-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathbetareg_phil_adjusted.R
182 lines (101 loc) · 5.41 KB
/
betareg_phil_adjusted.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
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
library(betareg)
SNP_list_parse <- function(string){
x <- strsplit(string, split = "-")[[1]]
y <- paste(x[1],x[2], sep=".")
return(y)
}
calc_allele_freq <- function(SNP){
x <- length(which(SNP == 0)) #AA
y <- length(which(SNP == 1)) #Aa
z <- length(which(SNP == 2)) #aa
a.freq <- (2*z + y)/(2*(x + y + z))
return(a.freq)
}
#setwd("/home/delores/Academic/LRRI2013/Phil/betareg/output")
#setwd("D:/Phil/betareg/output")
#unadjusted.reg.results <- read.csv("total_reg_results_unadjusted_postprocessed.csv", stringsAsFactors=FALSE)
setwd("D:/Phil/adjusted_betareg/data")
unadjusted.reg.results <- read.csv("sig_reg_results.csv", stringsAsFactors=FALSE)
ADC.meth.data <- read.csv("ADC_total_meth_data.csv", stringsAsFactors=FALSE)
SCC.meth.data <- read.csv("SCC_total_meth_data.csv", stringsAsFactors=FALSE)
SCC.meth.data <- SCC.meth.data[-anyDuplicated(SCC.meth.data$patient.name), ]
ADC.covariates <- read.csv("ADC_total_covariates.csv", stringsAsFactors=FALSE)
SCC.covariates <- read.csv("SCC_total_covariates.csv", stringsAsFactors=FALSE)
SCC.covariates <- SCC.covariates[-anyDuplicated(SCC.covariates$patient.ID), ]
results.table <- unadjusted.reg.results[ which(unadjusted.reg.results$ADC.or.SCC == "ADC"), c(1,2,3,4,5,6,10)]
M <- nrow(results.table)
results.table <- cbind(results.table, "ADC.or.SCC" = rep("", M), "Estimate" = rep(0, M), "P.Value" = rep(0, M), "Prevalence" = rep(0, M))
#remove 27k.
remove_27k <- function(data){
data <- data[which(data$platform.indicator == 1), -which(names(data) == platform.indicator)]
return(data)
}
#The function below runs beta regression on data subsets.
#Takes as parameters a list of genes, and a histology.
#The outer loops iterates over the methylation probes and the inner loop iterates ovet the SNP probes.
#Outputs a file for each iteration of the loop.
meth.data <- ADC.meth.data
covariates <- ADC.covariates
reg.results.table <- results.table
hist <- "ADC"
beta_reg <- function(meth.data, covariates, reg.results.table, hist){
#This chunk sets up results table.
N <- nrow(reg.results.table)
reg.results.table$ADC.or.SCC <- rep(hist, N)
for (i in 1:N){
meth.probe <- reg.results.table$Meth.Probe[i]
SNP.ID <- reg.results.table$SNP.ID[i]
SNP.ID <- SNP_list_parse(SNP.ID)
test.pair.ind <- list(match(meth.probe, names(meth.data)), match(SNP.ID, names(covariates)))
if (any(is.na(unlist(test.pair.ind) == TRUE))){
reg.results.table$Estimate[i] <- NA
reg.results.table$P.Value[i] <- NA
next
}
BV <- meth.data[c(test.pair.ind[[1]],5)]
row.names(BV) <- meth.data$patient.name
cov <- covariates[c(6, 7, 9, 10, 11, 12, test.pair.ind[[2]])]
row.names(cov) <- covariates$patient.ID
#Remove NAs, much of the 27k data is missing for probes of interest.
BV <- BV[which(is.na(BV[,1]) == FALSE), 1:2, drop=FALSE]
reg.table <- merge(BV, cov, by="row.names")
a.freq <- calc_allele_freq(reg.table[10])
reg.results.table$Prevalence[i] <- a.freq
#Calculate allele freq with SNP data being used in model.
if ((a.freq == 1)|(a.freq == 0)){
reg.results.table$Estimate[i] <- NA
reg.results.table$P.Value[i] <- NA
next
}
Beta_Value <- unlist(reg.table[2], use.names=FALSE)
Platform.Indicator <- unlist(reg.table[3], use.names=FALSE)
Ethnicity_NA <- unlist(reg.table[4], use.names=FALSE)
Ethnicity_Afro <- unlist(reg.table[5], use.names=FALSE)
Age <- unlist(reg.table[6], use.names=FALSE)
Gender <- unlist(reg.table[7], use.names=FALSE)
Smoking_Status <- as.factor(unlist(reg.table[8], use.names=FALSE))
Smoking_Status <- unlist(reg.table[8])
Stage <- unlist(reg.table[9], use.names=FALSE)
SNP <- unlist(reg.table[10], use.names=FALSE)
#Data with identical platform.indicator values throws error in regression.
if (all(Platform.Indicator == Platform.Indicator[1]) == TRUE){
model <- betareg(Beta_Value ~ SNP + Ethnicity_NA + Ethnicity_Afro + Age + Gender + Smoking_Status + Stage, link="logit")
reg.results.table$Estimate[i] <- model$coefficients$mean[2]
reg.results.table$P.Value[i] <- summary(model)$coefficients$mean[2,4]
} else{
model <- betareg(Beta_Value ~ SNP + Ethnicity_NA + Ethnicity_Afro + Age + Gender + Smoking_Status + Stage + Platform.Indicator, link="logit")
reg.results.table$Estimate[i] <- model$coefficients$mean[2]
reg.results.table$P.Value[i] <- summary(model)$coefficients$mean[2,4]
}
}
return(reg.results.table)
}
ADC.reg.results <- beta_reg(ADC.meth.data, ADC.covariates, results.table, "ADC")
#ADC.reg.results <- ADC.reg.results[-which(is.na(ADC.reg.results$P.Value) == TRUE), ]
SCC.reg.results <- beta_reg(SCC.meth.data, SCC.covariates, results.table, "SCC")
#SCC.reg.results <- SCC.reg.results[-which(is.na(SCC.reg.results$P.Value) == TRUE), ]
#setwd("/home/delores/Academic/LRRI2013/Phil/adjusted_betareg/output")
setwd("D:/Phil/adjusted_betareg/output")
names(ADC.reg.results) == names(SCC.reg.results)
x <- rbind(ADC.reg.results, SCC.reg.results)
write.csv(x, "adjusted_reg_results.csv", row.names=FALSE)