Finsih SequencePreprocessor.encode_integer() and update tests

breimanntools · Jun 28, 2024 · 8d38d41 · 8d38d41
1 parent f2da477
commit 8d38d41
Show file tree

Hide file tree

Showing 3 changed files with 244 additions and 7 deletions.
diff --git a/aaanalysis/data_handling/_seq_preproc.py b/aaanalysis/data_handling/_seq_preproc.py
@@ -231,7 +231,7 @@ def encode_integer(list_seq: Union[List[str], str] = None,
     def get_aa_window(seq: str = None,
                       pos_start: int = 0,
                       pos_stop: Optional[int] = None,
-                      window_size: Optional[int] = 5,
+                      window_size: Optional[int] = None,
                       index1: bool = False,
                       gap: str = '-',
                       accept_gap: bool = True,
@@ -250,7 +250,7 @@ def get_aa_window(seq: str = None,
             The starting position (>=0) of the window.
         pos_stop : int, optional
             The ending position (>=``pos_start``) of the window. If ``None``, ``window_size`` is used to determine it.
-        window_size : int, default=5
+        window_size : int, optional
             The size of the window (>=1) to extract. Only used if ``pos_stop`` is ``None``.
         index1 : bool, default=False
             Whether position index starts at 1 (if ``True``) or 0 (if ``False``),

diff --git a/examples/data_handling/sp_encode_integer.ipynb b/examples/data_handling/sp_encode_integer.ipynb
@@ -12,18 +12,255 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "outputs": [],
    "source": [
     "import aaanalysis as aa\n",
+    "import pandas as pd\n",
     "\n",
-    "seq = \"AACDEFGHII\"\n",
+    "list_seq = [\"AACDEFGHIY\", \"IIHGFECDAY\"]\n",
     "sp = aa.SequencePreprocessor()"
    ],
    "metadata": {
-    "collapsed": false
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:38.975453869Z",
+     "start_time": "2024-06-28T20:36:36.239449900Z"
+    }
    },
    "id": "6529c65f51e1c14f"
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "Provide the sequence as ``seq`` parameter to obtain a feature matrix (``X``) and the respective ``features``, which are integer amino acid representation at given residue positions: "
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "1163b1302190f5a5"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DataFrame shape: (2, 10)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "<IPython.core.display.HTML object>",
+      "text/html": "<style type=\"text/css\">\n#T_6d5a5 thead th {\n  background-color: white;\n  color: black;\n}\n#T_6d5a5 tbody tr:nth-child(odd) {\n  background-color: #f2f2f2;\n}\n#T_6d5a5 tbody tr:nth-child(even) {\n  background-color: white;\n}\n#T_6d5a5 th {\n  padding: 5px;\n  white-space: nowrap;\n}\n#T_6d5a5  td {\n  padding: 5px;\n  white-space: nowrap;\n}\n</style>\n<table id=\"T_6d5a5\" style='display:block; max-height: 300px; max-width: 100%; overflow-x: auto; overflow-y: auto;'>\n  <thead>\n    <tr>\n      <th class=\"blank level0\" >&nbsp;</th>\n      <th id=\"T_6d5a5_level0_col0\" class=\"col_heading level0 col0\" >P1</th>\n      <th id=\"T_6d5a5_level0_col1\" class=\"col_heading level0 col1\" >P2</th>\n      <th id=\"T_6d5a5_level0_col2\" class=\"col_heading level0 col2\" >P3</th>\n      <th id=\"T_6d5a5_level0_col3\" class=\"col_heading level0 col3\" >P4</th>\n      <th id=\"T_6d5a5_level0_col4\" class=\"col_heading level0 col4\" >P5</th>\n      <th id=\"T_6d5a5_level0_col5\" class=\"col_heading level0 col5\" >P6</th>\n      <th id=\"T_6d5a5_level0_col6\" class=\"col_heading level0 col6\" >P7</th>\n      <th id=\"T_6d5a5_level0_col7\" class=\"col_heading level0 col7\" >P8</th>\n      <th id=\"T_6d5a5_level0_col8\" class=\"col_heading level0 col8\" >P9</th>\n      <th id=\"T_6d5a5_level0_col9\" class=\"col_heading level0 col9\" >P10</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th id=\"T_6d5a5_level0_row0\" class=\"row_heading level0 row0\" >1</th>\n      <td id=\"T_6d5a5_row0_col0\" class=\"data row0 col0\" >1</td>\n      <td id=\"T_6d5a5_row0_col1\" class=\"data row0 col1\" >1</td>\n      <td id=\"T_6d5a5_row0_col2\" class=\"data row0 col2\" >2</td>\n      <td id=\"T_6d5a5_row0_col3\" class=\"data row0 col3\" >3</td>\n      <td id=\"T_6d5a5_row0_col4\" class=\"data row0 col4\" >4</td>\n      <td id=\"T_6d5a5_row0_col5\" class=\"data row0 col5\" >5</td>\n      <td id=\"T_6d5a5_row0_col6\" class=\"data row0 col6\" >6</td>\n      <td id=\"T_6d5a5_row0_col7\" class=\"data row0 col7\" >7</td>\n      <td id=\"T_6d5a5_row0_col8\" class=\"data row0 col8\" >8</td>\n      <td id=\"T_6d5a5_row0_col9\" class=\"data row0 col9\" >20</td>\n    </tr>\n    <tr>\n      <th id=\"T_6d5a5_level0_row1\" class=\"row_heading level0 row1\" >2</th>\n      <td id=\"T_6d5a5_row1_col0\" class=\"data row1 col0\" >8</td>\n      <td id=\"T_6d5a5_row1_col1\" class=\"data row1 col1\" >8</td>\n      <td id=\"T_6d5a5_row1_col2\" class=\"data row1 col2\" >7</td>\n      <td id=\"T_6d5a5_row1_col3\" class=\"data row1 col3\" >6</td>\n      <td id=\"T_6d5a5_row1_col4\" class=\"data row1 col4\" >5</td>\n      <td id=\"T_6d5a5_row1_col5\" class=\"data row1 col5\" >4</td>\n      <td id=\"T_6d5a5_row1_col6\" class=\"data row1 col6\" >2</td>\n      <td id=\"T_6d5a5_row1_col7\" class=\"data row1 col7\" >3</td>\n      <td id=\"T_6d5a5_row1_col8\" class=\"data row1 col8\" >1</td>\n      <td id=\"T_6d5a5_row1_col9\" class=\"data row1 col9\" >20</td>\n    </tr>\n  </tbody>\n</table>\n"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "X, features = sp.encode_integer(list_seq=list_seq)\n",
+    "\n",
+    "# Convert to DataFrame for visualization\n",
+    "df_encode = pd.DataFrame(X, columns=features)\n",
+    "aa.display_df(df=df_encode, show_shape=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:39.094402731Z",
+     "start_time": "2024-06-28T20:36:38.987669931Z"
+    }
+   },
+   "id": "ae99247d4a174b85"
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "You can adjust the used ``alphabet`` to change the considered characters:"
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "bd44a3454980db67"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DataFrame shape: (2, 3)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "<IPython.core.display.HTML object>",
+      "text/html": "<style type=\"text/css\">\n#T_9581a thead th {\n  background-color: white;\n  color: black;\n}\n#T_9581a tbody tr:nth-child(odd) {\n  background-color: #f2f2f2;\n}\n#T_9581a tbody tr:nth-child(even) {\n  background-color: white;\n}\n#T_9581a th {\n  padding: 5px;\n  white-space: nowrap;\n}\n#T_9581a  td {\n  padding: 5px;\n  white-space: nowrap;\n}\n</style>\n<table id=\"T_9581a\" style='display:block; max-height: 300px; max-width: 100%; overflow-x: auto; overflow-y: auto;'>\n  <thead>\n    <tr>\n      <th class=\"blank level0\" >&nbsp;</th>\n      <th id=\"T_9581a_level0_col0\" class=\"col_heading level0 col0\" >P1</th>\n      <th id=\"T_9581a_level0_col1\" class=\"col_heading level0 col1\" >P2</th>\n      <th id=\"T_9581a_level0_col2\" class=\"col_heading level0 col2\" >P3</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th id=\"T_9581a_level0_row0\" class=\"row_heading level0 row0\" >1</th>\n      <td id=\"T_9581a_row0_col0\" class=\"data row0 col0\" >1</td>\n      <td id=\"T_9581a_row0_col1\" class=\"data row0 col1\" >2</td>\n      <td id=\"T_9581a_row0_col2\" class=\"data row0 col2\" >3</td>\n    </tr>\n    <tr>\n      <th id=\"T_9581a_level0_row1\" class=\"row_heading level0 row1\" >2</th>\n      <td id=\"T_9581a_row1_col0\" class=\"data row1 col0\" >3</td>\n      <td id=\"T_9581a_row1_col1\" class=\"data row1 col1\" >2</td>\n      <td id=\"T_9581a_row1_col2\" class=\"data row1 col2\" >1</td>\n    </tr>\n  </tbody>\n</table>\n"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Show integer encoding with smaller alphabet\n",
+    "list_seq = [\"ABC\", \"CBA\"]\n",
+    "ALPHABET = \"ABC\"\n",
+    "X, features = sp.encode_integer(list_seq=list_seq, alphabet=ALPHABET)\n",
+    "\n",
+    "# Convert to DataFrame for visualization\n",
+    "df_encode = pd.DataFrame(X, columns=features)\n",
+    "aa.display_df(df=df_encode, show_shape=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:39.149782676Z",
+     "start_time": "2024-06-28T20:36:39.095604192Z"
+    }
+   },
+   "id": "280dd1afa9267756"
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "Change the ``gap`` symbol (default=``-``) as follows:"
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "f53bf29b98c3fd9e"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DataFrame shape: (2, 3)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "<IPython.core.display.HTML object>",
+      "text/html": "<style type=\"text/css\">\n#T_8e4b3 thead th {\n  background-color: white;\n  color: black;\n}\n#T_8e4b3 tbody tr:nth-child(odd) {\n  background-color: #f2f2f2;\n}\n#T_8e4b3 tbody tr:nth-child(even) {\n  background-color: white;\n}\n#T_8e4b3 th {\n  padding: 5px;\n  white-space: nowrap;\n}\n#T_8e4b3  td {\n  padding: 5px;\n  white-space: nowrap;\n}\n</style>\n<table id=\"T_8e4b3\" style='display:block; max-height: 300px; max-width: 100%; overflow-x: auto; overflow-y: auto;'>\n  <thead>\n    <tr>\n      <th class=\"blank level0\" >&nbsp;</th>\n      <th id=\"T_8e4b3_level0_col0\" class=\"col_heading level0 col0\" >P1</th>\n      <th id=\"T_8e4b3_level0_col1\" class=\"col_heading level0 col1\" >P2</th>\n      <th id=\"T_8e4b3_level0_col2\" class=\"col_heading level0 col2\" >P3</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th id=\"T_8e4b3_level0_row0\" class=\"row_heading level0 row0\" >1</th>\n      <td id=\"T_8e4b3_row0_col0\" class=\"data row0 col0\" >1</td>\n      <td id=\"T_8e4b3_row0_col1\" class=\"data row0 col1\" >2</td>\n      <td id=\"T_8e4b3_row0_col2\" class=\"data row0 col2\" >3</td>\n    </tr>\n    <tr>\n      <th id=\"T_8e4b3_level0_row1\" class=\"row_heading level0 row1\" >2</th>\n      <td id=\"T_8e4b3_row1_col0\" class=\"data row1 col0\" >3</td>\n      <td id=\"T_8e4b3_row1_col1\" class=\"data row1 col1\" >2</td>\n      <td id=\"T_8e4b3_row1_col2\" class=\"data row1 col2\" >0</td>\n    </tr>\n  </tbody>\n</table>\n"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Show integer encoding with other gap ('*')\n",
+    "list_seq = [\"ABC\", \"CB*\"]\n",
+    "ALPHABET = \"ABC\"\n",
+    "X, features = sp.encode_integer(list_seq=list_seq, alphabet=ALPHABET, gap=\"*\")\n",
+    "\n",
+    "# Convert to DataFrame for visualization\n",
+    "df_encode = pd.DataFrame(X, columns=features)\n",
+    "aa.display_df(df=df_encode, show_shape=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:39.172622592Z",
+     "start_time": "2024-06-28T20:36:39.111814958Z"
+    }
+   },
+   "id": "91341e3430083a63"
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "If one sequence is smaller than the other, gaps will be included either at the N-terminus or C-terminus (default), which is called padding. Adjust the padding using the ``pad_at`` (``N`` or ``C``) parameter:"
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "4691599de5bd9c99"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DataFrame shape: (2, 3)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "<IPython.core.display.HTML object>",
+      "text/html": "<style type=\"text/css\">\n#T_e676c thead th {\n  background-color: white;\n  color: black;\n}\n#T_e676c tbody tr:nth-child(odd) {\n  background-color: #f2f2f2;\n}\n#T_e676c tbody tr:nth-child(even) {\n  background-color: white;\n}\n#T_e676c th {\n  padding: 5px;\n  white-space: nowrap;\n}\n#T_e676c  td {\n  padding: 5px;\n  white-space: nowrap;\n}\n</style>\n<table id=\"T_e676c\" style='display:block; max-height: 300px; max-width: 100%; overflow-x: auto; overflow-y: auto;'>\n  <thead>\n    <tr>\n      <th class=\"blank level0\" >&nbsp;</th>\n      <th id=\"T_e676c_level0_col0\" class=\"col_heading level0 col0\" >P1</th>\n      <th id=\"T_e676c_level0_col1\" class=\"col_heading level0 col1\" >P2</th>\n      <th id=\"T_e676c_level0_col2\" class=\"col_heading level0 col2\" >P3</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th id=\"T_e676c_level0_row0\" class=\"row_heading level0 row0\" >1</th>\n      <td id=\"T_e676c_row0_col0\" class=\"data row0 col0\" >1</td>\n      <td id=\"T_e676c_row0_col1\" class=\"data row0 col1\" >2</td>\n      <td id=\"T_e676c_row0_col2\" class=\"data row0 col2\" >3</td>\n    </tr>\n    <tr>\n      <th id=\"T_e676c_level0_row1\" class=\"row_heading level0 row1\" >2</th>\n      <td id=\"T_e676c_row1_col0\" class=\"data row1 col0\" >2</td>\n      <td id=\"T_e676c_row1_col1\" class=\"data row1 col1\" >0</td>\n      <td id=\"T_e676c_row1_col2\" class=\"data row1 col2\" >0</td>\n    </tr>\n  </tbody>\n</table>\n"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Show default padding (at C-Termius)\n",
+    "list_seq = [\"ABC\", \"B\"]\n",
+    "ALPHABET = \"ABC\"\n",
+    "X, features = sp.encode_integer(list_seq=list_seq, alphabet=ALPHABET)\n",
+    "\n",
+    "# Convert to DataFrame for visualization\n",
+    "df_encode = pd.DataFrame(X, columns=features)\n",
+    "aa.display_df(df=df_encode, show_shape=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:39.174009891Z",
+     "start_time": "2024-06-28T20:36:39.156212641Z"
+    }
+   },
+   "id": "48ff3a51801c9d62"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DataFrame shape: (2, 3)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": "<IPython.core.display.HTML object>",
+      "text/html": "<style type=\"text/css\">\n#T_449d7 thead th {\n  background-color: white;\n  color: black;\n}\n#T_449d7 tbody tr:nth-child(odd) {\n  background-color: #f2f2f2;\n}\n#T_449d7 tbody tr:nth-child(even) {\n  background-color: white;\n}\n#T_449d7 th {\n  padding: 5px;\n  white-space: nowrap;\n}\n#T_449d7  td {\n  padding: 5px;\n  white-space: nowrap;\n}\n</style>\n<table id=\"T_449d7\" style='display:block; max-height: 300px; max-width: 100%; overflow-x: auto; overflow-y: auto;'>\n  <thead>\n    <tr>\n      <th class=\"blank level0\" >&nbsp;</th>\n      <th id=\"T_449d7_level0_col0\" class=\"col_heading level0 col0\" >P1</th>\n      <th id=\"T_449d7_level0_col1\" class=\"col_heading level0 col1\" >P2</th>\n      <th id=\"T_449d7_level0_col2\" class=\"col_heading level0 col2\" >P3</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th id=\"T_449d7_level0_row0\" class=\"row_heading level0 row0\" >1</th>\n      <td id=\"T_449d7_row0_col0\" class=\"data row0 col0\" >1</td>\n      <td id=\"T_449d7_row0_col1\" class=\"data row0 col1\" >2</td>\n      <td id=\"T_449d7_row0_col2\" class=\"data row0 col2\" >3</td>\n    </tr>\n    <tr>\n      <th id=\"T_449d7_level0_row1\" class=\"row_heading level0 row1\" >2</th>\n      <td id=\"T_449d7_row1_col0\" class=\"data row1 col0\" >0</td>\n      <td id=\"T_449d7_row1_col1\" class=\"data row1 col1\" >0</td>\n      <td id=\"T_449d7_row1_col2\" class=\"data row1 col2\" >2</td>\n    </tr>\n  </tbody>\n</table>\n"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Show N-terminal padding\n",
+    "list_seq = [\"ABC\", \"B\"]\n",
+    "ALPHABET = \"ABC\"\n",
+    "X, features = sp.encode_integer(list_seq=list_seq, alphabet=ALPHABET, pad_at=\"N\")\n",
+    "\n",
+    "# Convert to DataFrame for visualization\n",
+    "df_encode = pd.DataFrame(X, columns=features)\n",
+    "aa.display_df(df=df_encode, show_shape=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "end_time": "2024-06-28T20:36:39.616720327Z",
+     "start_time": "2024-06-28T20:36:39.477176504Z"
+    }
+   },
+   "id": "43798b8ac9b7e61c"
   }
  ],
  "metadata": {

diff --git a/examples/data_handling/sp_encode_one_hot.ipynb b/examples/data_handling/sp_encode_one_hot.ipynb
@@ -18,7 +18,7 @@
     "import aaanalysis as aa\n",
     "import pandas as pd\n",
     "\n",
-    "list_seq = [\"AACDEFGHII\", \"IIHGFECDAA\"]\n",
+    "list_seq = [\"AACDEFGHIY\", \"IIHGFECDAY\"]\n",
     "sp = aa.SequencePreprocessor()"
    ],
    "metadata": {
@@ -33,7 +33,7 @@
   {
    "cell_type": "markdown",
    "source": [
-    "Provide the sequence as ``seq`` parameter to obtain a feature matrix (``X``) and the respective ``features``, which are binary representation of amino acid at given residue positions: "
+    "Provide the sequence as ``seq`` parameter to obtain a feature matrix (``X``) and the respective ``features``, which are binary representation of each amino acid at given residue positions: "
    ],
    "metadata": {
     "collapsed": false