notebooks/semantic_similarity_wang.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Calculating Wang's Semantic Similarity between two GO Terms\n",
    "\n",
    "1. Setup\n",
    "2. Calculate Wang's semantic similarity using optional *part_of* relationship\n",
    "3. Calculate Wang's semantic similarity using researcher-set edge weights. \n",
    "4. Calculate Wang's semantic similarity usint *is_of* relationship only\n",
    "\n",
    "## 1) Setup\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a list of all GO IDs that will be compared"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Researcher-provided GO terms related to smell\n",
    "go_a = 'GO:0007608'\n",
    "go_b = 'GO:0050911'\n",
    "go_c = 'GO:0042221'\n",
    "\n",
    "# Optional relationships. (Relationship, is_a, is required and always used)\n",
    "relationships = {'part_of'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "goids = {go_a, go_b, go_c}\n",
    "\n",
    "# Annotations for plotting\n",
    "go2txt = {\n",
    "    go_a:'GO TERM A',\n",
    "    go_b:'GO TERM B',\n",
    "    go_c:'GO TERM C'}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read the GO DAG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  EXISTS: go-basic.obo\n",
      "go-basic.obo: fmt(1.2) rel(2020-11-18) 47,326 GO Terms; optional_attrs(relationship)\n"
     ]
    }
   ],
   "source": [
    "from goatools.base import get_godag\n",
    "\n",
    "godag = get_godag(\"go-basic.obo\", optional_attrs={'relationship'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a printing function for this notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_details(go_a, go_b, val):\n",
    "    \"\"\"Print concise and informative report: GO terms and their semantic similarity\"\"\"\n",
    "    pattern = ('go_a: {GOa} {GOa_name}\\n'\n",
    "               'go_b: {GOb} {GOb_name}\\n'\n",
    "               'wang: {VAL:.8f}\\n')\n",
    "    print(pattern.format(\n",
    "        GOa=go_a, GOa_name=godag[go_a].name,\n",
    "        GOb=go_b, GOb_name=godag[go_b].name,\n",
    "        VAL=val))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2) Calculate Wang's semantic similarity using *part_of* relationship"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Instantiate a Wang's Semantic Similarity object\n",
    "We choose to use the optional relationship, *part_of*, in addition the required *is_a* relationships for this example.    \n",
    "Load all GO IDs that you will be comparing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from goatools.semsim.termwise.wang import SsWang\n",
    "\n",
    "# goids: researcher-provided GO terms\n",
    "wang_r1 = SsWang(goids, godag, relationships)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize researcher GO terms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " GoSubDag:   3 sources in  14 GOs rcnt(True). 0 alt GO IDs\n",
      " GoSubDag: namedtuple fields: NS level depth reldepth GO alt GO_name dcnt D1 childcnt REL REL_short rel id\n",
      " GoSubDag: relationships: {'part_of'}\n",
      "    3 usr  14 GOs  WROTE: smell_r1.png\n"
     ]
    }
   ],
   "source": [
    "from goatools.gosubdag.gosubdag import GoSubDag\n",
    "from goatools.gosubdag.plot.gosubdag_plot import GoSubDagPlot\n",
    "\n",
    "r1_png = 'smell_r1.png'\n",
    "r1_gosubdag = GoSubDag(goids, godag, relationships)\n",
    "GoSubDagPlot(r1_gosubdag, go2txt=go2txt).plt_dag(r1_png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![smell_r1](images/smell_r1.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "wang: 0.59684773\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_a, go_b)\n",
    "print_details(go_a, go_b, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.14114914\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_a, go_c)\n",
    "print_details(go_a, go_c, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.38020047\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_b, go_c)\n",
    "print_details(go_b, go_c, val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3) Calculate Wang's semantic similarity using researcher-set edge weights\n",
    "\n",
    "### Print the default edge weights in the Wang configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wang Semantic Similarity Configuration:\n",
      "    Optional relationships: part_of\n",
      "    Edge weights:\n",
      "        0.80000000 is_a\n",
      "        0.60000000 part_of\n",
      "\n"
     ]
    }
   ],
   "source": [
    "wang_r1.prt_cfg()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use researcher-specified edge weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "relationship2weight = {\n",
    "    'is_a': 0.9,\n",
    "    'part_of': 0.9\n",
    "}\n",
    "wang_r1 = SsWang(goids, godag, relationships, relationship2weight)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "wang: 0.64019253\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_a, go_b)\n",
    "print_details(go_a, go_b, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.16922363\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_a, go_c)\n",
    "print_details(go_a, go_c, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.35627471\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r1.get_sim(go_b, go_c)\n",
    "print_details(go_b, go_c, val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4) Calculate Wang's semantic similarity with *is_a* relationships only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "wang_r0 = SsWang(goids, godag)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculate Wang's semantic similarity using *is_a* relationship only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " GoSubDag:   3 sources in  14 GOs rcnt(True). 0 alt GO IDs\n",
      " GoSubDag: namedtuple fields: NS level depth GO alt GO_name dcnt D1 id\n",
      " GoSubDag: relationships: set()\n",
      "    3 usr  14 GOs  WROTE: smell_r0.png\n"
     ]
    }
   ],
   "source": [
    "r0_png = 'smell_r0.png'\n",
    "r0_gosubdag = GoSubDag(goids, godag)\n",
    "GoSubDagPlot(r0_gosubdag, go2txt=go2txt).plt_dag(r0_png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![smell_r1](images/smell_r0.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Instantiate a Wang's Semantic Similarity object\n",
    "We use the required *is_a* relationships only.    \n",
    "\n",
    "Load all GO IDs that you will be comparing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "wang: 0.07122873\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r0.get_sim(go_a, go_b)\n",
    "print_details(go_a, go_b, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0007608 sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.14114914\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r0.get_sim(go_a, go_c)\n",
    "print_details(go_a, go_c, val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "go_a: GO:0050911 detection of chemical stimulus involved in sensory perception of smell\n",
      "go_b: GO:0042221 response to chemical\n",
      "wang: 0.32164130\n",
      "\n"
     ]
    }
   ],
   "source": [
    "val = wang_r0.get_sim(go_b, go_c)\n",
    "print_details(go_b, go_c, val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Copyright (C) DV Klopfenstein and Haibao Tang, All rights reserved."
   ]
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}