3. Supervised learning - Phenotypes.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Copyright (C) 2016 Glencoe Software, Inc.\n",
    "All Rights Reserved.\n",
    "\n",
    "This program is free software; you can redistribute it and/or modify\n",
    "it under the terms of the GNU General Public License as published by\n",
    "the Free Software Foundation; either version 2 of the License, or\n",
    "(at your option) any later version.\n",
    "\n",
    "This program is distributed in the hope that it will be useful,\n",
    "but WITHOUT ANY WARRANTY; without even the implied warranty of\n",
    "MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the\n",
    "GNU General Public License for more details.\n",
    "\n",
    "You should have received a copy of the GNU General Public License along\n",
    "with this program; if not, write to the Free Software Foundation, Inc.,\n",
    "51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Working with the analytical data stored on the OMERO server"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " * Train the classifier using positive and negtive control cells ([scikit-learn](http://scikit-learn.org/stable/modules/svm.html))\n",
    " * Use the trained classifier to count the number of negative and positive cells at each doseage level ([scikit-learn](http://scikit-learn.org/stable/modules/svm.html))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook demonstarates how to retrieve analytical data stored as an [OMERO.table](http://downloads.openmicroscopy.org/omero/5.2.4/api/slice2html/omero/grid/Table.html) from the [OMERO.server](http://www.openmicroscopy.org/site/products/omero) and use it to run supervised machine learning algorithms using [scikit-learn](http://scikit-learn.org/stable/modules/svm.html).\n",
    "\n",
    "The example below uses the output of the image processing performed using the [CellProfiler](http://cellprofiler.org/) on the [SBS Bioimage](http://cellprofiler.org/examples.html#SBS_Bioimage_CNT) dataset. Derived features (parameters describing cells' shape, instenisty and texture) are then used to build classifier using [scikit-learn]((http://scikit-learn.org/stable/modules/svm.html) that can distinguish between the positive control and the negative control cells. Finally the classifier is used to score the number of postive and negative cells at the different dose levels.\n",
    "\n",
    "The video tutorial for this notebook can be found on [Glencoe Software, Inc.](http://glencoesoftware.com/webinars.html) website (look for: OMERO Plus for High Content Screening & Analysis: The Gold Standard in Data Management & Integration)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Connect to OMERO server"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Username: your_username\n",
      "OMERO Password: ········\n",
      "Host: your.server.address\n",
      "Port: 4064\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "import omero\n",
    "from omero.gateway import BlitzGateway\n",
    "\n",
    "import gs_analytics\n",
    "from getpass import getpass\n",
    "\n",
    "from getpass import getpass\n",
    "conn = BlitzGateway(\n",
    "    raw_input(\"Username: \"), getpass(\"OMERO Password: \"),\n",
    "    host=raw_input(\"Host: \"), port=int(raw_input(\"Port: \")))\n",
    "conn.connect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "matplotlib.rcParams['figure.figsize'] = (12.0, 7.0)\n",
    "\n",
    "from IPython.display import display, HTML\n",
    "import pprint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "import sklearn.svm as svm\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Request OMERO.table object"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "OMERO.table file id: 66214\n",
      "First ten columns:\n",
      "['ImageNumber',\n",
      " 'ObjectNumber',\n",
      " 'Metadata_Column',\n",
      " 'Metadata_Row',\n",
      " 'Metadata_SBS_Controls',\n",
      " 'Metadata_SBS_Doses',\n",
      " 'Metadata_Well',\n",
      " 'AreaShape_Area',\n",
      " 'AreaShape_Center_X']\n",
      "\n",
      "Table contains 17876 rows\n"
     ]
    }
   ],
   "source": [
    "shared_resources = conn.c.sf.sharedResources()\n",
    "original_file_id = raw_input(\"OMERO.table file id: \")\n",
    "original_file = omero.model.OriginalFileI(original_file_id, False)\n",
    "omero_table = shared_resources.openTable(original_file)\n",
    "column_names = [col.name for col in omero_table.getHeaders()]\n",
    "number_of_rows = omero_table.getNumberOfRows()\n",
    "print \"First ten columns:\\n%s\\n\" % pprint.pformat(column_names[0:9])\n",
    "print \"Table contains %s rows\" % number_of_rows"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Select columns with numerical data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selected 78 columns.\n"
     ]
    }
   ],
   "source": [
    "black_list = [\n",
    "    'ImageNumber', 'ObjectNumber', 'Metadata_Column', 'Metadata_Row',\n",
    "    'Metadata_SBS_Controls', 'Metadata_SBS_Doses', 'Metadata_Well'\n",
    "]\n",
    "column_indices = []\n",
    "for column_name in column_names:\n",
    "    if column_name in black_list:\n",
    "        continue\n",
    "    column_indices.append(column_names.index(column_name))\n",
    "print \"Selected %i columns.\" % len(column_indices)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read in data and the column containing the class labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "table_data = omero_table.read(column_indices, 0, number_of_rows)\n",
    "data = []\n",
    "for index in range(number_of_rows):\n",
    "    row_values = [column.values[index] for column in table_data.columns]\n",
    "    data.append(row_values)\n",
    "label_column_index = column_names.index('Metadata_SBS_Doses')\n",
    "label_column = omero_table.read([label_column_index], 0, number_of_rows)\n",
    "label_column = label_column.columns[0].values\n",
    "class_labels = np.array(label_column)\n",
    "\n",
    "negative_dose = 0\n",
    "positive_dose = 10\n",
    "label_values = [negative_dose, positive_dose]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train the classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def train_with_feature_selection(x, y):\n",
    "    '''\n",
    "    Train the classifier with the feature selection.\n",
    "    '''\n",
    "    etc = LinearSVC(penalty=\"l1\", dual=False)\n",
    "    etc = etc.fit(x, y)\n",
    "    model = SelectFromModel(etc, prefit=True)\n",
    "    x_new = model.transform(x)\n",
    "    print \"Selected %i features using the linear SVC model.\" % x_new.shape[1]\n",
    "    clf = svm.SVC(kernel='linear', C=1)\n",
    "    clf.fit(x_new, y)\n",
    "    return {'clf': clf, 'model': model}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 3152 rows for label 0\n",
      "Found 1692 rows for label 10\n"
     ]
    }
   ],
   "source": [
    "clf_samples = []\n",
    "clf_labels = []\n",
    "for value in label_values:\n",
    "    label_query = [int(x) == int(value) for x in class_labels]\n",
    "    positions = [i[0] for i in enumerate(label_query) if i[1]]\n",
    "    samples = [data[i] for i in positions]\n",
    "    labels = [value] * len(samples)\n",
    "    print \"Found %i rows for label %s\" % (len(labels), value)\n",
    "    clf_samples.extend(samples)\n",
    "    clf_labels.extend(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selected 42 features using the linear SVC model.\n"
     ]
    }
   ],
   "source": [
    "clf = train_with_feature_selection(clf_samples, clf_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classify all the objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_scores(clf, data, class_labels, positive_value, negative_value):\n",
    "    '''\n",
    "    Find the number of positive and negative controls for each label.\n",
    "    '''\n",
    "    result_pos = []\n",
    "    result_neg = []\n",
    "    print \"Scoring positive values: %s and negative values: %s\" \\\n",
    "        % (positive_value, negative_value)\n",
    "    for i in range(11):\n",
    "        labels = [int(x) == int(i) for x in class_labels]\n",
    "        Y = [label[0] for label in enumerate(labels) if label[1]]\n",
    "        X = [data[x] for x in Y]\n",
    "        result = clf.predict(X)\n",
    "        neg = [a for a in clf.predict(X) if a == negative_value]\n",
    "        pos = [a for a in clf.predict(X) if a == positive_value]\n",
    "        result_pos.append(100.0 * float(len(pos)) / float(len(result)))\n",
    "        result_neg.append(100.0 * float(len(neg)) / float(len(result)))\n",
    "        print \"Label %s; Positive cells: %s, Negative cells: %s\" % \\\n",
    "            (i, result_pos[i], result_neg[i])\n",
    "    return {'negative': result_neg, 'positive': result_pos}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifing 17876 objects using 4844 samples\n",
      "Scoring positive values: 10 and negative values: 0\n",
      "Label 0; Positive cells: 0.697969543147, Negative cells: 99.3020304569\n",
      "Label 1; Positive cells: 3.84364820847, Negative cells: 96.1563517915\n",
      "Label 2; Positive cells: 2.79441117764, Negative cells: 97.2055888224\n",
      "Label 3; Positive cells: 11.8569636136, Negative cells: 88.1430363864\n",
      "Label 4; Positive cells: 48.5155195682, Negative cells: 51.4844804318\n",
      "Label 5; Positive cells: 91.7638483965, Negative cells: 8.2361516035\n",
      "Label 6; Positive cells: 96.6966966967, Negative cells: 3.3033033033\n",
      "Label 7; Positive cells: 97.7339181287, Negative cells: 2.26608187135\n",
      "Label 8; Positive cells: 97.2346786248, Negative cells: 2.76532137519\n",
      "Label 9; Positive cells: 96.3527851459, Negative cells: 3.64721485411\n",
      "Label 10; Positive cells: 97.6359338061, Negative cells: 2.36406619385\n"
     ]
    }
   ],
   "source": [
    "print \"Classifing %i objects using %i samples\" % \\\n",
    "    (len(data), len(clf_samples))\n",
    "new_data = clf['model'].transform(data)\n",
    "scores = get_scores(clf['clf'], new_data, class_labels, label_values[1], label_values[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualise the result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def plot_result(X_values, Y_values):\n",
    "    plt.figure()\n",
    "    plt.xlabel(\"Dose\")\n",
    "    plt.ylabel(\"% of cell type\")\n",
    "    plt.plot(X_values, Y_values['positive'], c='r', marker='v', label=\"positive\", linewidth=2)\n",
    "    plt.plot(X_values, Y_values['negative'], c='b', marker='o', label=\"negative\", linewidth=2)\n",
    "    plt.legend(\n",
    "        [\"Positive cells\", \"Negive cells (dose=0)\"],\n",
    "        bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)\n",
    "    plt.grid(True)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6IAAAG2CAYAAABlIDsjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112a78b10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_result(range(1,12), scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Disconnect from the OMERO server"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "conn._closeSession()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}