# install the latest version of the appliedepidata package
pak::pak("appliedepi/appliedepidata")2 Download handbook and data
2.1 Download offline handbook
You can download this handbook and read it without an internet connection. Each language is a zip file holding the whole site: every chapter, every image and the navigation sidebar.
- Download the zip for your language, then unzip it.
- Open
index.htmlin the unzipped folder to start reading. - See the Suggested packages page to install the R packages you need before you lose your connection.
To download the handbook:
The zip is rebuilt automatically, so it always follows the latest version of the handbook.
2.2 Download data to follow along
To “follow along” with the handbook pages, you can load the example data straight into R with our companion R package, appliedepidata. It contains every example dataset used in this handbook.
Install the package
appliedepidata is not on CRAN, so install it from its Github repository with pak:
Load a dataset directly
Use get_data() with the dataset’s name = to load it straight into R - no download, no file path, no import() step. The function returns the dataset as an R object (a data frame, an sf object, a phylo tree, etc., depending on the dataset).
# load the cleaned case linelist directly into R
linelist <- appliedepidata::get_data(name = "linelist_cleaned_rds")Each page of this handbook that uses example data shows the get_data(name = "...") call needed for that page, right where the data are first used.
Save a dataset to a real file
Some pages teach you how to import a file (e.g. an Excel workbook, or a folder of files) rather than an R object. For these, use save_data() to write the real file(s) to a folder of your choice with name = and path =:
# write the raw linelist Excel file to your working directory
appliedepidata::save_data(name = "case_linelists_linelist_raw", path = getwd())You can then import() it as you would any file on your computer - see the Import and export page for details. Note that save_data() does not automatically unzip datasets that are distributed as a .zip bundle (e.g. multi-file shapefiles, or folders of files); use utils::unzip() on the saved .zip afterwards.
Browse what’s available
-
appliedepidata::search_data()launches an interactive Shiny app so you can filter and search all the datasets in the package by language and keyword. -
appliedepidata::list_data()returns a data frame listing every dataset name programmatically, for use in scripts.
# browse datasets interactively
appliedepidata::search_data()
# list all dataset names programmatically
appliedepidata::list_data()If you wish, you can also review the raw data files in the “data” folder of the appliedepidata Github repository.
Dataset reference, by page
Below is the get_data()/save_data() name to use for the example data referenced on each page of this handbook.
Case linelist
This is a fictional Ebola outbreak, expanded by the handbook team from the ebola_sim practice dataset in the outbreaks package.
# the "raw" linelist - an Excel spreadsheet with messy data
# use this to follow along with the Cleaning data and core functions page
raw_linelist <- appliedepidata::get_data(name = "case_linelists_linelist_raw")
# the "clean" linelist - an R-specific .rds file that preserves column classes
# use this for all other pages of this handbook that use the linelist
linelist <- appliedepidata::get_data(name = "linelist_cleaned_rds")
# the "clean" linelist, as an Excel file instead
linelist_excel <- appliedepidata::get_data(name = "linelist_cleaned_excel")Part of the cleaning page uses a “cleaning dictionary” (.csv file). You can load it directly into R by running the following command:
cleaning_dict <- appliedepidata::get_data(name = "case_linelists_cleaning_dict")Working with dates
The Working with dates page uses counts of cases by district and date:
counts <- appliedepidata::get_data(name = "example_district_weekly_count_data")Writing functions
The Writing functions page uses a linelist of the 2013 H7N9 influenza outbreak in China:
flu_china <- appliedepidata::get_data(name = "fluH7N9_China_2013")Iteration, loops, and lists
The Iteration, loops, and lists page imports an Excel workbook with one sheet per hospital. Save it as a file first:
appliedepidata::save_data(name = "example_hospital_linelists", path = getwd())Import and export
The Import and export page imports the linelist from Excel files. Save them as files first:
# the raw linelist
appliedepidata::save_data(name = "linelist_raw", path = getwd())
# the clean linelist
appliedepidata::save_data(name = "linelist_cleaned", path = getwd())Malaria count data
These data are fictional counts of malaria cases by age group, facility, and day. A .rds file is an R-specific file type that preserves column classes. This ensures you will have only minimal cleaning to do after importing the data into R.
malaria_data <- appliedepidata::get_data(name = "malaria_facility_count_data")Likert-scale data
These are fictional data from a Likert-style survey, used in the page on Demographic pyramids and Likert-scales.
likert_data <- appliedepidata::get_data(name = "likert_data")Flexdashboard
Below are links to the files associated with the page on Dashboards with R Markdown. These are R Markdown/HTML source files, not example datasets, so they are not part of the appliedepidata package - download them directly from Github:
Contact Tracing
The Contact Tracing page analyses example contact tracing data from Go.Data. These data are part of appliedepidata:
# case investigation data
cases <- appliedepidata::get_data(name = "cases_clean")
# contact registration data
contacts <- appliedepidata::get_data(name = "contacts_clean")
# contact follow-up data
followups <- appliedepidata::get_data(name = "followups_clean")
# links between cases and contacts
relationships <- appliedepidata::get_data(name = "relationships_clean")NOTE: Structured contact tracing data from other software (e.g. KoBo, DHIS2 Tracker, CommCare) may look different. If you would like to contribute alternative sample data or content for this page, please contact us.
TIP: If you are deploying Go.Data and want to connect to your instance’s API, see the Import and export page (API section) and the Go.Data Community of Practice.
GIS
Shapefiles have many sub-component files, each with a different file extension. One file will have the “.shp” extension, but others may have “.dbf”, “.prj”, etc. appliedepidata bundles each shapefile’s component files together as a single .zip, so save_data() and then unzip:
# load the Sierra Leone admin-3 shapefile directly as an sf object
sle_adm3 <- appliedepidata::get_data(name = "sle_adm3")
# load the health facility points as an sf object
sle_hf <- appliedepidata::get_data(name = "sle_hf")
# load the ADM3 population table
sle_adm3_pop <- appliedepidata::get_data(name = "sle_admpop_adm3_2020")
# OR save the raw shapefile component files to a folder
shp_dir <- "sle_adm3_files"
dir.create(shp_dir)
appliedepidata::save_data(name = "sle_adm3", path = shp_dir)
utils::unzip(file.path(shp_dir, "sle_adm3.zip"), exdir = shp_dir)The GIS basics page loads all three data sets: the admin-3 boundaries sle_adm3, the health facility points sle_hf, and the population table sle_admpop_adm3_2020. To read a shapefile from your own file path, use read_sf() from the sf package. The GIS basics page also links to the Humanitarian Data Exchange website, where you can download further shapefiles as zipped files. The health facility points, for example, are here.
Phylogenetic trees
See the page on Phylogenetic trees. Newick file of phylogenetic tree constructed from whole genome sequencing of 299 Shigella sonnei samples and corresponding sample data (converted to a text file). The Belgian samples and resulting data are kindly provided by the Belgian NRC for Salmonella and Shigella in the scope of a project conducted by an ECDC EUPHEM Fellow, and will also be published in a manuscript. The international data are openly available on public databases (ncbi) and have been previously published.
# the phylogenetic tree, as a "phylo" object (ape package)
tree <- appliedepidata::get_data(name = "Shigella_tree")
# additional information on each sample
sample_data <- appliedepidata::get_data(name = "sample_data_Shigella_tree")
# the subset-tree created later in the page
subtree <- appliedepidata::get_data(name = "Shigella_subtree_2")Standardization
See the page on Standardised rates. You can load the data directly with the following commands:
##############
# Country A
##############
# demographics for country A
A_demo <- appliedepidata::get_data(name = "country_demographics")
# deaths for country A
A_deaths <- appliedepidata::get_data(name = "deaths_countryA")
##############
# Country B
##############
# demographics for country B
B_demo <- appliedepidata::get_data(name = "country_demographics_2")
# deaths for country B
B_deaths <- appliedepidata::get_data(name = "deaths_countryB")
###############
# Reference Pop
###############
# world standard population, by sex
standard_pop_data <- appliedepidata::get_data(name = "world_standard_population_by_sex")Time series and outbreak detection
See the page on Time series and outbreak detection. We use campylobacter cases reported in Germany 2002-2011, as available from the surveillance R package. (nb. this dataset has been adapted from the original, in that 3 months of data have been deleted from the end of 2011 for demonstration purposes)
counts <- appliedepidata::get_data(name = "campylobacter_germany")We also use climate data from Germany 2002-2011 (temperature in degrees celsius and rain fall in millimetres). These were downloaded from the EU Copernicus satellite reanalysis dataset using the ecmwfr package, and are bundled by appliedepidata as one .zip of ten yearly .nc files. get_data() reads the whole bundle in as one combined stars object directly; save_data() gives you the raw .nc files if you want to inspect or process them yourself.
# read the combined weather data directly as a stars object
weather_data <- appliedepidata::get_data(name = "germany_weather")
# OR save the raw yearly .nc files to a folder
weather_dir <- "germany_weather_files"
dir.create(weather_dir)
appliedepidata::save_data(name = "germany_weather", path = weather_dir)
utils::unzip(file.path(weather_dir, "germany_weather.zip"), exdir = weather_dir)Survey analysis
For the survey analysis page we use fictional mortality survey data based off MSF OCA survey templates. This fictional data was generated as part of the “R4Epis” project.
# fictional survey data
survey_data <- appliedepidata::get_data(name = "survey_data")
# fictional survey data dictionary
survey_dict <- appliedepidata::get_data(name = "survey_dict")
# fictional survey population data
population <- appliedepidata::get_data(name = "population")Shiny
The page on Dashboards with Shiny demonstrates the construction of a simple app to display malaria data. The app’s data are part of appliedepidata:
malaria_data <- appliedepidata::get_data(name = "malaria_facility_count_data")The app’s R scripts are not example data, so they are not part of appliedepidata - download them directly from Github:
You can click here to download the app.R file that contains both the UI and Server code for the Shiny app.
You can click here to download the global.R file that should run prior to the app opening, as explained in the page.
You can click here to download the plot_epicurve.R file that is sourced by global.R. Note that you may need to store it within a “funcs” folder for the here() file paths to work correctly.