-
Notifications
You must be signed in to change notification settings - Fork 0
/
coxRegressionSnp.R
70 lines (57 loc) · 2.25 KB
/
coxRegressionSnp.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
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
# Content: Cox regression for SNP score of each Olink feature
library(tidyverse)
library(data.table)
library(parallel)
library(pbmcapply)
library(survival)
RhpcBLASctl::blas_set_num_threads(2)
load("snpScoreData.RData")
allEvents <- c("Myeloid")
clinicalFeatures <- grep("indel_|cnv_|eid", colnames(clinical), value = TRUE, invert = TRUE)
indelFeatures <- grep("indel_", colnames(clinical), value = TRUE)
cnvFeatures <- grep("cnv_", colnames(clinical), value = TRUE)
chFeatures <- c(indelFeatures, cnvFeatures)
proteomicFeatures <- colnames(data)[2:ncol(data)]
featureSetList <- c(list(clinicalFeatures), list(c(clinicalFeatures, "indel_maxVaf", "cnv_allCnv")))
names(featureSetList) <- c("clinical", "cliVafCnv")
featureSetList <- featureSetList[sort(names(featureSetList))]
# Cox regression
for (featureSet in names(featureSetList)) {
allResult <- list()
for (event in allEvents) {
result <- pbmclapply(proteomicFeatures,
mc.cores = nCores, ignore.interactive = TRUE,
function(col) {
set.seed(1)
df <- data %>% dplyr::select(eid, all_of(col))
df <- df %>% left_join(clinical %>% dplyr::select(eid, any_of(featureSetList[[featureSet]])),
by = "eid"
)
tmp <- c(event, paste0("survtime_", event))
df <- df %>%
left_join(diagnosis %>% dplyr::select(eid, all_of(tmp)), by = "eid") %>%
dplyr::select(-eid)
df[[event]] <- replace_na(df[[event]], 0)
df <- df %>% drop_na()
survObj <- df %>% dplyr::select(all_of(tmp))
colnames(survObj) <- c("event", "time")
survObj <- Surv(survObj$time, survObj$event)
dfClinical <- df %>% dplyr::select(-all_of(tmp))
form <- as.formula("survObj ~ .")
model <- coxph(form, data = dfClinical)
modelSummary <- summary(model)
cIndex <- concordance(model)$concordance
list(
pVal = modelSummary$coefficients[col, "Pr(>|z|)"],
cIndex = cIndex,
coef = modelSummary$coefficients,
numCase = sum(df[[event]] == 1),
numControl = sum(df[[event]] == 0)
)
}
)
allResult[[event]] <- result
names(allResult[[event]]) <- proteomicFeatures
}
saveRDS(allResult, str_glue("Result/coxRegSnp_{featureSet}.rds"))
}