_counts(normalize=True))\n",
    "print (\"%%%%%%%%%%%%%%%%%%%%%%%%  Interface  %%%%%%%%%%%%%%%%%%%%%%%%%%%\")\n",
    "\n",
    "\n",
    "plt.figure(figsize=(7,5))\n",
    "prob_I = interface['RESIDUE'].value_counts(normalize=False)\n",
    "prob_I.plot(kind='bar',color='orange',rot=0)\n",
    "plt.title('Interface residues absolute count')\n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "print (interface['RESIDUE'].value_counts(normalize=False))\n",
    "\n",
    "print (interface['RESIDUE'].value_counts(normalize=True))\n",
    "print (\"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\")\n",
    "\n",
    "plt.figure(figsize=(7,5))\n",
    "\n",
    "Scop3P_allSTR_filt['Accessibility'].value_counts().plot.bar(color=['green','orange','steelblue'],rot=0)\n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "\n",
    "\n",
    "#######################################################################################################\n",
    "\n",
    "accessdata=access[['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS_y','RESIDUE','ASA', 'BSA','Ssec']]\n",
    "accessdata.columns=['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS','RESIDUE','ASA', 'BSA','Ssec']\n",
    "accessdata.PDBPOS=accessdata.PDBPOS.astype(int)\n",
    "accessdata=accessdata.merge(accdataSTY,on=['ACC_ID', 'PDBID', 'CHAINID','PDBPOS'])\n",
    "\n",
    "interfacedata=interface[['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS_y','RESIDUE','ASA', 'BSA','Ssec']]\n",
    "interfacedata.columns=['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS','RESIDUE','ASA', 'BSA','Ssec']\n",
    "interfacedata.PDBPOS=interfacedata.PDBPOS.astype(int)\n",
    "interfacedata=interfacedata.merge(accdataSTY,on=['ACC_ID', 'PDBID', 'CHAINID','PDBPOS'])\n",
    "\n",
    "burieddata=buried[['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS_y','RESIDUE','ASA', 'BSA','Ssec']]\n",
    "burieddata.columns=['ACC_ID', 'PDBID', 'CHAINID','UP_POS','PDBPOS','RESIDUE','ASA', 'BSA','Ssec']\n",
    "burieddata.PDBPOS=burieddata.PDBPOS.astype(int)\n",
    "burieddata=burieddata.merge(accdataSTY,on=['ACC_ID', 'PDBID', 'CHAINID','PDBPOS'])\n",
    "\n",
    "###########################################################################################################\n",
    "plt.rcParams['axes.labelsize'] = 16\n",
    "accessdata11_P=accessdata[accessdata['Accesibility'].str.count(',').add(1)==11]\n",
    "accessP=accessdata11_P[['Accesibility']]\n",
    "accessP[['P-5','P-4','P-3','P-2','P-1','P','P1','P2','P3','P4','P5']]=accessP.Accesibility.str.split(\",\",expand=True) \n",
    "accessP=accessP.drop(['Accesibility'],axis=1)\n",
    "accessP=accessP.apply(pd.value_counts, axis=0,normalize=True)\n",
    "accessP.T.plot(kind='bar',rot=0,stacked=True,legend=False,color=['green','steelblue','orange'],title='Accessible residues',figsize=(7,5)) \n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "\n",
    "interfacedata11_P=interfacedata[interfacedata['Accesibility'].str.count(',').add(1)==11]\n",
    "interfaceP=interfacedata11_P[['Accesibility']]\n",
    "interfaceP[['P-5','P-4','P-3','P-2','P-1','P','P1','P2','P3','P4','P5']]=interfaceP.Accesibility.str.split(\",\",expand=True) \n",
    "interfaceP=interfaceP.drop(['Accesibility'],axis=1)\n",
    "interfaceP=interfaceP.apply(pd.value_counts, axis=0,normalize=True)\n",
    "interfaceP.T.plot(kind='bar',rot=0,stacked=True,legend=False,color=['green','steelblue','orange'],title='Accessible residues',figsize=(7,5)) \n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "\n",
    "burieddata11_P=burieddata[burieddata['Accesibility'].str.count(',').add(1)==11]\n",
    "buriedP=burieddata11_P[['Accesibility']]\n",
    "buriedP[['P-5','P-4','P-3','P-2','P-1','P','P1','P2','P3','P4','P5']]=buriedP.Accesibility.str.split(\",\",expand=True) \n",
    "buriedP=buriedP.drop(['Accesibility'],axis=1)\n",
    "buriedP=buriedP.apply(pd.value_counts, axis=0,normalize=True)\n",
    "buriedP.T.plot(kind='bar',stacked=True,rot=0,legend=False,color=['green','steelblue','orange'],title='Accessible residues',figsize=(7,5)) \n",
    "plt.tick_params(axis='x', labelsize=12)\n",
    "\n",
    "\n",
    "print (\"Access, interface and buried 11 data points\",accessdata11_P.shape[0],interfacedata11_P.shape[0],\\\n",
    "      burieddata11_P.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "C    0.564921\n",
      "H    0.293196\n",
      "E    0.140722\n",
      "-    0.001161\n",
      "Name: Ssec, dtype: float64\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Buried'}, ylabel='Ssec'>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATEAAAEeCAYAAADinyKLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAAqS0lEQVR4nO3deXwUVb428KeqestG9oQkJICsRhYhQQVEtoiCsiqyOwiCC15n5o76znjVUUc/OnqdGV5lFUVHw67sKiiyY8JOCHtIgGyE7Evv1VX3DxKGJYRO0p1T1fX7fkRIdyf9KOGh6tSpczhZlmUQQohK8awDEEJIc1CJEUJUjUqMEKJqVGKEEFWjEiOEqBqVGCFE1ajECCGqRiVGCFE1KjFCiKpRiRFCVI1KjBCialRihBBVoxIjhKgalRghRNWoxAghqkYlRghRNSoxQoiq6VgHIGxZrVasXLkSW7duRVZWFsxmM4KDg3HPPfdg+PDhGDVqFHQ6+jYhysXR8tTadfHiRcyePRsXLlxAv3790L9/f4SGhqK0tBS//fYb9u3bh5kzZ+K1115jHZWQ26IS0yibzYYxY8YgNzcX//znPzFs2LBbXpORkYHjx49jypQpDBIS4h46T9Co1atXIycnB7Nmzaq3wACgR48e6NGjRwsnI6RxaGBfo7Zs2QIAmDBhAuMkhDQPlZhGnTt3DoGBgYiPj2cdhZBmoRLTqJqaGgQEBLCOQUizUYlpVGBgIMxmM+sYhDQblZhGderUCTU1NcjNzWUdhZBmoRLTqLorkqtXr2achJDmoRLTqPHjx6N9+/b48ssv8csvv9T7mszMTKSmprZwMkIahya7atj1M/YffPBB9OvXDyEhISgrK0N6ejr27NmDZ599Fq+88grrqITcFpWYxtXdO7llyxZkZWXBYrEgODgY3bp1w4gRIzBy5EgIgsA6JiG3RSVGCFE1GhMjhKgalRghRNXoBnDicbIsQXbZIUtOAADH6cAJRsiSE5KjAi57GSTRAo7TAbwOHK/7z685AeB04HgB4ITajwXwOhNkSbz6NWs/5jgaqyNUYqSJZEm8WkS8HoAM0VwA0VoEp7kAoiUfovUKXNYSiLZiuGzFcNlKILvszXhHDoJfFPSB8dAHxEMf1BaG4E4wBLWDzj8GvL4VZJcNsuwCx+vA6/w99Z9KFI4G9skdybILktMMTjBCclTDXnES1uKDcJSdgL38JETrZdYRAV4PvX8s9IEJ0AW2gSn8XvhFPQB9QAwk0Qpe51dbuMTXUImRW8iSE5JoAy8YYa/KgrVwD6zF+2EtOQzJXs46XqNwugCYwnvAFJGEgJiHYAzrDllyghdM4AQqNV9AJUYAAJLLAcgiXI5KVOesR03eFtjLMgFZZB3NszgBpvAe8I95CAFthsHYqiMklx28PgAcR9e51IhKTMMklx2QJbhsJajKWYuaixvhqMpiHatFcYIf/KIfQHCHp+AfOwiQXOD1tESRmlCJaYwk2gAAoqUAVdnfoSb3BzirL7ANpRCcLgCB8Y8gpNM0GEK7AjLA60ysY5E7oBLTAEkSAUmEs+YiqrLXoCb3R4jmfNaxFE3n1xpB7cchuNMUCIZgcIIRHE8X85WISsyHSS47ABk1l35A2Yl5cFZls46kSsbQbmjVcSKC2o0CZEAwBLGORK5DJeaDJOfVFVsrzn6NijNL4bKVME7kIzgdAmIHIqzbf8EQ3Amc4AeO41in0jwqMR/ictZAFq0oO/EZqs6vhuyyso7ks0yR9yGi119gDOlMZcYYlZjKybIEWbTCac5H2fF/oiZvKyBLrGNphimyT22ZdaEyY4RKTKVkWYbsssFWfAilx/8FW8kh1pE0zRSZjIh7/wJjaFcqsxZGJaZCktMM0VKIy2mvwl56lHUcch1TRNLVI7PQu8EJJppA2wKoxFREql0ZouTIB6g6v4JOGxXMFNELEb3egDGkK3g93YzuTVRiKlC3tE31xY0oOfIBJEcF60jETYFtRyGqz3vgBAN4wcg6jk+iElM4yWmG05yLot9egb38BOs4pAl4Yyii+ryPgNiBtESQF1CJKZQk2iBLDhQfehfVOd8DoN8mtfOPGYjovp+A1wXQ7UweRCWmQJJoQ1X2GpQe+wiSs5p1HOJBnC4AEb1eR6v2Y8Hr/FjH8QlUYgoiiTZIzkoU7JwNe1kG6zjEi0wRvdC6/6cQDKE08N9MVGIKIYkWmAt2oijtVciimXUc0gI43oCwbi8jpOuMqzeY03SMJqESY0yWXJAlO64ceLN27ItojSG4C2IHLYVgCqMrmE1AJcaQJNogWgpQsHMmremlcbw+CK0f/Ax+kcl0BbORqMQYkUQLzPnbrp4+NmsXIOI7OITeMwdh97xIg/6NQCXGgCRaUXLkA1Se+4Z1FKJAftH9ETNgAXid/9X9N0mDqMRakORyQhZrULBjBmx0zyNpgC6gDdoMXQ7BL5LGye6ASqyFSC47xJo85G2bSIsUErfw+iDEDlwCY1g3GidrAJVYC5BEGxxV55G/bRJNXiWNwwmI6vMegtqNoiK7DSoxL5NEK+zlJ5H/6zRaaZU0WXCXZxDR81Ua8K8HlZgXSaIF1uJDKNz5LGTJwToOUbmg9uMQ1ec9KrKb0BRhL5GcFpgLd6FgxwwqMOIR1Tnfo/jg25BEOqK/HpWYF0iiBdW5P+LynjmALLKOQ3xIVfYqFB9+j4rsOlRiHiY5Lag6vxpX0l6hlVeJV1RlLUPJkQ+pyGrRlsYeJIkWlJ/5EmUZn7COQnxc5bl/A7wOET3/pPmrllRiHiKJFpRlzkf5yXmsoxCNqDzzJXheh7Duv9d0kdHppAdIogXVFzdRgZEWV35qMcpOzIckWlhHYYZKrJkklx32shO4sv911lGIRpWfmIfyk4s1W2RUYs0gSy64bCUo2DkTkF2s4xANK8uci8pzyyA5tVdkVGLNIIkW5P1CtxIRZSg5+gHs5ZmQNLa0E5VYE0miFQU7Z0A057KOQshVsoSCXc9BclRDSzfiUIk1gSRacOXAG7AVH2QdhZAbSI4KFOyYDtllYx2lxVCJNZIkWlBx9htaD58olr38BK4c/KtmBvqpxBpBEm2wFqWh9OjfWUchpEHV2atRfWmzJmb1U4m5SZZcEM35KNwzB7QbN1GD4v1vwFlzEZLk2/fvUom5SZYcKNg1W1NjDUTdZMmBgu3PQPbx00oqMTdIogVlx+fCWZ3NOgohjSJaL6Nw12yfPq2kErsDWXLBWX0B5ac/Zx2FkCaxXklH2fF/+exAP5XYHciSA4V7XqJldYiqlZ/6HI6qHMiS730fU4k1QBItKM34B5zVOayjENJMMop++yNkyfdm81OJ3YYsueCoykbF6S9YRyHEIxyV51B5LtXnxseoxG5Dlhy4vOcl0HQK4ktKMz6BJJpZx/AoKrF6SE4LSo5+DGfNRdZRCPEo2WVDUdqrPjXITyV2E1kS4ajKQuXZr1hHIcQrLAU7YL2SDllyso7iEVRiN5El59WrkXQaSXxYUfpfqMR8kSTaUHH237S8DvF5LmsRSo/9wydOK6nEriPLIsozP2Mdg5AWUXF2KURLoerXHqMSqyU5zSjN+AcksYZ1FEJahizh8r4/qv5+YCqxWpJoQeW5b1nHIKRF2cuOw5y/TdUrXVCJ4epRWPGhdwAfGegkpDFKMz4BqMTUTbSVoObSD6xjEMKEs/oCzAXqPRrTfIlJTjNKj34ImlJBtKz02P+q9mhM8yXmspehJncL6xiEMHX1aOxXVR6NabrEJKcZJUfoKIwQACg9ps6xMU2XmMtejprcH1nHIEQRnNXZsBYfgKyytfM0W2KSswalGf8AHYUR8h9lx+dCVtlSPZotMXA8anLpiiQh17OVHILTnMc6RqNossQkSUT1hY2QXb63yiUhzVV2/F9wOdVz54omSwySA5XnvmGdghBFqsnbqqpt3jRZYqK1GPbyE6xjEKJMsoSKc99CUsmZiuZKTBItqDjzFesYhChazYUNgEpWt9BciYETUH1hLesUhCias+YiRHM+6xhu0VSJybIES+FuSI5K1lEIUbzK8ytUsTOSpkqMTiUJcV/1xY0Ax7GOcUeaKjHZZYe1aB/rGISogstaBEfledYx7kgzJSa57Kg8lwqaoU+I+6qylkFyKnu6hWZKDJBRlb2KdQhCVKUm9yeAF1jHaJBmSsxlK1fN1RZClMJlL4O97DjrGA3SRIldvSq5g3UMQlSpMms5JKeZdYzb0kSJSU4zzPnbWccgRJXMeT8DvI51jNvSRInxghHWK2msYxCiSpKzGvZS5Z5SaqLEHDWXIDmrWccgRLXMl3dBVuhuYD5fYrIkwpz/C+sYhKiarfigYmfv+3yJSaIVlsJdrGMQomq20mPgBRPrGPXy+RLjBCNsxYdYxyBE1WTRAqelkHWMevl8idnLT0KWHKxjEKJ61qI0yApcnsenS0xy2WHO28o6BiE+wXrlN8ii8uaL+XSJyZITlst7WMcgxCdYiw8BnPIqQ3mJPIgXjLQMNSEeIprzICtwc12fLjHRWgyobCNQQpTMXpbBOsItfLrEnNU5rCMQ4lPMBbsguZR1ocxnS0yWZdjLT7GOQYhPsZUcUtx+rb5bYqIFjspzrGMQ4lMclefAC0bWMW7guyUmu+Cg00lCPEqJ9yC7XWJOpxMWS/3L1FosFjidyro5lOMNNCZGiBe47GWsI9zA7RJ744038Oabb9b73FtvvYW33nrLY6E8guPgspWwTkGIzxGtRawj3MDtEktPT8fQoUPrfW7IkCFIS1PWel2iuYB1BEJ8krMml3WEG7hdYqWlpQgLC6v3ubCwMJSUKOuox1Gl/K2mCFEjR/UFyAqaf+l2iYWHh+Ps2bP1Pnf27FmEhIR4KlOzybIL9vKTrGMQ4pNEcz5kl411jGvcLrFBgwZh/vz5OH369A2PnzlzBgsXLsTgwYM9Hq6pJNEKJx2JEeIVoqVQUbcfub36/8svv4x9+/bhiSeeQPfu3REdHY2ioiIcP34cbdq0wR/+8AcvxmwkWYJovcI6BSE+SbQUAhzHOsY1bh+JhYWFYc2aNZg9ezZkWb52RPb8889jzZo1tx0vY0WJ81kI8QWipRA8r5wJr5ysxFXOmkly1uDSj4/DWXORdRRCfFKHCacVM3O/0ZvJlZWV4dixY6ioqMDgwYMREhICu90OvV4PnlfIDQCcDpKzhnUKQnyWZC8H79+adQwAjSgxWZbx0Ucf4dtvv4XT6QTHcVizZg1CQkLw4osvonfv3pgzZ443s7qN4/V0OkmIF0kKWuHV7UOnRYsWITU1FXPmzMGqVatuWGt78ODB2LFjhzfyNRmtq0+I96jy6uTq1asxZ84cPPfcc3C5XDc8l5CQgEuXLnk8XFPJsrLu4yTE1yipxNw+EisqKkLPnj3rfU6v18NqVdDGmpLrzq8hhDSdgnYDd7vEoqOjce5c/etznTlzBm3atPFYqOZS0i0RhPgiJR2JuX06+eijj2LevHlITEzEvffeCwDgOA45OTn48ssv8dRTT3krY+NRiakGb2iFiN5vI/iusS3+3lfHdWVABmr/BdQ9du2hm15TOxYs3/z6m56v+7V8w9e/+XVy7T/yTe8rX/f6/zwn3/B+t/78n+e9T+cf1SLv4w6354nZbDbMmDEDR44cQWxsLPLz8xEfH4/CwkL06tULX3zxBQwGg7fzukW0lSLn+2TWMUgDjKHdEZL0DnSh3XD0yhU8EBMDu90Oh8MBjuOu/QDQ4K8BgLv6wQ0/1z138+O47vNJy3K5XCgru3EtMpPJhKCgoGZ93UZNdnW5XNi4cSP27NmDsrIyhISEYMCAARg5ciR0ukZPOfMa0VqMnLX3sY5B6tGqwySY7n4Jer9I/PvECfzvwYM4X1GBV5OT8e4DD2DhwoWorKxkHVP1eJ6/5UfdXwD1PV7fr6//+PrP5TgOQ4cORURERKMy5eXl3bKc19ixY/Hhhx8267/VJ2fsu+zlyP6uN+sYpBanC0BEr7+AbzMaNS7go/378WVmJqodN06D2TBmDPqGhGDRokUQReWMuZBbPfvss4iLi2vU59jtdhw6dOiGx6KiotCxY8dmZXG7xMrKymCz2RAbG3vtsRUrVuDcuXN48MEHFbWKhexyIGtlF9YxNM8Q3AVhSe9ACO+FY1eK8V56On7MzkZD33DZM2bAVlSEVatWtVhO0njPP/88oqOjWccA0Iirk6+//joWL1587eN58+bh7bffxqZNm/Diiy/ihx9+8ErAJuEEcAq5r0uLgtqNRfRjuxA1bCPWlPgj+dtU9Fu+HD/cocAAoHdqKtq0a4f+/fu3SFbSNIIgsI5wjdsllpmZib59+177eMWKFXjuueeQnp6OKVOmYOnSpV4J2BSyZIdgCGUdQ1t4EyJ6v4moMcfA3/se3juajdiFCzHjpy04VVrq9pepsNsx9Lvv8NBDD6FDhw5eDEyaQ5UlVllZifDwcABXV3ItKSnB2LFXL4sPHToUOTnK2VlIllzgjVRiLUEX1B7Rg75G7LhjyA4ZgWd+3o7YBQvw8YEDqLA3bZPVA5cv44+7dmH8+PEIDaXfRyVSUom5fUkxJCQERUVXdzlJS0tDVFQU2rVrBwAQRRGSpKS5WTIEYwjrED4tIP5RBHR/DYbANlh7Lgsf/bwcx4qLPfb15x89igFxcZg2bRoWLlwIh4PuhVUSPz8/1hGucbvE+vXrh08//RTl5eVYunQpUlJSrj2XnZ3d6CsV3sXR6aQ38DqEd/8jhPaT4eKN+OTQYSw4+gNKvHTL2aTNm3F6+nSMHz8eqampXnkP0nhGo1E5y26hEaeTr776KmJiYvDJJ58gPj7+hmV3Nm7ciN69lTOlgeN1EOh00mN0/rGIHrAEMeMyURA1Hs9t243o+Qvwt99+81qB1UlKTUVkbKyirn5rXatWrRS1WbbbR2IRERG3Hbz/6quvFDNbHwA4wUBjYh7gHzMQQT3/B/qgdvgxJwcfrlqNA5cvt2gGs9OJAatW4cDkySgoKMCZM2da9P3JrZS0sxnQhJVd61RXV+PChQuIjIxE69bKWOGxDscJ0Pkp594udeER2u0l6Dv8DtAF4LMjR/DpkV9w2cxuEbzjJSV4bts2LB43Dp9//rni9jjVmuDgYPUM7O/evRvp6el45ZVXbnh8wYIFmDdv3rV1xUaMGIG///3virr1iEqscXhTJCKT3gHXejAumy14b1c6Vpw+DYdLGcsafX3iBB6MjcW0adOwYMEC2GzK2fdQa0JDQ5X1Z72hJ1esWHHLzbJ79+7F3Llz0blzZ4wfPx7nz5/HypUrcc8992DGjBleDdsYgimcdQRV8It6AK16vQl9cCf8evESPvh+Lfbm57OOVa9ZP/+MPlOnYsLEifj311/DB++YU4Xw8HBF3UTfYImdOnUKL7zwwg2Pff/99zAajfjiiy8QGRl57fFNmzYpqsT0AcpZ30yJQrrOgqHzLPCGYCzOyMDcQ18it1r5+xLcv3w5CmbNwsMPP4ytW7eyjqNJSpu712CJlZaWIiEh4YbH9u7di6SkpBsKbNCgQVi/fr13EjaRYAoHJ/hBdiloxVnGeEMIInq/BS72EZTaHfhgbzq+PXUKVhXdbG13ufDAypXImDoV+fn5OHHiBOtImtPcpXM8rcESCwgIuGHZ6QsXLqCiouKWZaoDAwMVNtkVkEQrDCFdYC89yjoKc8bwXgjp/VfoQxKxryAf76/fiO25uaxjNdm58nJM27IFqaNGobS0FJdb+IqplvE8D6NRWfclNzhP7K677sK2bduufbxt2zZwHHfLzbl5eXnXbklSCo7XwRSayDoGU606TUPUyN8QPmQFvsmTcM9XXyFl9RpVF1idNWfPYl5GBqZMmQJ/f3/WcTQjKChIccskNXgkNn36dLz00kvX7ptcu3YtOnfujKSkpBtet3PnTnTt2tWrQRuL1/nDFNEblVnLWEdpUbwuEBG9/wdoMxLVTgl/278fX2dmokZBkxM95b937EDfmBhMmjQJS5cuVdzZgC8KDQ1V3P/nBo/EUlJS8Prrr+P48eNYt24devbsiblz595wZaK4uBj79u3DwIEDvR62sUzh9e/O5IsMoYmIHroaMWMP47jpQUzc/BMSFi3CvCNHfLLA6gxYuRKG4GAMHz6cdRRNaNOmjaKmVwA+urJrHdnlQNaqu31645Cgu8bDL/H30PtFYfmpU/j44EGcuWkdc1+XEBSE09OnY+tPP+Ho0aOs4/i03/3ud9cWflAKZVWqh8mSA/qgdnBWZbOO4lk6f0T0fBV8whOwSTz+duAAlhz/HpVNXPpG7S5VV+PJTZuwduRIFBcXI1+h89x8wfUrOyuFb5cYZBhDE32mxPRBdyEs+V0IEX1wsqQU7/30CzZlZ0Py3YNpt/2Qk4OPDh3Cf0+ejAULFqCmpoZ1JJ8TERGhyAnGyllPwwt4IQCmMPWPiwXEP4boETsQ/ehP2FAeggdSl+G+1FRsOH+eCuw6b+7di/QrVzBl6lRF3dvnKxISEhQ1U7+OTx+JcTwPU2TSnV+oRLwBET3+BL7dBIicAX8/eBALj21EGd0z2KCUNWuQO2sWRo0ahbVr17KO41M6dOigqNVq6vh0iQGAMbgT6wiNoguIR3jyu+Aj++J8RSXe+2Un1mVlQVTYZW2lknB1s5HsGTPQp08fHDhwgHUkn3Hz3TtK4fMlBnAwhNwNR8Up1kEaFBCXgsAef4Y+MAGbsrPx4a+rcLh2OXDSOEUWCx5btw5bxo1DUVERLl26xDqS6gUGBsJkMrGOUS/fLzFej8D4YShTZInxCOvxB+jaT4Us+GHukSP47MgWXLFYWAdTvR15eXjzt9/wzqRJWLBgAaqqqlhHUrX4+HiIoqi4OWKAj88Tq+OoysbFTUPv/MIWIpiiEZH8DvjogcitMeP9tDSsOnMGTjpl9LiNY8bgAdpVvNlGjBiBpKQkRa2tX0d5ibxAFxALwY/9bsV+rfsj+pEfETNqN3Y6OuDhNd8hcelSpJ46RQXmJSPXrUONIGDcuHGso6jaXXfdpcgCAzRSYpAlBMaxOxILTXwBUaMOoFX/L7DwfBU6LfkCo9atQ1phIbNMWtLr228RT7uKN5m/vz+Cg4NZx7gt5Z3gegGv80dQuzEtejM4bwhDZNJfgZiHUWyz4f096Vh++jRsdErT4irsdqR8/z12jR+Py5cv4/z586wjqUpiYqLibvq+njaOxAAYw3uA0wV4/X1MEUmIHrYBsaPTkIbuGLluPTou+QJLMzOpwBhKLyzEn3bvpl3FmyA5OVmR88PqaKbEZMmBgJgBXvv6IV2eQdTINIQOWoavLjlw99KleOS777ArL89r70ka57MjR7A+JwfTpk1T9B9KJQkODkZYWBjrGA3SxNXJOtWXfsDlPXPu/EI38YZWiOj1Jri4Eah0iPggPR3/PnkSZh9e+sYXnJk+HVxFBZbRruJ3NGDAAAwYMAB6vZ51lNvSxJhYnYCYgQAnAHLztiEzhnZHSNLb0IV2x4HLl/H+xs34+eJFD6Uk3pa0bBnynn0WgwcPxvbt21nHUbTevXsrusAAjZWYLEswRSTBVry/SZ/fqsMEmO5+GXq/SHxz8iT+d8PXyKqo8GxI4nU1DgceWr0a+ydNQn5+Ps6ePcs6kiJFRUWpYulvTZUYr/NDq/ZjG1diOn9E3vsXcPFjYHEBb+/fjy8z16DK4fBeUOJ1GcXFeGHbNiwYNw5LliyhXcXr0bNnT1WsBqKpMTHg6i5I2d8nQxYbvrXHENwZoUnvQBfeGxnFxXgvLR0/5uTQ0jc+ZsmwYXiyXTssmD8fdo0uKnk7r7zyCgICvH9Fv7k0c3XyGllCUNtRt306sO1oRD22E1HDNmFtaSD6fJuKvsuWYzMtPuiTnt26FRetVkycOFGRa2WxEh8fr8j7JOujuSMxAHBU5eDipiH/eYA3IuLeV8EnjIcDOvzj4EEsyshAOa3dpQkmnQ75s2bh9LFj+Pnnn1nHUYRRo0ahZ8+eir3V6HrqqFoP0/lFwRh+L1z2coQnvwsh4gGcLS/Hez9vx/qsLLi01+uaZhNFPLBiBTKmTkVBQYHmdxUXBAGJiYmqKDBAoyXGCSaEP/QVeL0/NmRl4cNfVuDolSusYxGGzpWX43dbtuCbUaNQUlKCIg2v5dazZ09VnVprs8R4AQZDIJK++QbH6aoUqbXq7Fn0i43FzKlTMX/+fFitVtaRWhzHcRg4cKCq7mjQ5JgYcPUUYu6hQ/jz7t2soxCFSZ80CbGyrMldxbt27YoxY8bAaDS69fr09HQ8/fTTt31eEAScPHnSU/HqpckjMeDqYO6LvXrhb2lpdJsQuUH/lSuRP2sWhg8fjs2bN7OO06IGDx7sdoFd7/HHH8dDDz10y+MtMa6m2RIDAA7As927Y+7hw6yjEAURJQl9li3D6enTkZ+fr5ldxRMSEhASEtKkz01MTMTo0aM9G8hN6rj84CWBBgNev/9+CCoaxCQto25X8eHDhyMuLo51nBaRkpKi+Psk66PpEgMAP70eE7p2ZR2DKNAPOTn4+PBhTJ48GYGBgazjeFV8fDyio6ObfFXSarWirKzslh8tsRO7Zgf2r1dssaDt4sWw0qKFpB7bx49HV6MRS5YsgcvVvBVQlGrmzJmIi4trdIndaWB/0KBBWLRoUXPjNUjTY2J1/HU6/Pn++/HXvXtZRyEKNHT1auTOmoWRI0di3bp1rON4XEJCAqKiopo1N2zChAl49NFHb3m8JRZUpBIDEGAw4JXkZHyekYG86mrWcYjCSACSUlNx3kd3FR82bFizx8Latm2Lfv36eShR42h+TKyOnufx/4cMufMLiSZdtlgwcsMGpKSkICEhgXUcj+nYsSMiIyNVNUP/ZlRitfSCgGHt2qFfbCzrKEShfr10CX9NS8OkSZPQqlUr1nGaTafTYfTo0aqanV8fKrHr+Ot0WPLII1Dv30nE2z7cvx/b8vMxdepU1SxVcztDhgxp0sRWpVH374KHcRyH+KAgTO/WDUszM1nHIQr1+Nq1uDBzJsaOHYvVq1ezjtMkUVFRSE5O9ti8sJMnT2L9+vX1PpeSkuLVxRVpikU9ym02tF28GNW0BDW5jTCTCRdnzsRve/Zg3759rOM0CsdxeO6555p9RRK48xQLANi6dSvatm3brPdpCJVYPSxOJxYeO4Y/7djBOgpRsPtjYrBr/HisXLkS2dnZrOO4rU+fPkhJSVH9WFgdGhOrh79ej+d79sRdwcGsoxAFq9tV/KmnnlLNruKBgYE+VWAAldhtGXgeXw0fToP8pEGfHTmCDTk5mDptmiruOxw5cqQqdjBqDCqx29AJAnpFReH/3Xcf6yhE4Z7atAklLhfGjx/POkqDOnXqhHbt2lGJaUmgwYA3+/bFAzExrKMQhUtatgxRcXEYOHAg6yj10uv1PjEnrD5UYnfgr9dj/dixCPGB+TTEe2ocDgxcvRp9+/VD586dWce5xSOPPOKTBQZQibkl2GDAiscfZx2DKNyx2l3Fx40bh4iICNZxrunRowe6d++uijG7pqASc4NRp0P/uDj8vndv1lGIwi09cQKpZ85g6tSpipgN37p1azz22GM+exQGUIm5LdBgwPsDBqBXVBTrKEThZm7dios2G/Ndxf38/DBlyhSfPQKrQyXWCP46HTaOG4cgH/5bjXjG/cuXIzgyEikpKUzen+M4TJo0CSaTSdUrVLiDSqwROI5DmNGIfw8fzjoKUTibKKLvihVISk5GYmJii7//ww8/jOjoaNXfpO4OKrFG8tPrkdK2LWb16ME6ClG4M+XleHrLFowePRrR0dEt9r6JiYlITk726XGw61GJNUGgwYB/DR5M88fIHa06cwaLjh/HlKlT4efn5/X3i4yMxOjRo31+HOx6VGJN5K/XY8v48eiuoEvpRJle3r4dp6uqMGnyZK+OTxmNRkydOlVTBQZQiTVLkF6PHRMnomMTNxwl2tFvxQqYgoMxfMQIr3x9juMwYcIE+Pv7+/xA/s2oxJqB4ziEGAzYO3ky2gQFsY5DFEyUJPRZvhzdundHz549Pfq1OY7Dk08+ibi4OE0M5N+MSqyZeJ5HmNGIvZMmIaIFxjyIel2sqsKEzZsxYsQIxHpoLweO4/DEE0+gY8eOmhnIvxmVmAfoBAGtAwKwZ9IktNLoNxJxz8bsbHx8+DCmTJnS7CWbOY7DuHHj0KlTJ80WGEAl5jEGQUDbVq2wfcIE+GnwkJ647409e3CguBhTpk5t1rI4Y8eORefOnTVdYACVmEeZdDrcHRaGH594Anqe/teS2xuyejXg54fHm7iwwJgxY9ClSxfNFxhAJeZxfno9+rRuje9GjwavsatExH0SgN6pqejctSuSk5Mb9bljxozB3XffTQVWi0rMC/z1egyJj8fyxx+Hjo7IyG3U7Sr+8MMPu72r+KhRo6jAbkJ/wrwkwGDAY3fdhW1PPYVAjU0+JO779dIlvJ2ejokTJ95xV/GRI0finnvuoQK7CW3Z5mVWpxO51dUYtHIlCs1m1nGIQv0wdiySW7XC4sWLIYriDc/xPI/Ro0eja9euVGD1oBJrAU6XC+V2OwavXImTpaWs4xCFujBzJmoKC7Hmul3F69YEi4yMpAK7DTqdbAF6QUCEnx/SpkzBEDfHPoj29E5NRdv27dGvXz8AQHh4OF544QVER0dTgTWAjsRamMXpxOu7d2Pu4cOsoxAF6hsbix1PPok9e/agf//+0Ov14OniUIOoxBgwOxxYl5WFGVu2wOFysY5DFGbTuHEYGhcHkwLW6FcDKjFGLE4nsioq8MiaNbhMA/4EQFDtrloD27RBAJ0+uo1KjCGHy4VqhwNPbNiAnbm5rOMQhhLDw/HTE08g0t8fJrptrVGoxBTA7HRi2alT+OP27TA7nazjkBY2qWtXfP7II/DT6egujyagElMIq9OJSocDkzZtwg46KtOE1gEB+HzYMAxOSEAATYhuMioxhbE4nVh+6hR+T0dlPosDMLtHD3w8aBCMggBDM1ayIFRiimRxOlFFR2U+qWtYGFIfewydQ0MRSIP3HkElpmBmpxMrT5/Gy7/+SkdlKmcQBLzVty/+mJQEoyBAoLlfHkMlpnDW647KttNRmSoNaNMG344YgXCTiaZOeAGVmEqYnU5syMrCn3fvxqWqKtZxiBtCjEbMHTIET3buDH8auPcaKjEVcbhccEkSlp06hbf27UNBTQ3rSKQeAsdh8t13Y+6QIfDT6Wjel5dRiamQw+WCS5bxVWYm3tm3D0UWC+tIBFfHvX6XmIh3H3wQAXo9gujUsUVQiamYXRThkmUszsjA+2lpKLFaWUfSJD+dDrN79MCbffvCIAhUXi2MSswH2GrLbP6RI/hg/36U22ysI2lCkMGA/+rdG6/16QOB42jKBCNUYj7E6nTCJcv41+HD+Hj/flQ5HKwj+aQwkwl/Sk7Gy717g+M4mm3PGJWYD7LUzilbn5WFhceOYXdeHug3ufmi/f3x5/vvx+wePcDh6s5WhD0qMR8mSRLMogiL04nPMzLwZWYmciorWcdSlQC9HqM7dsTsHj1wf0wMANDVRoWhEtMImyhCBnCqtBSfHjmCNWfOoIbuAqiXURDwaPv2eLZ7d6S0bQuHy4VWtEChYlGJaVC1wwEdz2Pz+fOYf/QoduTmav50U+A4DElIwIzu3TGyQwe4JImKSyWoxDRMkmXUOBywu1xYcfo0NmdnY3d+/rUxNV/H4eqa9tO7dcOELl0AAIEGA63ppTJUYgQA4JIk1Did8NPpkFlSgrXnzmHrhQs4WFQEyUe+RTgAiREReKhNG4xo3x4PxsVB4Hn46XS0U7uKUYmRejlcLthEEXpBwPHiYmy9eBF78vKQVliISruddTy3BOr16BUdjftat8bw9u2vDczzHEf3MvoQKjHiFlGSYK49UrtsNmN/YSEyS0uRU1l57UdBTQ2TozaDICAmIAAJrVqhV1QUBsbH477WrRHl7w+LKMIkCHRF0YdRiZEmc0kSLKIIlyRBLwgwCgJKrFbkVlfjbFkZTpWWIruyEheqqnCxqgrW2te6ZBkuWYYky9c+rg8HIMLfH3GBgYgNDERc7Y+OISFoFxyMmIAARPr7w1+ng1UUIUoSjIJA87c0hkpMI9LT0/H000/jtddew8yZM+t9TZcuXTBo0CAsWrTII+8pShIsTickWYZBEMBzHDiOAw+A4zhwtT/zHAdJliHLMiTg6s+yDB3PwylJcLhckGUZep6Hn15PA+/kBnSMTbxGx/NuT1PgOQ7gONy82ryuduCdkNuhSzKEEFWjEiOEqBodp2uM1WpFWVkZ6xiEeAyVmMZ8+umn+PTTT1nHIMRjqMQ0ZsKECXj00Ufrfe6ZZ55p4TSENB+VmMa0bdsW/fr1Yx2DEI+hgX1CiKpRiRFCVI1KjBCialRihBBVoxIjhKga3QBOCFE1OhIjhKgalRghRNWoxAghqkYlRghRNSoxQoiqUYkRQlSNSowQompUYoQQVaMSI4SoGpUYIUTVqMQIIapGJUYIUTUqMUKIqlGJEUJUjUqMEKJqVGKEEFWjEiOEqBqVGCFE1ajECCGqRiVGCFE1KjFCiKr9H6LG4SR7A5BfAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "######################################################## Secondary structure overall and A/B/I \n",
    "plt.figure(figsize=(7,5))\n",
    "print (Scop3P_allSTR_filt.Ssec.value_counts(normalize=True))\n",
    "labels=['C','H','E','-']\n",
    "Scop3P_allSTR_filt.Ssec.value_counts(normalize=True).plot(kind='pie', fontsize=18,colors=['goldenrod','darkcyan','grey','red'])\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(1, 3,figsize=(15,10))\n",
    "access.Ssec.value_counts(normalize=True).plot(kind='pie', fontsize=10,ax=ax[0],\\\n",
    "                                              labels=labels,colors=['goldenrod','darkcyan','grey'],title='Access')\n",
    "\n",
    "interface.Ssec.value_counts(normalize=True).plot(kind='pie', fontsize=10,ax=ax[1],\\\n",
    "                                                labels=labels,colors=['goldenrod','darkcyan','grey'],title='Interface')\n",
    "\n",
    "\n",
    "buried.Ssec.value_counts(normalize=True).plot(kind='pie', fontsize=10,ax=ax[2],\\\n",
    "                                             labels=labels,colors=['goldenrod','darkcyan','grey'],title='Buried')\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}