From 0313cb52cc030a693c5679a0dcc61b28dd883221 Mon Sep 17 00:00:00 2001
From: Mark Wilkinson <markw@illuminae.com>
Date: Wed, 7 Aug 2024 12:54:16 +0200
Subject: [PATCH] working environment for phenotypes

---
 content/FLAIR-GG/phenotypefrequency.ipynb | 113 ++++++++++++++++------
 1 file changed, 82 insertions(+), 31 deletions(-)

diff --git a/content/FLAIR-GG/phenotypefrequency.ipynb b/content/FLAIR-GG/phenotypefrequency.ipynb
index 5cce723..4f71e83 100644
--- a/content/FLAIR-GG/phenotypefrequency.ipynb
+++ b/content/FLAIR-GG/phenotypefrequency.ipynb
@@ -1,20 +1,13 @@
 {
  "cells": [
-  {
-   "cell_type": "markdown",
-   "id": "039f7a57-a580-4d29-b126-25b51f67140e",
-   "metadata": {},
-   "source": [
-    "# Results explorer for the EJP Counting Case\n",
-    "<img src = \"https://vp.bgv.cbgp.upm.es/images/flair-gg-logo.png\" width = 700/>"
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "ed7ed56c-0088-4335-8d0c-7b9fdf2def85",
    "metadata": {},
    "source": [
-    "## Welcome to the FLAIR-GG Counting Case Analytics Notebook\n",
+    "## Welcome to the FLAIR-GG Phenotype Frequency Analytics Notebook\n",
+    "\n",
+    "The Phenotype Frequency Data Service will count the number of times a phenotype appears in a registry (on a per-patient basis - only counts once per patient)\n",
     "\n",
     "Please run the first cell to set-up the analytics environment\n",
     "\n",
@@ -71,25 +64,25 @@
     "# type,frequency\n",
     "# http://purl.obolibrary.org/obo/NCIT_C131922,31\n",
     "# http://purl.obolibrary.org/obo/NCIT_C136154,22\n",
-    "# http://www.orpha.net/ORDO/Orphanet_98896,32\n",
-    "# http://purl.obolibrary.org/obo/NCIT_C20197,76\n",
-    "# http://purl.obolibrary.org/obo/NCIT_C68615,87\n",
-    "# http://purl.obolibrary.org/obo/NCIT_C83164,87\n",
-    "# http://purl.obolibrary.org/obo/NCIT_C13306,24\n",
     "\n",
     "response = requests.get(url)\n",
     "response = json.loads(response.content)\n",
-    "# print(response)\n",
-    "site = [\"sample1\", \"sample2\"]\n",
-    "count = [100, 3]\n",
+    "#print(response)\n",
+    "\n",
+    "data = {}\n",
     "for provider in response.keys():\n",
     "    print(\"Provider: {}\".format(provider))\n",
-    "    data = response[provider]\n",
+    "    alllines = response[provider]\n",
+    "    data[provider] = []\n",
     "\n",
-    "    data = data.splitlines().pop()\n",
+    "    lines = alllines.splitlines()  # this is the CSV\n",
+    "    lines.pop(0)  # get rid of header\n",
+    "    # print(data)\n",
+    "    for line in iter(lines):\n",
+    "        [pheno, freq] = line.split(\",\")\n",
+    "        data[provider].append([pheno, freq])\n",
     "\n",
-    "    site.append(provider)\n",
-    "    count.append(int(data))\n",
+    "# print(data)\n",
     "print(\"DONE\")"
    ]
   },
@@ -100,15 +93,73 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "source = pd.DataFrame({\n",
-    "    'site': site,\n",
-    "    'count': count\n",
-    "})\n",
-    "\n",
-    "alt.Chart(source).mark_bar().encode(\n",
-    "    x='count',\n",
-    "    y='site'\n",
-    ")"
+    "data_list = []\n",
+    "for provider, phenotypes in data.items():\n",
+    "    for phenotype, frequency in phenotypes:\n",
+    "        data_list.append({\"Provider\": provider, \"Phenotype\": phenotype, \"Frequency\": frequency})\n",
+    "\n",
+    "# Create a DataFrame\n",
+    "df = pd.DataFrame(data_list)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ac295f7c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Simple Bar Chart\n",
+    "bar_chart = alt.Chart(df).mark_bar().encode(\n",
+    "    x='Phenotype:N',\n",
+    "    y='Frequency:Q',\n",
+    "    color='Provider:N',\n",
+    "    column='Provider:N'\n",
+    ").properties(\n",
+    "    title='Phenotype Frequencies by Provider'\n",
+    ")\n",
+    "\n",
+    "bar_chart"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "86bee1fb",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Grouped Bar Chart\n",
+    "grouped_bar_chart = alt.Chart(df).mark_bar().encode(\n",
+    "    x=alt.X('Phenotype:N', title='Phenotype'),\n",
+    "    y=alt.Y('Frequency:Q', title='Frequency'),\n",
+    "    color=alt.Color('Provider:N', title='Provider'),\n",
+    "    column=alt.Column('Provider:N', title='Provider')\n",
+    ").properties(\n",
+    "    title='Grouped Phenotype Frequencies by Provider'\n",
+    ")\n",
+    "\n",
+    "grouped_bar_chart"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "54916d07",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Heatmap\n",
+    "heatmap = alt.Chart(df).mark_rect().encode(\n",
+    "    x='Phenotype:N',\n",
+    "    y='Provider:N',\n",
+    "    color=alt.Color('Frequency:Q', scale=alt.Scale(scheme='viridis')),\n",
+    "    tooltip=['Provider', 'Phenotype', 'Frequency']\n",
+    ").properties(\n",
+    "    title='Phenotype Frequencies Heatmap'\n",
+    ")\n",
+    "\n",
+    "heatmap"
    ]
   },
   {