Data Visualization Using Mosquito and Malaria Data
Allan
2025-05-05
Introduction to data visualization in R
0. Getting started ——–
Libraries
library(tidyverse)TRUE ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
TRUE ✔ dplyr 1.1.4 ✔ readr 2.1.5
TRUE ✔ forcats 1.0.0 ✔ stringr 1.5.1
TRUE ✔ ggplot2 3.5.2 ✔ tibble 3.2.1
TRUE ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
TRUE ✔ purrr 1.0.2
TRUE ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
TRUE ✖ dplyr::filter() masks stats::filter()
TRUE ✖ dplyr::lag() masks stats::lag()
TRUE ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorsCreate a data folder
dir.create("data")Download example data
download.file("https://raw.githubusercontent.com/AMMnet/AMMnet-Hackathon/main/01_data-vis/data/mockdata_cases.csv", destfile = "data/mockdata_cases.csv")
download.file("https://raw.githubusercontent.com/AMMnet/AMMnet-Hackathon/main/01_data-vis/data/mosq_mock.csv", destfile = "data/mosq_mock.csv")Load example data
malaria_data <- read_csv("data/mockdata_cases.csv")TRUE Rows: 514 Columns: 10
TRUE ── Column specification ────────────────────────────────────────────────────────
TRUE Delimiter: ","
TRUE chr (2): location, ages
TRUE dbl (8): month, year, total, positive, xcoord, ycoord, prev, time_order_loc
TRUE
TRUE ℹ Use `spec()` to retrieve the full column specification for this data.
TRUE ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.mosquito_data <- read_csv("data/mosq_mock.csv")TRUE Rows: 104 Columns: 19
TRUE ── Column specification ────────────────────────────────────────────────────────
TRUE Delimiter: ","
TRUE chr (4): Village, Method, Location, hour
TRUE dbl (15): session, Compound.ID, ag.Male, Ag.unfed, Ag.halffed, Ag.fed, Ag.gr...
TRUE
TRUE ℹ Use `spec()` to retrieve the full column specification for this data.
TRUE ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.1. Characterizing our data ——–
#’ Before we start visualizing our data, we need to understand the #’ characteristics of our data. The goal is to get an idea of the #’ data structure and to understand the relationships between variables.
Explore the structure and summary of the datasets
dim(malaria_data) TRUE [1] 514 10head(malaria_data)summary(malaria_data)TRUE location month year ages
TRUE Length:514 Min. : 1.000 Min. :2018 Length:514
TRUE Class :character 1st Qu.: 4.000 1st Qu.:2018 Class :character
TRUE Mode :character Median : 7.000 Median :2019 Mode :character
TRUE Mean : 6.486 Mean :2019
TRUE 3rd Qu.: 9.000 3rd Qu.:2020
TRUE Max. :12.000 Max. :2020
TRUE total positive xcoord ycoord
TRUE Min. : 20.0 Min. : -1.00 Min. :-21.84 Min. :28.52
TRUE 1st Qu.: 46.0 1st Qu.: 14.00 1st Qu.:-20.39 1st Qu.:29.64
TRUE Median :103.0 Median : 33.00 Median :-20.06 Median :29.99
TRUE Mean :141.5 Mean : 47.81 Mean :-20.04 Mean :30.00
TRUE 3rd Qu.:206.0 3rd Qu.: 67.00 3rd Qu.:-19.71 3rd Qu.:30.32
TRUE Max. :611.0 Max. :264.00 Max. :-18.79 Max. :31.81
TRUE prev time_order_loc
TRUE Min. :-0.04545 Min. : 1.00
TRUE 1st Qu.: 0.24615 1st Qu.: 9.00
TRUE Median : 0.33016 Median :18.00
TRUE Mean : 0.31518 Mean :17.65
TRUE 3rd Qu.: 0.39024 3rd Qu.:26.00
TRUE Max. : 0.53488 Max. :35.00Explore individual columns/variables
malaria_data$location # values for a single columnTRUE [1] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [6] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [11] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [16] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [21] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [26] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [31] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [36] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [41] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [46] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [51] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [56] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [61] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [66] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [71] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [76] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [81] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [86] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [91] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [96] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [101] "mordor" "mordor" "mordor" "mordor" "mordor"
TRUE [106] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [111] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [116] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [121] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [126] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [131] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [136] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [141] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [146] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [151] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [156] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [161] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [166] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [171] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [176] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [181] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [186] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [191] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [196] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [201] "narnia" "narnia" "narnia" "narnia" "narnia"
TRUE [206] "narnia" "narnia" "narnia" "narnia" "neverwhere"
TRUE [211] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [216] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [221] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [226] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [231] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [236] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [241] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [246] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [251] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [256] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [261] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [266] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [271] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [276] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [281] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [286] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [291] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [296] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [301] "neverwhere" "neverwhere" "neverwhere" "neverwhere" "neverwhere"
TRUE [306] "oz" "oz" "oz" "oz" "oz"
TRUE [311] "oz" "oz" "oz" "oz" "oz"
TRUE [316] "oz" "oz" "oz" "oz" "oz"
TRUE [321] "oz" "oz" "oz" "oz" "oz"
TRUE [326] "oz" "oz" "oz" "oz" "oz"
TRUE [331] "oz" "oz" "oz" "oz" "oz"
TRUE [336] "oz" "oz" "oz" "oz" "oz"
TRUE [341] "oz" "oz" "oz" "oz" "oz"
TRUE [346] "oz" "oz" "oz" "oz" "oz"
TRUE [351] "oz" "oz" "oz" "oz" "oz"
TRUE [356] "oz" "oz" "oz" "oz" "oz"
TRUE [361] "oz" "oz" "oz" "oz" "oz"
TRUE [366] "oz" "oz" "oz" "oz" "oz"
TRUE [371] "oz" "oz" "oz" "oz" "oz"
TRUE [376] "oz" "oz" "oz" "oz" "oz"
TRUE [381] "oz" "oz" "oz" "oz" "oz"
TRUE [386] "oz" "oz" "oz" "oz" "oz"
TRUE [391] "oz" "oz" "oz" "oz" "oz"
TRUE [396] "oz" "oz" "oz" "oz" "oz"
TRUE [401] "oz" "oz" "oz" "oz" "oz"
TRUE [406] "oz" "oz" "oz" "oz" "wonderland"
TRUE [411] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [416] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [421] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [426] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [431] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [436] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [441] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [446] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [451] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [456] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [461] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [466] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [471] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [476] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [481] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [486] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [491] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [496] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [501] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [506] "wonderland" "wonderland" "wonderland" "wonderland" "wonderland"
TRUE [511] "wonderland" "wonderland" "wonderland" "wonderland"unique(malaria_data$location) # unique values for a single columnTRUE [1] "mordor" "narnia" "neverwhere" "oz" "wonderland"table(malaria_data$location) # frequencies for a single columnTRUE
TRUE mordor narnia neverwhere oz wonderland
TRUE 105 104 96 104 105table(malaria_data$location, malaria_data$ages) # frequencies for multiple columnsTRUE
TRUE 15_above 5_to_14 under_5
TRUE mordor 35 35 35
TRUE narnia 35 35 34
TRUE neverwhere 32 32 32
TRUE oz 35 35 34
TRUE wonderland 35 35 35Check for missing values in the data and in each column
sum(is.na(malaria_data))TRUE [1] 0colSums(is.na(malaria_data)) TRUE location month year ages total
TRUE 0 0 0 0 0
TRUE positive xcoord ycoord prev time_order_loc
TRUE 0 0 0 0 02. Exploratory Visualizations Using Base R Functions ————
#’ First, we will look at some exploratory data visualization #’ techniques using base R functions. The purpose of these plots #’ is to help us understand the relationships between variables and #’ characteristics of our data. They are useful for quickly exploring #’ the data and understanding the relationships, but they are not #’ are not great for sharing in scientific publications/presentations.
One variable comparison
Histograms
hist(malaria_data$prev)
hist(malaria_data$prev,
breaks = 10,
main = "Distribution of Malaria Prevalence",
xlab = "Malaria Prevalence",
ylab = "Frequency",
col = "purple",
border = "black")
Barplot
barplot(table(malaria_data$ages))
table(malaria_data$ages)TRUE
TRUE 15_above 5_to_14 under_5
TRUE 172 172 170barplot(table(malaria_data$location))
barplot(table(malaria_data$year))
Multiple variables
Scatterplot
plot(malaria_data$total, malaria_data$positive)
plot(malaria_data$month, malaria_data$prev)
plot_jan<-filter(malaria_data, month==1) #scatter plot for a single month of january
plot(plot_jan$month, plot_jan$prev)
Boxplots
boxplot(malaria_data$prev ~ malaria_data$month) 
boxplot(malaria_data$prev ~ malaria_data$location)
3. Data Visualization with ggplot2 ————
#’ ggplot2 is a popular visualization package for R. It provides #’ an easy-to-use interface for creating data visualizations. #’ The ggplot2 package is based on the “grammar of graphics” #’ and is a powerful way to create complex visualizations that #’ are useful for creating scientific and publication-quality #’ figures. #’ #’ The “grammar of graphics” used in ggplot2 is a set of rules that are #’ used to develop data visualizations using a layering approach. Layers #’ are added using the “+” operator. #’ # Components of a ggplot #’ There are three main components of a ggplot: #’ 1. The data: the dataset we want to visualize #’ 2. The aesthetics: the visual properties from the data used in the plot #’ 3. The geometries: the visual representations of the data (e.g., points, lines, bars)
The data
#’ All ggplot2 plots require a data frame as input. Just running this #’ line will produce a blank plot because we have stated which elements from #’ the data we want to visualize or how we want to visualize them.
ggplot(data = malaria_data) 
The aesthetics
#’ Next, we need to specify the visual properties of the plot that are determined #’ by the data. The aesthetics are specified using the aes() function. The output should #’ now produce a blank plot but with determined visual properties (e.g., axes labels).
ggplot(data = malaria_data, aes(x = total, y = positive))
The geometries
#’ Finally, we need to specify the visual representation of the data. The #’ geometries are specified using the geom_ function. There are many #’ different geometries that can be used in ggplot2. We will use geom_point #’ in this example and we will append it to the previous plot using the #’ “+” operator. The output should now produce a plot with the specified visual #’ representation of the data.
ggplot(data = malaria_data, aes(x = total, y = positive)) +
geom_point()
Here are some examples of different geom functions
ggplot(data = malaria_data, aes(x = prev)) +
geom_histogram(bins = 20) # the "bins" argument specifies the number of bars
ggplot(data = malaria_data, aes(x = year)) +
geom_bar(fill = "tomato") # the "fill" argument specifies the color of the bars
ggplot(data = malaria_data, aes(x = location, y = prev)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) # geom_jitter adds jittered points to the plot, and the "alpha" argument specifies the transparency
ggplot(data = malaria_data, aes(x = location, y = prev)) +
geom_violin() + # Violin plot are similar to boxplots, but illustrate the distribution of the data
geom_jitter(alpha = 0.2)
ggplot(data = malaria_data, aes(x = total, y = positive)) +
geom_point() +
geom_smooth(method = "lm") # The smooth geom add a smoothed line to the plot, using the "lm" or other methodsTRUE `geom_smooth()` using formula = 'y ~ x'
#’ Expanding the aes() function #’ Addition visual properties, such as color, size, and shape, can be defined #’ from our input data using the aes() function. Here is an example of adding #’ color to a previous plot using the color aesthetic.
ggplot(data = malaria_data, aes(x = total, y = positive, color = location)) +
geom_point()
#’ Note that this is different then defining a color directly within the geom_point, #’ which would only apply a single color to all points.
ggplot(data = malaria_data, aes(x = total, y = positive)) +
geom_point(color = "tomato")
#’ When using the aes() function, the visual properties will be determined by a #’ variable in the dataset. This allows us to visualize relationships between #’ multiple variables at the same time.
ggplot(data = malaria_data, aes(x = prev, fill = ages)) +
geom_histogram(color = "black")TRUE `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ggplot(data = malaria_data, aes(x = prev, fill = ages)) +
geom_histogram(color = "black", bins = 12)
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2)
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2,aes(color=location))
ggplot(data = malaria_data, aes(x = total, y = positive, color = location), alpha = 0.5) +
geom_point() +
geom_smooth(method = "lm", se = FALSE)TRUE `geom_smooth()` using formula = 'y ~ x'
ggplot(data = malaria_data, aes(x = xcoord, y = ycoord, color = location)) +
geom_point(alpha = 0.5)
4. Customizing ggplot Graphics for Presentation and Communication ——–
#’ In this section, we will using additional features of ggplot2 to customize and #’ develop high-quality plots that can used in scientific publications and presentations.
Themes
#’ There are many different themes that can be used in ggplot2. #’ The “theme” function is used to specify the theme of the plot. There are many #’ preset theme functions, and further custom themes can be created using the #’ generic theme() function. #’ Typically you will want to set the theme at the end of your plot.
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
theme_classic()
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
theme_bw()
ggplot(data = malaria_data, aes(x = location, y = prev, fill = ages)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
theme_classic() +
theme(legend.position = "bottom")
Labels
#’ Labels can be added to plots using the labs() function.
ggplot(data = malaria_data, aes(x = location, y = prev, fill = ages)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
labs(title = "Malaria prevalence by location and age group",
subtitle = "Data from 2018 - 2020",
x = "Location",
y = "Prevalence",
fill = "Age group") +
theme_classic() +
theme(legend.position = "bottom")
Custom color palettes
#’ There are many different color palettes that can be used in ggplot2. #’ The “scale_color” function is used to specify the color of the plot. There are many #’ preset color palettes, and further custom color palettes can be created using the #’ generic scale_color() function.
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
scale_fill_brewer(palette = "Set1")
We can also set our own colors
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
scale_fill_manual(values = c("#C6E0FF", "#136F63", "#E0CA3C", "#F34213", "#3E2F5B"))
We can also use custom color palettes for continuous variables
ggplot(data = malaria_data, aes(x = total, y = positive, color = prev)) +
geom_point() +
scale_color_gradient(low = "blue", high = "red")
ggplot(data = malaria_data, aes(x = total, y = positive, color = prev)) +
geom_point() +
scale_color_viridis_c(option = "magma") # use viridis package to create custom color palettes
Facets
#’ Facets are a powerful feature of ggplot2 that allow us to create multiple plots #’ based on a single variable. This “small multiple” approach is another effective #’ way to visualize relationships between mutliple variables.
ggplot(data = malaria_data, aes(x = total, y = positive, color = prev)) +
geom_point() +
scale_color_viridis_c(option = "magma") +
facet_wrap(~ location)
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
facet_wrap(~ ages) +
coord_flip() + # flips the x and y axes
scale_fill_manual(values = c("#C6E0FF", "#136F63", "#E0CA3C", "#F34213", "#3E2F5B")) +
labs(title = "Malaria prevalence by location and age group",
subtitle = "Data from 2018 - 2020",
x = "Location",
y = "Prevalence",
fill = "Age group") +
theme_classic()
ggplot(data = malaria_data, aes(x = prev, fill = ages)) +
geom_histogram(bins = 10) +
scale_fill_viridis_d() +
facet_grid(year ~ .)
Exporting plots
#’ ggplot2 can be exported to a variety of formats using the ggsave() function. #’ You can specify which plot to export by saving in an object and then calling the #’ object in the ggsave() function, otherwise ggsave() will save the current/last plot. #’ The width and height of the output image using the width and height can be set using #’ the width and height arguments, and the resolution of the image using the dpi argument. #’ #’ The file type can be set using the format argument, or by using a specific file extension. #’ I recommend using informative names for the output file.
ggplot(data = malaria_data, aes(x = location, y = prev, fill = location)) +
geom_boxplot() +
geom_jitter(alpha = 0.2) +
facet_wrap(~ ages) +
coord_flip() + # flips the x and y axes
scale_fill_manual(values = c("#C6E0FF", "#136F63", "#E0CA3C", "#F34213", "#3E2F5B")) +
labs(title = "Malaria prevalence by location and age group",
subtitle = "Data from 2018 - 2020",
x = "Location",
y = "Prevalence",
fill = "Age group") +
theme_classic()
ggsave("malaria-prevalence-age-boxplot.png", width = 10, height = 6, dpi = 300)5. Something to try————————-
#’ CHALLENGE 1: Create a figure showing how the Anopheles gambiae total counts #’ vary each day and by location.