inst/data-raw/process/PMID29415062_Simeonovic-2018/process.R

In avucoh/nPOD: What the Package Does (One Line, Title Case)

library(readxl)
library(tidyr)
library(dplyr)
library(purrr)

# S1 Table downloaded from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5802856/bin/pone.0191360.s006.xlsx

normal <- read_excel("pone.0191360.s006.xlsx", range = "A4:F100") %>%
  filter(!Reduce(`&`, lapply(., is.na))) %>%
  rename(ID = `pancreas IDb`) %>%
  mutate(ID = gsub("#", "", ID),
         ID = purrr::accumulate(ID, function(i1, i2) if(is.na(i2)) i1 else i2)) %>%
  mutate_all(as.numeric) %>%
  group_by(ID) %>%
  summarize_at(c("HS", "Col18", "Sdc1", "Insulin", "Glucagon"), list(`prct` = mean, `prct_SD` = sd), na.rm = T) %>%
  mutate_all(~replace(., is.nan(.), NA)) %>%
  rename_all(function(x) gsub("_prct", ".prct", x))

t1d <- read_excel("pone.0191360.s006.xlsx", range = "G4:L100") %>%
  filter(!Reduce(`&`, lapply(., is.na))) %>%
  rename(ID = `pancreas ID`) %>%
  mutate(ID = gsub("#", "", ID),
         ID = purrr::accumulate(ID, function(i1, i2) if(is.na(i2)) i1 else i2)) %>%
  mutate_all(as.numeric) %>%
  group_by(ID) %>%
  summarize_at(c("HS", "Col18", "Sdc1", "Insulin", "Glucagon"), list(`prct` = mean, `prct_SD` = sd), na.rm = T) %>%
  mutate_all(~replace(., is.nan(.), NA)) %>%
  rename_all(function(x) gsub("_prct", ".prct", x))

dataset <- rbind(normal, t1d) %>%
  mutate_all(round, digits = 3)
write.table(dataset, "PMID29415062_1_Simeonovic-2018.tsv", sep = "\t", quote = F, row.names = F)