{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Barplot of Samples by Country" ] }, { "cell_type": "markdown", "metadata": { "id": "0YZ_s4Vpg8kc" }, "source": [ "## Introduction" ] }, { "cell_type": "markdown", "metadata": { "id": "skdQa-hRsqS5" }, "source": [ "This notebook recreates [Supplement Figure 1 from the Pf7 paper](https://figshare.com/ndownloader/files/38478530) - a bar plot which shows the number of samples in the Pf7 release, broken down by country. Each bar also details the number of samples passing QC (or not) per country.\n", "\n", "**This notebook should take approximately 1 minute to run.**" ] }, { "cell_type": "markdown", "metadata": { "id": "bKys7ae1s60J" }, "source": [ "## Setup" ] }, { "cell_type": "markdown", "metadata": { "id": "SbXBdw7ItfvI" }, "source": [ "Install and import the malariagen Python package:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "SpUWpCBOmYqO" }, "outputs": [], "source": [ "!pip install -q --no-warn-conflicts malariagen_data\n", "import malariagen_data" ] }, { "cell_type": "markdown", "metadata": { "id": "aiUVx5GXtlCI" }, "source": [ "Import required python libraries that are installed at colab by default." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "CjDVXchOn_CD" }, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import collections\n", "from google.colab import drive" ] }, { "cell_type": "markdown", "metadata": { "id": "BNKb5_mjs-ZE" }, "source": [ "## Access Pf7 Data" ] }, { "cell_type": "markdown", "metadata": { "id": "_UZmQjrB1IYa" }, "source": [ "We use the malariagen data package to load the release data." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "wRf5_frsoA_M" }, "outputs": [], "source": [ "release_data = malariagen_data.Pf7()\n", "sample_metadata = release_data.sample_metadata()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 278 }, "id": "LdU0QNRz4ZP7", "outputId": "2f4114fa-7f72-4b67-9f87-55e8d71d4655" }, "outputs": [ { "data": { "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SampleStudyCountryAdmin level 1Country latitudeCountry longitudeAdmin level 1 latitudeAdmin level 1 longitudeYearENAAll samples same casePopulation% callableQC passExclusion reasonSample typeSample was in Pf6
0FP0008-C1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR1081237FP0008-CAF-W82.16TrueAnalysis_setgDNATrue
1FP0009-C1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR1081238FP0009-CAF-W88.85TrueAnalysis_setgDNATrue
2FP0010-CW1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR2889621FP0010-CWAF-W86.46TrueAnalysis_setsWGAFalse
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", " \n", "\n", "\n", "\n", " \n", "
\n", "
\n", "
\n" ], "text/plain": [ " Sample Study Country Admin level 1 Country latitude \\\n", "0 FP0008-C 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "1 FP0009-C 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "2 FP0010-CW 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "\n", " Country longitude Admin level 1 latitude Admin level 1 longitude Year \\\n", "0 -10.337093 16.565426 -9.832345 2014.0 \n", "1 -10.337093 16.565426 -9.832345 2014.0 \n", "2 -10.337093 16.565426 -9.832345 2014.0 \n", "\n", " ENA All samples same case Population % callable QC pass \\\n", "0 ERR1081237 FP0008-C AF-W 82.16 True \n", "1 ERR1081238 FP0009-C AF-W 88.85 True \n", "2 ERR2889621 FP0010-CW AF-W 86.46 True \n", "\n", " Exclusion reason Sample type Sample was in Pf6 \n", "0 Analysis_set gDNA True \n", "1 Analysis_set gDNA True \n", "2 Analysis_set sWGA False " ] }, "execution_count": 55, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sample_metadata.head(3)" ] }, { "cell_type": "markdown", "metadata": { "id": "_-JV7B-J2st0" }, "source": [ "We can start exploring the data by answering these questions:\n", "\n", "\n", "* How many samples with QC pass\n", "* How many samples in each country\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "41-XrTGxVEwf", "outputId": "dde988d4-470a-4d51-e78c-d348be984e0e" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "We see 20864 samples of which 16203 QC-pass and 4661 QC fail in the overall Pf7 dataset.\n" ] } ], "source": [ "# Calculate the total number of samples\n", "total_sample_number = sample_metadata.Sample.count()\n", "\n", "# Calculate the number of samples that passed QC\n", "qc_pass_count = (sample_metadata['QC pass'] == True).sum()\n", "\n", "# Calculate the number of samples that failed QC\n", "qc_fail_count = (sample_metadata['QC pass'] == False).sum()\n", "\n", "print(f\"We see {total_sample_number} samples of which {qc_pass_count} QC-pass and {qc_fail_count} QC fail in the overall Pf7 dataset.\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "jQv8qpfw2ra4", "outputId": "5e9df9bd-aaa8-431f-8867-73a1f00b041a" }, "outputs": [ { "data": { "text/plain": [ "Ghana 4090\n", "Mali 1804\n", "Vietnam 1733\n", "Cambodia 1723\n", "Bangladesh 1658\n", "Myanmar 1260\n", "Gambia 1247\n", "Thailand 1106\n", "Laos 1052\n", "Kenya 726\n", "Tanzania 697\n", "Democratic Republic of the Congo 573\n", "Malawi 371\n", "Benin 334\n", "India 316\n", "Cameroon 294\n", "Papua New Guinea 251\n", "Sudan 203\n", "Guinea 199\n", "Colombia 159\n", "Senegal 155\n", "Nigeria 140\n", "Indonesia 133\n", "Mauritania 104\n", "Mozambique 91\n", "Côte d'Ivoire 71\n", "Gabon 59\n", "Burkina Faso 58\n", "Ethiopia 34\n", "Madagascar 25\n", "Peru 21\n", "Uganda 15\n", "Venezuela 2\n", "Name: Country, dtype: int64" ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Calculate the number of samples in each country\n", "sample_metadata['Country'].value_counts()" ] }, { "cell_type": "markdown", "metadata": { "id": "FsFWFPXftxSv" }, "source": [ "## Figure preparation: Defining populations" ] }, { "cell_type": "markdown", "metadata": { "id": "Y5IZIuX51x63" }, "source": [ "Countries are grouped into ten major sub-populations based on their geographic and genetic characteristics.\n", "\n", "The dataframe has a `Population` column that contains abbreviated names, for clarity, we want to display the full name in the figure." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 295 }, "id": "6CA3lBb7oO5N", "outputId": "71a39b6f-d2ef-4d42-dc14-c7070f001bc0" }, "outputs": [ { "data": { "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SampleStudyCountryAdmin level 1Country latitudeCountry longitudeAdmin level 1 latitudeAdmin level 1 longitudeYearENAAll samples same casePopulation% callableQC passExclusion reasonSample typeSample was in Pf6Continent
0FP0008-C1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR1081237FP0008-CAF-W82.16TrueAnalysis_setgDNATrueWest Africa
1FP0009-C1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR1081238FP0009-CAF-W88.85TrueAnalysis_setgDNATrueWest Africa
2FP0010-CW1147-PF-MR-CONWAYMauritaniaHodh el Gharbi20.265149-10.33709316.565426-9.8323452014.0ERR2889621FP0010-CWAF-W86.46TrueAnalysis_setsWGAFalseWest Africa
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", " \n", "\n", "\n", "\n", " \n", "
\n", "
\n", "
\n" ], "text/plain": [ " Sample Study Country Admin level 1 Country latitude \\\n", "0 FP0008-C 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "1 FP0009-C 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "2 FP0010-CW 1147-PF-MR-CONWAY Mauritania Hodh el Gharbi 20.265149 \n", "\n", " Country longitude Admin level 1 latitude Admin level 1 longitude Year \\\n", "0 -10.337093 16.565426 -9.832345 2014.0 \n", "1 -10.337093 16.565426 -9.832345 2014.0 \n", "2 -10.337093 16.565426 -9.832345 2014.0 \n", "\n", " ENA All samples same case Population % callable QC pass \\\n", "0 ERR1081237 FP0008-C AF-W 82.16 True \n", "1 ERR1081238 FP0009-C AF-W 88.85 True \n", "2 ERR2889621 FP0010-CW AF-W 86.46 True \n", "\n", " Exclusion reason Sample type Sample was in Pf6 Continent \n", "0 Analysis_set gDNA True West Africa \n", "1 Analysis_set gDNA True West Africa \n", "2 Analysis_set sWGA False West Africa " ] }, "execution_count": 58, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Define populations in an ordered dictionary\n", "populations = collections.OrderedDict()\n", "populations['SA'] = 'South America'\n", "populations['AF-W'] = 'West Africa'\n", "populations['AF-C'] = 'Central Africa'\n", "populations['AF-NE'] = 'Northeast Africa'\n", "populations['AF-E'] = 'East Africa'\n", "populations['AS-S-E'] = 'Eastern South Asia'\n", "populations['AS-S-FE'] = 'Far-eastern South Asia'\n", "populations['AS-SE-W'] = 'Western Southeast Asia'\n", "populations['AS-SE-E'] = 'Eastern Southeast Asia'\n", "populations['OC-NG'] = 'Oceania'\n", "\n", "# Map continent names into the df by using Population column and populations dictionary\n", "sample_metadata['Continent'] = sample_metadata['Population'].map(populations)\n", "sample_metadata.head(3)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "SxOT-rALoXo6" }, "outputs": [], "source": [ "# Create an ordered dictionary which maps the codes for major sub-populations -from west to east- to a colour code.\n", "population_colours = collections.OrderedDict()\n", "population_colours['SA'] = \"#4daf4a\"\n", "population_colours['AF-W'] = \"#e31a1c\"\n", "population_colours['AF-C'] = \"#fd8d3c\"\n", "population_colours['AF-NE'] = \"#bb8129\"\n", "population_colours['AF-E'] = \"#fecc5c\"\n", "population_colours['AS-S-E'] = \"#dfc0eb\"\n", "population_colours['AS-S-FE'] = \"#984ea3\"\n", "population_colours['AS-SE-W'] = \"#9ecae1\"\n", "population_colours['AS-SE-E'] = \"#3182bd\"\n", "population_colours['OC-NG'] = \"#f781bf\"\n", "\n", "# Map population colours into the df by using Population column and population_colours dictionary\n", "sample_metadata['population_colour'] = sample_metadata['Population'].map(population_colours)" ] }, { "cell_type": "markdown", "metadata": { "id": "kDnv7x4oEqfR" }, "source": [ "## Figure preparation: Sort countries in geographic order" ] }, { "cell_type": "markdown", "metadata": { "id": "pbnbhXZOE8Lu" }, "source": [ "We want to sort the countries on the x-axis in geographic order, which means arranging them from left to right on the chart based on their geographical location, from west to east or by continents.\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "rL3tRRZXN05f" }, "source": [ "### Using longitudes to locate country" ] }, { "cell_type": "markdown", "metadata": { "id": "MiVyYjI6NzGh" }, "source": [ "To do this arrangement, we will use longitude coordinate countries which can be found in the dataset column `Country longitude`." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "8tO2kRKxDvMQ" }, "outputs": [], "source": [ "# Find the average of longitude of samples in each country\n", "mean_population_longitude = sample_metadata.groupby('Population')['Country longitude'].mean()\n", "\n", "# Add a new column that conveys mean population values for each sample\n", "sample_metadata['Population_long'] = sample_metadata['Population'].map(mean_population_longitude)" ] }, { "cell_type": "markdown", "metadata": { "id": "QyCWHC02xG1U" }, "source": [ "## Splitting countries with multi-populations" ] }, { "cell_type": "markdown", "metadata": { "id": "LgwsfEIcMw8-" }, "source": [ "We identified three countries (Kenya, India, and Thailand) where the sampling locations are associated with more than one major sub-population\". For example, Kenya has sampling locations from AF-NE and AF-E, and this causes problems with ordering on country longitude because AF-NE and AF-E become mixed up in the table.\n", "\n", "To accurately represent this diversity, we created a new column called `Country_or_admin1` and `Country_or_admin1_long` in our sample metadata.\n", "\n", "These columns categorizes these countries based on their first-level administrative divisions." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "fc8O-qlsRJSu" }, "outputs": [], "source": [ "# Create a duplicate column with country names\n", "sample_metadata['Country_or_admin1'] = sample_metadata['Country']\n", "sample_metadata['Country_or_admin1_long'] = sample_metadata['Country longitude']\n", "\n", "# Rename each 'Admin level 1' of split-countries\n", "sample_metadata.loc[(sample_metadata['Country'] == 'Kenya') & (sample_metadata['Admin level 1'] == 'Kilifi'), 'Country_or_admin1'] = 'Kenya, Kilifi'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'Kenya') & (sample_metadata['Admin level 1'] == 'Kisumu'), 'Country_or_admin1'] = 'Kenya, Kisumu'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'India') & (sample_metadata['Admin level 1'] == 'Tripura'), 'Country_or_admin1'] = 'India, Tripura'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'India') & (sample_metadata['Admin level 1'] == 'Odisha'), 'Country_or_admin1'] = 'India, Odisha or West Bengal'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'India') & (sample_metadata['Admin level 1'] == 'West Bengal'), 'Country_or_admin1'] = 'India, Odisha or West Bengal'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'Thailand') & (sample_metadata['Admin level 1'] == 'Sisakhet'), 'Country_or_admin1'] = 'Thailand, Sisakhet'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'Thailand') & (sample_metadata['Admin level 1'] == 'Tak'), 'Country_or_admin1'] = 'Thailand, Tak or Ranong'\n", "sample_metadata.loc[(sample_metadata['Country'] == 'Thailand') & (sample_metadata['Admin level 1'] == 'Ranong'), 'Country_or_admin1'] = 'Thailand, Tak or Ranong'\n", "\n", "# Set longitude to that of admin1 for split countries\n", "sample_metadata.loc[\n", " sample_metadata['Country_or_admin1'] != sample_metadata['Country'],\n", " 'Country_or_admin1_long'\n", "] = sample_metadata.loc[\n", " sample_metadata['Country_or_admin1'] != sample_metadata['Country'],\n", " 'Admin level 1 longitude'\n", "]\n", "\n", "# Set longitude to that of admin1 with most samples for countries with more than one admin1 in population\n", "sample_metadata.loc[\n", " sample_metadata['Country_or_admin1'] == 'India, Odisha or West Bengal',\n", " 'Country_or_admin1_long'\n", "] = sample_metadata.loc[\n", " ( sample_metadata['Country'] == 'India' )\n", " & ( sample_metadata['Admin level 1'] == 'Odisha' ),\n", " 'Country_or_admin1_long'\n", "].values[0]\n", "sample_metadata.loc[\n", " sample_metadata['Country_or_admin1'] == 'Thailand, Tak or Ranong',\n", " 'Country_or_admin1_long'\n", "] = sample_metadata.loc[\n", " ( sample_metadata['Country'] == 'Thailand' )\n", " & ( sample_metadata['Admin level 1'] == 'Tak' ),\n", " 'Country_or_admin1_long'\n", "].values[0]" ] }, { "cell_type": "markdown", "metadata": { "id": "KQAMuIsW-YIt" }, "source": [ "Next, we want to arrange the divisions from the same countries adjacent to each other in order to facilitate meaningful comparisons when we look at the figure.\n", "\n", "In order to do that we simply adjust their longitude values." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "TRYqiYzQ-W6y" }, "outputs": [], "source": [ "# Adjust the longitude values to appear first or last\n", "sample_metadata.loc[sample_metadata['Country_or_admin1'] == 'Kenya, Kisumu', 'Country_or_admin1_long'] = 40 # Want it to appear last in AF-NE\n", "sample_metadata.loc[sample_metadata['Country_or_admin1'] == 'Kenya, Kilifi', 'Country_or_admin1_long'] = 34 # Want it to appear first in AF-E\n", "sample_metadata.loc[sample_metadata['Country_or_admin1'] == 'India, Tripura', 'Country_or_admin1_long'] = 90 # Want it to appear first in AS-S-FE\n", "sample_metadata.loc[sample_metadata['Country_or_admin1'] == 'Thailand, Sisakhet', 'Country_or_admin1_long'] = 103 # Want it to appear first in AS-SE-E" ] }, { "cell_type": "markdown", "metadata": { "id": "77H4qbrANbmF" }, "source": [ "Finally, we don't want Kenya returning travellers shown on plot, as we don't know which admin division(s) they could have visited." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "bVz2LvfvNbxl" }, "outputs": [], "source": [ "sample_metadata.loc[sample_metadata['Country_or_admin1'] == 'Kenya', 'Population'] = None" ] }, { "cell_type": "markdown", "metadata": { "id": "NtcJtlPTNjAp" }, "source": [ "### Sorting countries" ] }, { "cell_type": "markdown", "metadata": { "id": "eybfYkWyOiqY" }, "source": [ "Now the countries are ready to sort geographically." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "7M6F-Yv3RD6P", "outputId": "77450a1b-9773-442b-b2f7-663d7f2a27e6" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(36, 6)\n" ] }, { "data": { "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
ContinentPopulationpopulation_colourPopulation_longCountry_or_admin1_longFrequency (number of samples)
Country_or_admin1
PeruSouth AmericaSA#4daf4a-73.157010-74.35684221
ColombiaSouth AmericaSA#4daf4a-73.157010-73.086731159
VenezuelaSouth AmericaSA#4daf4a-73.157010-66.1459362
GambiaWest AfricaAF-W#e31a1c-3.519507-15.3729101247
SenegalWest AfricaAF-W#e31a1c-3.519507-14.470363155
GuineaWest AfricaAF-W#e31a1c-3.519507-10.936960199
MauritaniaWest AfricaAF-W#e31a1c-3.519507-10.337093104
Côte d'IvoireWest AfricaAF-W#e31a1c-3.519507-5.55444671
MaliWest AfricaAF-W#e31a1c-3.519507-3.5221521804
Burkina FasoWest AfricaAF-W#e31a1c-3.519507-1.74566058
GhanaWest AfricaAF-W#e31a1c-3.519507-1.2107114090
BeninWest AfricaAF-W#e31a1c-3.5195072.339713334
NigeriaWest AfricaAF-W#e31a1c-3.5195078.097575140
GabonWest AfricaAF-W#e31a1c-3.51950711.78498959
CameroonWest AfricaAF-W#e31a1c-3.51950712.741504294
Democratic Republic of the CongoCentral AfricaAF-C#fd8d3c23.66075823.660758573
SudanNortheast AfricaAF-NE#bb812932.74189430.005646203
UgandaNortheast AfricaAF-NE#bb812932.74189432.39193215
EthiopiaNortheast AfricaAF-NE#bb812932.74189439.62619534
Kenya, KisumuNortheast AfricaAF-NE#bb812932.74189440.00000064
Kenya, KilifiEast AfricaAF-E#fecc5c35.99970734.000000662
MalawiEast AfricaAF-E#fecc5c35.99970734.300482371
TanzaniaEast AfricaAF-E#fecc5c35.99970734.825685697
MozambiqueEast AfricaAF-E#fecc5c35.99970735.55143791
MadagascarEast AfricaAF-E#fecc5c35.99970746.69861825
India, Odisha or West BengalEastern South AsiaAS-S-E#dfc0eb79.62252584.418059244
India, TripuraFar-eastern South AsiaAS-S-FE#984ea389.83394590.00000072
BangladeshFar-eastern South AsiaAS-S-FE#984ea389.83394590.2773841658
MyanmarWestern Southeast AsiaAS-SE-W#9ecae198.49670296.5102011260
Thailand, Tak or RanongWestern Southeast AsiaAS-SE-W#9ecae198.49670298.791050994
Thailand, SisakhetEastern Southeast AsiaAS-SE-E#3182bd105.173981103.000000112
LaosEastern Southeast AsiaAS-SE-E#3182bd105.173981103.7681571052
CambodiaEastern Southeast AsiaAS-SE-E#3182bd105.173981104.9168731723
VietnamEastern Southeast AsiaAS-SE-E#3182bd105.173981106.5517961733
IndonesiaOceaniaOC-NG#f781bf135.577208117.314980133
Papua New GuineaOceaniaOC-NG#f781bf135.577208145.254007251
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", " \n", "\n", "\n", "\n", " \n", "
\n", "
\n", "
\n" ], "text/plain": [ " Continent Population \\\n", "Country_or_admin1 \n", "Peru South America SA \n", "Colombia South America SA \n", "Venezuela South America SA \n", "Gambia West Africa AF-W \n", "Senegal West Africa AF-W \n", "Guinea West Africa AF-W \n", "Mauritania West Africa AF-W \n", "Côte d'Ivoire West Africa AF-W \n", "Mali West Africa AF-W \n", "Burkina Faso West Africa AF-W \n", "Ghana West Africa AF-W \n", "Benin West Africa AF-W \n", "Nigeria West Africa AF-W \n", "Gabon West Africa AF-W \n", "Cameroon West Africa AF-W \n", "Democratic Republic of the Congo Central Africa AF-C \n", "Sudan Northeast Africa AF-NE \n", "Uganda Northeast Africa AF-NE \n", "Ethiopia Northeast Africa AF-NE \n", "Kenya, Kisumu Northeast Africa AF-NE \n", "Kenya, Kilifi East Africa AF-E \n", "Malawi East Africa AF-E \n", "Tanzania East Africa AF-E \n", "Mozambique East Africa AF-E \n", "Madagascar East Africa AF-E \n", "India, Odisha or West Bengal Eastern South Asia AS-S-E \n", "India, Tripura Far-eastern South Asia AS-S-FE \n", "Bangladesh Far-eastern South Asia AS-S-FE \n", "Myanmar Western Southeast Asia AS-SE-W \n", "Thailand, Tak or Ranong Western Southeast Asia AS-SE-W \n", "Thailand, Sisakhet Eastern Southeast Asia AS-SE-E \n", "Laos Eastern Southeast Asia AS-SE-E \n", "Cambodia Eastern Southeast Asia AS-SE-E \n", "Vietnam Eastern Southeast Asia AS-SE-E \n", "Indonesia Oceania OC-NG \n", "Papua New Guinea Oceania OC-NG \n", "\n", " population_colour Population_long \\\n", "Country_or_admin1 \n", "Peru #4daf4a -73.157010 \n", "Colombia #4daf4a -73.157010 \n", "Venezuela #4daf4a -73.157010 \n", "Gambia #e31a1c -3.519507 \n", "Senegal #e31a1c -3.519507 \n", "Guinea #e31a1c -3.519507 \n", "Mauritania #e31a1c -3.519507 \n", "Côte d'Ivoire #e31a1c -3.519507 \n", "Mali #e31a1c -3.519507 \n", "Burkina Faso #e31a1c -3.519507 \n", "Ghana #e31a1c -3.519507 \n", "Benin #e31a1c -3.519507 \n", "Nigeria #e31a1c -3.519507 \n", "Gabon #e31a1c -3.519507 \n", "Cameroon #e31a1c -3.519507 \n", "Democratic Republic of the Congo #fd8d3c 23.660758 \n", "Sudan #bb8129 32.741894 \n", "Uganda #bb8129 32.741894 \n", "Ethiopia #bb8129 32.741894 \n", "Kenya, Kisumu #bb8129 32.741894 \n", "Kenya, Kilifi #fecc5c 35.999707 \n", "Malawi #fecc5c 35.999707 \n", "Tanzania #fecc5c 35.999707 \n", "Mozambique #fecc5c 35.999707 \n", "Madagascar #fecc5c 35.999707 \n", "India, Odisha or West Bengal #dfc0eb 79.622525 \n", "India, Tripura #984ea3 89.833945 \n", "Bangladesh #984ea3 89.833945 \n", "Myanmar #9ecae1 98.496702 \n", "Thailand, Tak or Ranong #9ecae1 98.496702 \n", "Thailand, Sisakhet #3182bd 105.173981 \n", "Laos #3182bd 105.173981 \n", "Cambodia #3182bd 105.173981 \n", "Vietnam #3182bd 105.173981 \n", "Indonesia #f781bf 135.577208 \n", "Papua New Guinea #f781bf 135.577208 \n", "\n", " Country_or_admin1_long \\\n", "Country_or_admin1 \n", "Peru -74.356842 \n", "Colombia -73.086731 \n", "Venezuela -66.145936 \n", "Gambia -15.372910 \n", "Senegal -14.470363 \n", "Guinea -10.936960 \n", "Mauritania -10.337093 \n", "Côte d'Ivoire -5.554446 \n", "Mali -3.522152 \n", "Burkina Faso -1.745660 \n", "Ghana -1.210711 \n", "Benin 2.339713 \n", "Nigeria 8.097575 \n", "Gabon 11.784989 \n", "Cameroon 12.741504 \n", "Democratic Republic of the Congo 23.660758 \n", "Sudan 30.005646 \n", "Uganda 32.391932 \n", "Ethiopia 39.626195 \n", "Kenya, Kisumu 40.000000 \n", "Kenya, Kilifi 34.000000 \n", "Malawi 34.300482 \n", "Tanzania 34.825685 \n", "Mozambique 35.551437 \n", "Madagascar 46.698618 \n", "India, Odisha or West Bengal 84.418059 \n", "India, Tripura 90.000000 \n", "Bangladesh 90.277384 \n", "Myanmar 96.510201 \n", "Thailand, Tak or Ranong 98.791050 \n", "Thailand, Sisakhet 103.000000 \n", "Laos 103.768157 \n", "Cambodia 104.916873 \n", "Vietnam 106.551796 \n", "Indonesia 117.314980 \n", "Papua New Guinea 145.254007 \n", "\n", " Frequency (number of samples) \n", "Country_or_admin1 \n", "Peru 21 \n", "Colombia 159 \n", "Venezuela 2 \n", "Gambia 1247 \n", "Senegal 155 \n", "Guinea 199 \n", "Mauritania 104 \n", "Côte d'Ivoire 71 \n", "Mali 1804 \n", "Burkina Faso 58 \n", "Ghana 4090 \n", "Benin 334 \n", "Nigeria 140 \n", "Gabon 59 \n", "Cameroon 294 \n", "Democratic Republic of the Congo 573 \n", "Sudan 203 \n", "Uganda 15 \n", "Ethiopia 34 \n", "Kenya, Kisumu 64 \n", "Kenya, Kilifi 662 \n", "Malawi 371 \n", "Tanzania 697 \n", "Mozambique 91 \n", "Madagascar 25 \n", "India, Odisha or West Bengal 244 \n", "India, Tripura 72 \n", "Bangladesh 1658 \n", "Myanmar 1260 \n", "Thailand, Tak or Ranong 994 \n", "Thailand, Sisakhet 112 \n", "Laos 1052 \n", "Cambodia 1723 \n", "Vietnam 1733 \n", "Indonesia 133 \n", "Papua New Guinea 251 " ] }, "execution_count": 64, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_country_or_admin1 = (\n", " pd.DataFrame(\n", " sample_metadata\n", " .groupby(['Continent', 'Population', 'population_colour',\n", " 'Country_or_admin1', 'Population_long',\n", " 'Country_or_admin1_long'])\n", " .size()\n", " )\n", " .reset_index()\n", " .set_index('Country_or_admin1')\n", " .sort_values(['Population_long', 'Country_or_admin1_long'])\n", " .rename(columns={0: 'Frequency (number of samples)'})\n", ")\n", "\n", "print(df_country_or_admin1.shape)\n", "df_country_or_admin1" ] }, { "cell_type": "markdown", "metadata": { "id": "CrgcgTvbTuE_" }, "source": [ "We want to seperate and sort QC pass samples which will help to distinguish QC-fail samples in the figure." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "pMQKZlsSozvX", "outputId": "d757823e-0ffd-4886-f93c-4aabb47c5aae" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(36, 6)\n" ] }, { "data": { "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
ContinentPopulationpopulation_colourPopulation_longCountry_or_admin1_longFrequency (number of samples)
Country_or_admin1
PeruSouth AmericaSA#4daf4a-73.157010-74.35684221
ColombiaSouth AmericaSA#4daf4a-73.157010-73.086731135
VenezuelaSouth AmericaSA#4daf4a-73.157010-66.1459362
GambiaWest AfricaAF-W#e31a1c-3.519507-15.372910863
SenegalWest AfricaAF-W#e31a1c-3.519507-14.470363150
GuineaWest AfricaAF-W#e31a1c-3.519507-10.936960151
MauritaniaWest AfricaAF-W#e31a1c-3.519507-10.33709392
Côte d'IvoireWest AfricaAF-W#e31a1c-3.519507-5.55444671
MaliWest AfricaAF-W#e31a1c-3.519507-3.5221521167
Burkina FasoWest AfricaAF-W#e31a1c-3.519507-1.74566057
GhanaWest AfricaAF-W#e31a1c-3.519507-1.2107113131
BeninWest AfricaAF-W#e31a1c-3.5195072.339713150
NigeriaWest AfricaAF-W#e31a1c-3.5195078.097575110
GabonWest AfricaAF-W#e31a1c-3.51950711.78498955
CameroonWest AfricaAF-W#e31a1c-3.51950712.741504264
Democratic Republic of the CongoCentral AfricaAF-C#fd8d3c23.66075823.660758520
SudanNortheast AfricaAF-NE#bb812932.74189430.00564676
UgandaNortheast AfricaAF-NE#bb812932.74189432.39193212
EthiopiaNortheast AfricaAF-NE#bb812932.74189439.62619521
Kenya, KisumuNortheast AfricaAF-NE#bb812932.74189440.00000063
Kenya, KilifiEast AfricaAF-E#fecc5c35.99970734.000000627
MalawiEast AfricaAF-E#fecc5c35.99970734.300482265
TanzaniaEast AfricaAF-E#fecc5c35.99970734.825685589
MozambiqueEast AfricaAF-E#fecc5c35.99970735.55143734
MadagascarEast AfricaAF-E#fecc5c35.99970746.69861824
India, Odisha or West BengalEastern South AsiaAS-S-E#dfc0eb79.62252584.418059233
India, TripuraFar-eastern South AsiaAS-S-FE#984ea389.83394590.00000067
BangladeshFar-eastern South AsiaAS-S-FE#984ea389.83394590.2773841310
MyanmarWestern Southeast AsiaAS-SE-W#9ecae198.49670296.510201985
Thailand, Tak or RanongWestern Southeast AsiaAS-SE-W#9ecae198.49670298.791050895
Thailand, SisakhetEastern Southeast AsiaAS-SE-E#3182bd105.173981103.00000059
LaosEastern Southeast AsiaAS-SE-E#3182bd105.173981103.768157991
CambodiaEastern Southeast AsiaAS-SE-E#3182bd105.173981104.9168731267
VietnamEastern Southeast AsiaAS-SE-E#3182bd105.173981106.5517961404
IndonesiaOceaniaOC-NG#f781bf135.577208117.314980121
Papua New GuineaOceaniaOC-NG#f781bf135.577208145.254007221
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", " \n", "\n", "\n", "\n", " \n", "
\n", "
\n", "
\n" ], "text/plain": [ " Continent Population \\\n", "Country_or_admin1 \n", "Peru South America SA \n", "Colombia South America SA \n", "Venezuela South America SA \n", "Gambia West Africa AF-W \n", "Senegal West Africa AF-W \n", "Guinea West Africa AF-W \n", "Mauritania West Africa AF-W \n", "Côte d'Ivoire West Africa AF-W \n", "Mali West Africa AF-W \n", "Burkina Faso West Africa AF-W \n", "Ghana West Africa AF-W \n", "Benin West Africa AF-W \n", "Nigeria West Africa AF-W \n", "Gabon West Africa AF-W \n", "Cameroon West Africa AF-W \n", "Democratic Republic of the Congo Central Africa AF-C \n", "Sudan Northeast Africa AF-NE \n", "Uganda Northeast Africa AF-NE \n", "Ethiopia Northeast Africa AF-NE \n", "Kenya, Kisumu Northeast Africa AF-NE \n", "Kenya, Kilifi East Africa AF-E \n", "Malawi East Africa AF-E \n", "Tanzania East Africa AF-E \n", "Mozambique East Africa AF-E \n", "Madagascar East Africa AF-E \n", "India, Odisha or West Bengal Eastern South Asia AS-S-E \n", "India, Tripura Far-eastern South Asia AS-S-FE \n", "Bangladesh Far-eastern South Asia AS-S-FE \n", "Myanmar Western Southeast Asia AS-SE-W \n", "Thailand, Tak or Ranong Western Southeast Asia AS-SE-W \n", "Thailand, Sisakhet Eastern Southeast Asia AS-SE-E \n", "Laos Eastern Southeast Asia AS-SE-E \n", "Cambodia Eastern Southeast Asia AS-SE-E \n", "Vietnam Eastern Southeast Asia AS-SE-E \n", "Indonesia Oceania OC-NG \n", "Papua New Guinea Oceania OC-NG \n", "\n", " population_colour Population_long \\\n", "Country_or_admin1 \n", "Peru #4daf4a -73.157010 \n", "Colombia #4daf4a -73.157010 \n", "Venezuela #4daf4a -73.157010 \n", "Gambia #e31a1c -3.519507 \n", "Senegal #e31a1c -3.519507 \n", "Guinea #e31a1c -3.519507 \n", "Mauritania #e31a1c -3.519507 \n", "Côte d'Ivoire #e31a1c -3.519507 \n", "Mali #e31a1c -3.519507 \n", "Burkina Faso #e31a1c -3.519507 \n", "Ghana #e31a1c -3.519507 \n", "Benin #e31a1c -3.519507 \n", "Nigeria #e31a1c -3.519507 \n", "Gabon #e31a1c -3.519507 \n", "Cameroon #e31a1c -3.519507 \n", "Democratic Republic of the Congo #fd8d3c 23.660758 \n", "Sudan #bb8129 32.741894 \n", "Uganda #bb8129 32.741894 \n", "Ethiopia #bb8129 32.741894 \n", "Kenya, Kisumu #bb8129 32.741894 \n", "Kenya, Kilifi #fecc5c 35.999707 \n", "Malawi #fecc5c 35.999707 \n", "Tanzania #fecc5c 35.999707 \n", "Mozambique #fecc5c 35.999707 \n", "Madagascar #fecc5c 35.999707 \n", "India, Odisha or West Bengal #dfc0eb 79.622525 \n", "India, Tripura #984ea3 89.833945 \n", "Bangladesh #984ea3 89.833945 \n", "Myanmar #9ecae1 98.496702 \n", "Thailand, Tak or Ranong #9ecae1 98.496702 \n", "Thailand, Sisakhet #3182bd 105.173981 \n", "Laos #3182bd 105.173981 \n", "Cambodia #3182bd 105.173981 \n", "Vietnam #3182bd 105.173981 \n", "Indonesia #f781bf 135.577208 \n", "Papua New Guinea #f781bf 135.577208 \n", "\n", " Country_or_admin1_long \\\n", "Country_or_admin1 \n", "Peru -74.356842 \n", "Colombia -73.086731 \n", "Venezuela -66.145936 \n", "Gambia -15.372910 \n", "Senegal -14.470363 \n", "Guinea -10.936960 \n", "Mauritania -10.337093 \n", "Côte d'Ivoire -5.554446 \n", "Mali -3.522152 \n", "Burkina Faso -1.745660 \n", "Ghana -1.210711 \n", "Benin 2.339713 \n", "Nigeria 8.097575 \n", "Gabon 11.784989 \n", "Cameroon 12.741504 \n", "Democratic Republic of the Congo 23.660758 \n", "Sudan 30.005646 \n", "Uganda 32.391932 \n", "Ethiopia 39.626195 \n", "Kenya, Kisumu 40.000000 \n", "Kenya, Kilifi 34.000000 \n", "Malawi 34.300482 \n", "Tanzania 34.825685 \n", "Mozambique 35.551437 \n", "Madagascar 46.698618 \n", "India, Odisha or West Bengal 84.418059 \n", "India, Tripura 90.000000 \n", "Bangladesh 90.277384 \n", "Myanmar 96.510201 \n", "Thailand, Tak or Ranong 98.791050 \n", "Thailand, Sisakhet 103.000000 \n", "Laos 103.768157 \n", "Cambodia 104.916873 \n", "Vietnam 106.551796 \n", "Indonesia 117.314980 \n", "Papua New Guinea 145.254007 \n", "\n", " Frequency (number of samples) \n", "Country_or_admin1 \n", "Peru 21 \n", "Colombia 135 \n", "Venezuela 2 \n", "Gambia 863 \n", "Senegal 150 \n", "Guinea 151 \n", "Mauritania 92 \n", "Côte d'Ivoire 71 \n", "Mali 1167 \n", "Burkina Faso 57 \n", "Ghana 3131 \n", "Benin 150 \n", "Nigeria 110 \n", "Gabon 55 \n", "Cameroon 264 \n", "Democratic Republic of the Congo 520 \n", "Sudan 76 \n", "Uganda 12 \n", "Ethiopia 21 \n", "Kenya, Kisumu 63 \n", "Kenya, Kilifi 627 \n", "Malawi 265 \n", "Tanzania 589 \n", "Mozambique 34 \n", "Madagascar 24 \n", "India, Odisha or West Bengal 233 \n", "India, Tripura 67 \n", "Bangladesh 1310 \n", "Myanmar 985 \n", "Thailand, Tak or Ranong 895 \n", "Thailand, Sisakhet 59 \n", "Laos 991 \n", "Cambodia 1267 \n", "Vietnam 1404 \n", "Indonesia 121 \n", "Papua New Guinea 221 " ] }, "execution_count": 65, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_country_or_admin1_pass = (\n", " pd.DataFrame(\n", " sample_metadata\n", " .loc[sample_metadata['QC pass']]\n", " .groupby(['Continent', 'Population',\n", " 'population_colour', 'Country_or_admin1',\n", " 'Population_long', 'Country_or_admin1_long'])\n", " .size()\n", " )\n", " .reset_index()\n", " .set_index('Country_or_admin1')\n", " .sort_values(['Population_long', 'Country_or_admin1_long'])\n", " .rename(columns={0: 'Frequency (number of samples)'})\n", ")\n", "print(df_country_or_admin1_pass.shape)\n", "df_country_or_admin1_pass" ] }, { "cell_type": "markdown", "metadata": { "id": "UqZWlN5pStvT" }, "source": [ "Finally, we rename some countries with long names to shorter names to prevent the restriction of figure size." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "SSFOLh4cSmk-" }, "outputs": [], "source": [ "# rename the long-name countries in total samples df\n", "df_country_or_admin1.rename(index={'Democratic Republic of the Congo': 'DRC'},inplace=True)\n", "df_country_or_admin1.rename(index={'India, Odisha or West Bengal': 'India, Odisha\\nor West Bengal'},inplace=True)\n", "df_country_or_admin1.rename(index={'Thailand, Tak or Ranong': 'Thailand, Tak\\nor Ranong'},inplace=True)\n", "\n", "# rename the long-name countries in QC-pass samples df\n", "df_country_or_admin1_pass.rename(index={'Democratic Republic of the Congo': 'DRC'},inplace=True)\n", "df_country_or_admin1_pass.rename(index={'India, Odisha or West Bengal': 'India, Odisha\\nor West Bengal'},inplace=True)\n", "df_country_or_admin1_pass.rename(index={'Thailand, Tak or Ranong': 'Thailand, Tak\\nor Ranong'},inplace=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "DsfW-53dtHWg" }, "source": [ "## Make the figure" ] }, { "cell_type": "markdown", "metadata": { "id": "sUSSS1hrAgXP" }, "source": [ "We have the following considerations when making of this figure:\n", "1. While QC failed samples shown as outline only, others should have a solid-background to distinguish from each other\n", "2. Lines and annotations at bottom for both continent and population.\n", "3. The y-axis is truncated at 2,000 samples for visual clarity. With over 3,000 samples, Ghana is affected by this truncation. Therefore, specific annotations for QC pass and fail are positioned above Ghana's bar to highlight its significance." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 454 }, "id": "R8VW-P63T8Hg", "outputId": "6befef17-90dc-41ce-9b03-b662b2edbb49" }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Adjust the figure size\n", "fig, ax = plt.subplots(1, 1, figsize=(23.3, 10))\n", "\n", "# Create the bars for all samples with a white backgound\n", "ax.bar(\n", " np.arange(len(df_country_or_admin1)),\n", " df_country_or_admin1['Frequency (number of samples)'],\n", " facecolor='white',\n", " edgecolor = df_country_or_admin1['population_colour'],\n", ")\n", "# Create the bars for QC pass with solid-colour background\n", "ax.bar(\n", " np.arange(len(df_country_or_admin1_pass)),\n", " df_country_or_admin1_pass['Frequency (number of samples)'],\n", " color = df_country_or_admin1['population_colour'],\n", " edgecolor = df_country_or_admin1['population_colour'],\n", ")\n", "# Set x-axis labels and rotate them for readability\n", "ax.set_xticks(np.arange(len(df_country_or_admin1)))\n", "ax.set_xticklabels(df_country_or_admin1.index, rotation=90)\n", "ax.grid(True, axis='y')\n", "# Set the y-axis limit to truncate bars at a maximum of 2000\n", "ax.set_ylim(0, 2000)\n", "# Set axis labels\n", "ax.set_xlabel('Country or region',fontsize=15)\n", "ax.set_ylabel('Frequency (number of samples)',fontsize=15)\n", "trans = ax.get_xaxis_transform()\n", "# Add specific annotation to Ghana\n", "total_samples = collections.OrderedDict()\n", "pass_samples = collections.OrderedDict()\n", "x_pos = collections.OrderedDict()\n", "# Set the index number for Ghana\n", "x_pos['Ghana'] = 10\n", "for country in x_pos:\n", " total_samples[country] = df_country_or_admin1.loc[country, 'Frequency (number of samples)']\n", " pass_samples[country] = df_country_or_admin1_pass.loc[country, 'Frequency (number of samples)']\n", " ax.annotate(f\"{pass_samples[country]:,}/\", xy=(x_pos[country], 1.1), xycoords=trans, ha=\"center\", va=\"top\")\n", " ax.annotate(f\"{total_samples[country] - pass_samples[country]:,}\", xy=(x_pos[country], 1.05), xycoords=trans, ha=\"center\", va=\"top\")\n", "y_offset = -0.6\n", "text_offset = 0.05\n", "x_offset = 0.3\n", "\n", "# Add annotations for Continents\n", "ax.annotate('Continent', xy=(-3, y_offset-text_offset), xycoords=trans, ha=\"left\", va=\"top\",fontsize=14)\n", "ax.annotate('South\\nAmerica', xy=(1, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([0-x_offset, 2+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('Africa', xy=(13.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([3-x_offset, 24+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('Asia', xy=(29, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([25-x_offset, 33+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('Oceania', xy=(34.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([34-x_offset, 35+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "y_offset = -0.45\n", "text_offset = 0.05\n", "x_offset = 0.3\n", "\n", "# Add annotations for Populations\n", "ax.annotate('Population', xy=(-3, y_offset-text_offset), xycoords=trans, ha=\"left\", va=\"top\",fontsize=14)\n", "ax.annotate('SA', xy=(1, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([0-x_offset, 2+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AF-W', xy=(8.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([3-x_offset, 14+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AF-C', xy=(15, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([15-x_offset, 15+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AF-NE', xy=(17.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([16-x_offset, 19+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AF-E', xy=(22, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([20-x_offset, 24+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AS-S-E', xy=(25, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([25-x_offset, 25+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AS-S-FE', xy=(26.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([26-x_offset, 27+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AS-SE-W', xy=(28.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([28-x_offset, 29+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('AS-SE-E', xy=(31.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "ax.plot([30-x_offset, 33+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "ax.annotate('OC-NG', xy=(34.5, y_offset-text_offset), xycoords=trans, ha=\"center\", va=\"top\",fontsize=12)\n", "_ = ax.plot([34-x_offset, 35+x_offset],[y_offset, y_offset], color=\"k\", transform=trans, clip_on=False)\n", "\n", "# Customize tick label fonts and spacing\n", "for tick in ax.xaxis.get_major_ticks():\n", " tick.label1.set_fontsize(9)\n", "ax.tick_params(axis='x', pad=-2)\n", "\n", "fig.tight_layout()" ] }, { "cell_type": "markdown", "metadata": { "id": "TX7cb_DG-jek" }, "source": [ "**Figure Legend. Breakdown of samples by country.** Solid bars indicate samples which passed QC. Unfilled bars represent samples that failed QC. The y-axis is truncated at 2,000 samples, with the numbers of QC pass/QC fail samples in Ghana shown above the bar. Bars are coloured according to the major sub-population to which the location is assigned." ] }, { "cell_type": "markdown", "metadata": { "id": "6H4B534mtOmQ" }, "source": [ "### Save the figure" ] }, { "cell_type": "markdown", "metadata": { "id": "BEn5aMlx8wWV" }, "source": [ "We can output this to a location in Google Drive\n", "\n", "First we need to connect Google Drive by running the following:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "fsTmWqOlUcdj", "outputId": "80944b33-cd00-4748-c921-af339474c247" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mounted at /content/drive\n" ] } ], "source": [ "# You will need to authorise Google Colab access to Google Drive\n", "drive.mount('/content/drive')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "5Xc5KzXF80N9" }, "outputs": [], "source": [ "# This will send the file to your Google Drive, where you can download it from if needed\n", "# Change the file path if you wish to send the file to a specific location\n", "# Change the file name if you wish to call it something else\n", "\n", "fig.savefig('/content/drive/My Drive/SamplesByCountry_Barplot_20210720.pdf')\n", "fig.savefig('/content/drive/My Drive/SamplesByCountry_Barplot_20210720.png', dpi=480) # increase the dpi for higher resolution" ] } ], "metadata": { "colab": { "provenance": [] }, "kernelspec": { "display_name": "Python 3", "name": "python3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 0 }