data-raw/geo_prep.R

In PublicHealthEngland/hcaidcs: R functions for the mandatory surveillance of bacteraemia and C. difficile infection

library("rgdal") # librarys sp, will use proj.4 if installed
library("maptools")
library("ggplot2")
library("plyr")
library("rgeos")
library("dplyr")

setwd("H:\\spatial\\")

area_teams_sp <- readOGR(dsn = ".", layer = "AT_Eng")
area_teams_sp@data$id = rownames(area_teams_sp@data)
at_points <- fortify(area_teams_sp)
at_sp_df <- join(at_points, area_teams_sp@data, by = "id")

ggplot(at_sp_df, aes(long, lat, group = group)) + geom_polygon() +
  geom_path(colour = "white") + coord_equal()

save(at_sp_df, file = "at_sp_df.RData")

rm(list = ls())

# Create subregions ####

subregions <- readOGR(dsn = ".", layer = "nhs_subregions_and_das_2018")
subregions@data$id = rownames(subregions@data)
subregions_points <- fortify(subregions)
subregions_sp_df <- join(subregions_points, subregions@data, by = "id")
subregions_sp_df <- subregions_sp_df[!(subregions_sp_df$CTRYNM %in% c("Northern Ireland", "Scotland", "Wales")), ]
table(subregions_sp_df$CTRYNM)
table(subregions_sp_df$GSS_CD)
table(subregions_sp_df$GSS_CD, subregions_sp_df$GSS_NM) %>% View()

centroids_df <- data.frame(gCentroid(subregions, byid=TRUE, id = subregions@data$id))
centroids_df <- data.frame(rownames(centroids_df),centroids_df)
rownames(centroids_df) <- c(1:nrow(centroids_df))
names(centroids_df) <- c("id" ,"centroid_long", "centroid_lat")
# offset east and central mids
centroids_df$centroid_long[centroids_df$id == 6] <- centroids_df$centroid_long[centroids_df$id == 6] - (centroids_df$centroid_long[centroids_df$id == 6] * 0.05)
# Q78 (E39000030) could be more central, i.e. shifted left
centroids_df$centroid_long[centroids_df$id == 5] <- centroids_df$centroid_long[centroids_df$id == 5] - (centroids_df$centroid_long[centroids_df$id == 5] * 0.05)
# And Q70
centroids_df$centroid_long[centroids_df$id == 1] <- centroids_df$centroid_long[centroids_df$id == 1] - (centroids_df$centroid_long[centroids_df$id == 1] * 0.01)
centroids_df$centroid_lat[centroids_df$id == 1] <- centroids_df$centroid_lat[centroids_df$id == 1] - (centroids_df$centroid_lat[centroids_df$id == 1] * 0.05)
subregions_sp_df <- left_join(subregions_sp_df, centroids_df)

# Add in ODS codes here
# Need ODS parent code and ODS code for region
# codes is manually created from ODS and PHE GIS data
codes <- read.csv("H:/hcaidcs/data-raw/subregion_lut_2018.csv",
                  stringsAsFactors = FALSE)

subregions_sp_df <- subregions_sp_df %>%
  left_join(., select(codes, -GSS_NM))

names(subregions_sp_df)
length(unique(subregions_sp_df$ODS_CD))
table(subregions_sp_df$ODS_CD)

rm(codes)

ggplot(subregions_sp_df, aes(long, lat, group = group, label = GSS_CD)) +
  geom_polygon() +
  geom_path(colour = "white") + coord_equal() +
  geom_text(aes(x = centroid_long, y = centroid_lat), colour = "white")

ggplot(subregions_sp_df, aes(long, lat, group = group, label = ODS_CD)) +
  geom_polygon() +
  geom_path(colour = "white") + coord_equal() +
  geom_text(aes(x = centroid_long, y = centroid_lat), colour = "white")

subregions_sp_df$CTRYCD  <- NULL
subregions_sp_df$CTRYCDO  <- NULL
subregions_sp_df$CTRYNM <- NULL

setwd("H:/hcaidcs/")
devtools::use_data(subregions_sp_df, overwrite = TRUE)
# save(subregions_sp_df, file = "H:\\hcaidcs\\data\\subregions_sp_df.RData")

subregion_test_data <- data.frame(stringsAsFactors=FALSE,
           ODS_CD = c("Q71", "Q75", "Q72", "Q78", "Q79", "Q76", "Q77", "Q83", "Q74",
                      "Q84", "Q87", "Q88", "Q86", "Q85"),
           rand_val = c(0.302020973, 0.006223132, 0.156590506, 0.045990521,
                        0.35983688, 0.148314024, 0.497859566, 0.673062958,
                        0.317183955, 0.757254947, 0.800067725, 0.926615495, 0.591537639,
                        0.951676355)
)

devtools::use_data(subregion_test_data, overwrite = TRUE)

setwd("H:/spatial/")
rm(list = ls())