所有作品合集传送门: Tidy Tuesday
2018 年合集传送门: 2018
Global Mortality
What do people die from?
在过去的几个世纪里,世界发生了很大的变化–这就是《我们的世界》的数据所显示的。然而,有一件事在这种转变中一直保持不变:我们都必须在某个时候死亡。然而,随着生活水平的提高、医疗保健的进步和生活方式的改变,死亡的原因正在发生变化。
在这篇博客中,我们试图回答 “人们死于什么?”,首先看一下全球死因的数据,然后选择国家层面的例子。
世界各地的主要死因仍有很大差异,因此,也可以选择了一些国家,以突出这种异质性。
本次示例通过一些可视化方式来展示这些信息。
1. 一些环境设置
# 设置为国内镜像, 方便快速安装模块
options("repos" = c(CRAN = "https://mirrors.tuna.tsinghua.edu.cn/CRAN/"))
2. 设置工作路径
wkdir <- '/home/user/R_workdir/TidyTuesday/2018/2018-04-16_Global_Mortality/src-f'
setwd(wkdir)
3. 加载 R 包
library(scales)
library(tidyverse)
library(glue)
library(gridExtra)
library(ggpubr)
library(showtext)
# 在 Ubuntu 系统上测试的, 不加这个我画出来的汉字会乱码 ~
showtext_auto()
4. 加载数据
df_input <- readxl::read_excel("../data/global_mortality.xlsx")
# 简要查看数据内容
glimpse(df_input)
## Rows: 6,156
## Columns: 35
## $ country <chr> "Afghanistan", "Afghanistan…
## $ country_code <chr> "AFG", "AFG", "AFG", "AFG",…
## $ year <dbl> 1990, 1991, 1992, 1993, 199…
## $ `Cardiovascular albertes (%)` <dbl> 17.61040, 17.80181, 18.3868…
## $ `Cancers (%)` <dbl> 4.025975, 4.054145, 4.17395…
## $ `Respiratory diseases (%)` <dbl> 2.106626, 2.134176, 2.20829…
## $ `Diabetes (%)` <dbl> 3.832555, 3.822228, 3.90012…
## $ `Dementia (%)` <dbl> 0.5314287, 0.5324973, 0.540…
## $ `Lower respiratory infections (%)` <dbl> 10.886362, 10.356968, 10.09…
## $ `Neonatal deaths (%)` <dbl> 9.184653, 8.938897, 8.84138…
## $ `Diarrheal diseases (%)` <dbl> 2.497141, 2.572228, 2.70774…
## $ `Road accidents (%)` <dbl> 3.715944, 3.729142, 3.81635…
## $ `Liver disease (%)` <dbl> 0.8369093, 0.8455159, 0.874…
## $ `Tuberculosis (%)` <dbl> 5.877075, 5.891704, 6.03466…
## $ `Kidney disease (%)` <dbl> 1.680611, 1.671115, 1.70098…
## $ `Digestive diseases (%)` <dbl> 1.058771, 1.049322, 1.06288…
## $ `HIV/AIDS (%)` <dbl> 0.01301948, 0.01451458, 0.0…
## $ `Suicide (%)` <dbl> 0.4366105, 0.4422802, 0.456…
## $ `Malaria (%)` <dbl> 0.4488863, 0.4550191, 0.460…
## $ `Homicide (%)` <dbl> 1.287020, 1.290991, 1.32616…
## $ `Nutritional deficiencies (%)` <dbl> 0.3505045, 0.3432123, 0.345…
## $ `Meningitis (%)` <dbl> 3.037603, 2.903202, 2.84064…
## $ `Protein-energy malnutrition (%)` <dbl> 0.3297599, 0.3221711, 0.323…
## $ `Drowning (%)` <dbl> 0.9838624, 0.9545860, 0.951…
## $ `Maternal albert (%)` <dbl> 1.769213, 1.749264, 1.76424…
## $ `Parkinson disease (%)` <dbl> 0.02515859, 0.02545063, 0.0…
## $ `Alcohol disorders (%)` <dbl> 0.02899828, 0.02917152, 0.0…
## $ `Intestinal infectious diseases (%)` <dbl> 0.1833303, 0.1781074, 0.176…
## $ `Drug disorders (%)` <dbl> 0.04120540, 0.04203340, 0.0…
## $ `Hepatitis (%)` <dbl> 0.1387378, 0.1350081, 0.134…
## $ `Fire (%)` <dbl> 0.1741567, 0.1706712, 0.171…
## $ `Heat-related (hot and cold exposure) (%)` <dbl> 0.1378229, 0.1348266, 0.139…
## $ `Natural disasters (%)` <dbl> 0.00000000, 0.79760256, 0.3…
## $ `Conflict (%)` <dbl> 0.932, 2.044, 2.408, NA, 4.…
## $ `Terrorism (%)` <dbl> 0.007, 0.040, 0.027, NA, 0.…
# 检查数据的列名
colnames(df_input)
## [1] "country"
## [2] "country_code"
## [3] "year"
## [4] "Cardiovascular diseases (%)"
## [5] "Cancers (%)"
## [6] "Respiratory diseases (%)"
## [7] "Diabetes (%)"
## [8] "Dementia (%)"
## [9] "Lower respiratory inalbertons (%)"
## [10] "Neonatal deaths (%)"
## [11] "Diarrheal diseases (%)"
## [12] "Road accidents (%)"
## [13] "Liver disease (%)"
## [14] "Tuberculosis (%)"
## [15] "Kidney disease (%)"
## [16] "Digestive diseases (%)"
## [17] "HIV/AIDS (%)"
## [18] "Suicide (%)"
## [19] "Malaria (%)"
## [20] "Homicide (%)"
## [21] "Nutritional deficiencies (%)"
## [22] "Meningitis (%)"
## [23] "Protein-energy malnutrition (%)"
## [24] "Drowning (%)"
## [25] "Maternal deaths (%)"
## [26] "Parkinson disease (%)"
## [27] "Alcohol disorders (%)"
## [28] "Intestinal infectious diseases (%)"
## [29] "Drug disorders (%)"
## [30] "Hepatitis (%)"
## [31] "Fire (%)"
## [32] "Heat-related (hot and cold exposure) (%)"
## [33] "Natural disasters (%)"
## [34] "Conflict (%)"
## [35] "Terrorism (%)"
5. 数据预处理
df_tidy = df_input %>%
# pivot_longer() 从宽数据透视到长数据转换
pivot_longer(names_to = "Cause", cols = contains("%"), values_to = "Percent") %>%
# 建议使用 dplyr::mutate 形式调用函数, 不然容易与 plyr 中的函数冲突 (因为我自己就报错了...)
dplyr::mutate(
Percent = Percent/100,
Cause = str_trim(str_to_title(sub(" [(]%[)]", "", Cause)))
)
# 删除缺失值的观测
df_tidy <- na.omit(df_tidy)
# 简要查看数据内容
glimpse(df_tidy)
## Rows: 167,808
## Columns: 5
## $ country <chr> "Afghanistan", "Afghanistan", "Afghanistan", "Afghanistan…
## $ country_code <chr> "AFG", "AFG", "AFG", "AFG", "AFG", "AFG", "AFG", "AFG", "…
## $ year <dbl> 1990, 1990, 1990, 1990, 1990, 1990, 1990, 1990, 1990, 199…
## $ Cause <chr> "Cardiovascular Diseases", "Cancers", "Respialbert Diseas…
## $ Percent <dbl> 0.1761039712, 0.0402597540, 0.0210662613, 0.0383255475, 0…
6. 生成绘图所需数据
# 从中挑选一些需要关注的点
Year <- 2016
Disease <- 'Cardiovascular Diseases'
# 根据给定的年份和致死因素筛选数据
df_plot <- df_tidy %>% filter(Cause == Disease, year == Year)
# 获得各个国家的地图信息
country.map <- df_plot %>%
inner_join(maps::iso3166, by = c(country_code = "a3")) %>%
inner_join(map_data("world"), by = c(mapname = "region"))
# 获得给定的致死因素致死人数占各个国家死亡人口的百分比
country_bar <- df_plot %>% slice_max(Percent, n = 15) %>%
dplyr::mutate(country = fct_reorder(country, Percent))
7. 利用 ggplot2 绘图
# PS: 方便讲解, 我这里进行了拆解, 具体使用时可以组合在一起
gg <- ggplot(data = country.map,
aes(long, lat, group = group, text = paste(round(Percent*100, 2) , "% of Deaths in", country, "were from", Disease)))
# geom_polygon() 绘制的是多边形
gg <- gg + geom_polygon(aes(fill = Percent))
gg <- gg + borders("world")
# coord_cartesian() 默认坐标系统是笛卡尔坐标系
gg <- gg + coord_cartesian(ylim = c(-50, 90))
# scale_fill_gradientn() 将颜色比例转换为概率转换颜色分布
gg <- gg + scale_fill_gradientn(colours = c("#98FB98", "#FF4500", "#191970"), labels = percent_format())
gg <- gg + labs(x = "", y = "")
# theme_minimal() 去坐标轴边框的最小化主题
gg <- gg + theme_minimal()
gg <- gg + theme(
# legend.position 设置图例位置, "bottom" 表示图例置于下方
legend.position = "bottom",
# legend.title 设置图例标题
legend.title = element_blank(),
# plot.margin 调整图像边距, 上-右-下-左
plot.margin = unit(c(1, 1.5, .5, .5), "cm"))
pal = colorRampPalette(colors = c("#98FB98", "#FF4500", "#191970"))
hh <- ggplot(data = country_bar, aes(country, Percent, fill = country), alpha = 0.8)
# geom_col() 绘制条形图
hh <- hh + geom_col()
# coord_flip() 倒置坐标系
hh <- hh + coord_flip()
# scale_y_continuous() 对连续变量设置坐标轴显示范围
hh <- hh + scale_y_continuous(labels = percent_format())
# scale_fill_manual() 采取的是手动赋值的方法, 也就是直接把颜色序列赋值给它的参数 value
hh <- hh + scale_fill_manual(values = rev(pal(15)))
# guides() 设置图例信息, 4 列, 按行排序
hh <- hh + guides(fill = guide_legend(nrow = 4, byrow = TRUE))
hh <- hh + labs(x = "", y = "")
# theme_minimal() 去坐标轴边框的最小化主题
hh <- hh + theme_minimal()
hh <- hh + theme(
# legend.direction 设置图例的方向
legend.direction = "horizontal",
# legend.title 设置图例标题
legend.title = element_blank(),
# legend.position 设置图例位置, 这里指定精确的坐标
legend.position = c(0.22, -0.25),
# plot.margin 调整图像边距, 上-右-下-左
plot.margin = unit(c(1, 1.5, 3.5, 0.5), "cm"),
# panel.background 面板背景 数据下面
panel.background = element_rect(colour = "#CAEAFF", fill = "#CAEAFF"))
8. 保存图片到 PDF 和 PNG
filename = '20180416-F-01'
png(paste0(filename, ".png"), width = 920, height = 500)
grid.arrange(gg, hh, ncol = 2, widths = c(3, 2),
top = text_grob(label = glue("全世界每个国家由心血管疾病造成的死亡百分比·{Year}年"),
hjust = .5, vjust = 1.0,
size = 20,
face = "bold" ))
dev.off()
9. session-info
sessionInfo()
## R version 4.2.1 (2022-06-23)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 20.04.5 LTS
##
## Matrix products: default
## BLAS: /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/liblapack.so.3
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] showtext_0.9-5 showtextdb_3.0 sysfonts_0.8.8 ggpubr_0.4.0
## [5] gridExtra_2.3 plotly_4.10.0 maps_3.4.0 glue_1.6.2
## [9] extrafont_0.18 tidytext_0.3.4 ggthemes_4.2.4 forcats_0.5.2
## [13] stringr_1.4.1 dplyr_1.0.10 purrr_0.3.4 readr_2.1.2
## [17] tidyr_1.2.1 tibble_3.1.8 ggplot2_3.3.6 tidyverse_1.3.2
## [21] ggbump_0.1.0 scales_1.2.1
##
## loaded via a namespace (and not attached):
## [1] fs_1.5.2 lubridate_1.8.0 httr_1.4.4
## [4] SnowballC_0.7.0 tools_4.2.1 backports_1.4.1
## [7] bslib_0.4.0 utf8_1.2.2 R6_2.5.1
## [10] DBI_1.1.3 lazyeval_0.2.2 colorspace_2.0-3
## [13] withr_2.5.0 tidyselect_1.1.2 compiler_4.2.1
## [16] extrafontdb_1.0 cli_3.3.0 rvest_1.0.3
## [19] xml2_1.3.3 labeling_0.4.2 sass_0.4.2
## [22] digest_0.6.29 rmarkdown_2.16 pkgconfig_2.0.3
## [25] htmltools_0.5.3 dbplyr_2.2.1 fastmap_1.1.0
## [28] htmlwidgets_1.5.4 rlang_1.0.5 readxl_1.4.1
## [31] rstudioapi_0.14 farver_2.1.1 jquerylib_0.1.4
## [34] generics_0.1.3 jsonlite_1.8.0 tokenizers_0.2.3
## [37] car_3.1-0 googlesheets4_1.0.1 magrittr_2.0.3
## [40] Matrix_1.5-1 Rcpp_1.0.9 munsell_0.5.0
## [43] fansi_1.0.3 abind_1.4-5 lifecycle_1.0.1
## [46] stringi_1.7.8 yaml_2.3.5 carData_3.0-5
## [49] grid_4.2.1 crayon_1.5.1 lattice_0.20-45
## [52] haven_2.5.1 hms_1.1.2 knitr_1.40
## [55] pillar_1.8.1 ggsignif_0.6.3 reprex_2.0.2
## [58] evaluate_0.16 data.table_1.14.2 modelr_0.1.9
## [61] vctrs_0.4.1 tzdb_0.3.0 Rttf2pt1_1.3.10
## [64] cellranger_1.1.0 gtable_0.3.1 assertthat_0.2.1
## [67] cachem_1.0.6 xfun_0.32 broom_1.0.1
## [70] rstatix_0.7.0 janeaustenr_1.0.0 googledrive_2.0.0
## [73] viridisLite_0.4.1 gargle_1.2.1 ellipsis_0.3.2
测试数据
配套数据下载:global_mortality.xlsx