-
Notifications
You must be signed in to change notification settings - Fork 0
/
00_HLA_prep.R
45 lines (38 loc) · 1.52 KB
/
00_HLA_prep.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
rm(list = ls())
source("packages_load.R") #packages
# HLA data ####
setwd("")
# call HLA data
...
##generates HLA 2 digit alleles from the 4 digit alleles
# for (i in 1:(ncol(hladata %>% dplyr::select(-COHORT,-freq,-rs333_T,-starts_with("PC"))))) {
# # hladata[,i] <- str_pad(hladata[,i], 4, pad="0")
# hladata[,i] <- substr(hladata[,i], 1,4)
# }
# remove letters
hladata <- hladata %>% ungroup() %>%
dplyr::mutate(across(2:17, ~ str_replace_all(.,"[N|Q]","")))
# binomial
hlaBin <- hladata %>%
dplyr::select(-COHORT, -rs333_T, -starts_with("PC")) %>%
pivot_longer(cols = -ID, names_to = "variable", values_to = "value") %>%
drop_na() %>%
dplyr::mutate(variable = ifelse(str_detect(variable, "_"),
str_remove(variable, "_.*"),
str_remove(variable, "\\d+")),
cmb = paste(variable, value, sep = "_")) %>%
group_by(ID, cmb) %>%
dplyr::summarise(dm = 1) %>% # ignore homolozygous alleles (=2)
dplyr::arrange(cmb) %>%
pivot_wider(names_from = cmb, values_from = dm, values_fill = 0L) %>%
# mutate_all(~ ifelse(. > 0, 1, 0)) %>%
replace(is.na(.), 0L) %>%
dplyr::arrange(ID)
# adding PCs again
hlaBin <- full_join(hlaBin, hladata %>% dplyr::select(ID,starts_with("PC"),COHORT,rs333_T), by="ID")
fwrite(hlaBin, "hlaBin_4dig_corrected.csv",na = "NA")
# number of alleles ####
for (i in seq(2,16,by=2)) {
nr_alleles <- unique(c(hladata[i]),c(hladata[i+1])) %>% table() %>% length()
cat(colnames(hladata[i]), "has", nr_alleles, "alleles\n")
}