Cameroon SCD Blood Traits Description
Kevin Esoh
May 13, 2025
1 Load package
list.of.packages <- c("tidyverse","bestNormalize")
new.packages <- list.of.packages[!(list.of.packages %in% installed.packages()[,"Package"])]
if(length(new.packages))
install.packages(new.packages)
lapply(list.of.packages , require, character.only = T)## [[1]]
## [1] TRUE
##
## [[2]]
## [1] TRUE2 Load phenotype data and extract blood traits with relevant additional columns
pheno <- read.table(params$data, sep="\t", h=T)
keep_colmns <- c(
"FID","IID","SAMPLEID","GENEMAP_CODE",
"AGE","SEX", "GENDER","HEIGHT","WEIGHT","BMI","SBP","DBP",
"STATUS","HbA","HbA2","HbF","nHbF","HbS","HbC",
"RBC","Hb","MCV","HCT","MCHC","MCH","WBC",
"LYMPHOCYTE","MONOCYTE","NEUTROPHIL",
"EOSINOPHIL","BASOPHIL","PLATELET",
"RETICULOCYTE","RETICULOCYTES"
)
bt_pheno <- pheno[, which(names(pheno) %in% keep_colmns)]
bt_pheno$BMI <- (bt_pheno$WEIGHT)/(bt_pheno$HEIGHT)^23 Demographic data
3.1 AGE
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.67 10.00 16.00 17.60 23.00 66.00 10hist(bt_pheno$AGE, main="AGE destribution", xlab="AGE")
#plot(density(na.omit(bt_pheno$AGE)))
#polygon(density(na.omit(bt_pheno$AGE)), col = "coral")
#abline(v=c(min(bt_pheno$AGE), max(bt_pheno$AGE)), lty=2, lwd=1)
3.2 SEX
bt_pheno %>%
select(SEX) %>%
table() %>%
as.data.frame() %>%
na.omit() %>%
mutate(Prop = (Freq/sum(Freq))*100) -> sex # Sex
sexcolnames(sex) <- c("cat", "freqRBC", "prop")
par(mar=c(4,4,3,4))
barplot(
sex$prop,
ylim=c(0, 60),
col=c(2,4)
)
text(
x=c(0.7,1.9),
y=30,
labels=c("Female", "Male")
)
mtext(
text = "Proportion (%)",
side = 2,
line = 3
)
mtext(
text = "Sex",
side = 1,
line = 1
)
4 Normalize traits
4.1 RBC
# figures-side,
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$RBC, main="RBC (million/uL)")
plot(density(na.omit(bt_pheno$RBC)), main="RBC (million/uL)")
polygon(density(na.omit(bt_pheno$RBC)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$RBC)
bt_pheno$nRBC <- BNphen$x.t
plot(density(na.omit(bt_pheno$nRBC)), main="Normalized RBC")
polygon(density(na.omit(bt_pheno$nRBC)), col = "lightblue")
# fit smooth spline for Age against RBC
rbc_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","RBC"))])
rbc_plot <- qplot(AGE, RBC, data = rbc_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
rbc_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.2 Hb
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$Hb, main="Hb (g/dL)", xlab="Hb")
plot(density(na.omit(bt_pheno$Hb)), main="Hb (g/dL)")
polygon(density(na.omit(bt_pheno$Hb)), col = "coral")
# removing outliers
#abline(v=13, lwd=1, lty=2)
#bt_pheno$Hb <- as.numeric(ifelse(bt_pheno$Hb > 13, "NA", bt_pheno$Hb))
#plot(density(na.omit(bt_pheno$Hb)), main="Hb (>13) outlier pruned")
#polygon(density(na.omit(bt_pheno$Hb)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$Hb)
bt_pheno$nHb <- BNphen$x.t
plot(density(na.omit(bt_pheno$nHb)), main="Normalized Hb")
polygon(density(na.omit(bt_pheno$nHb)), col = "lightblue")
# fit smooth spline for Age against Hb
hb_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","Hb"))])
hb_plot <- qplot(AGE, Hb, data = hb_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
hb_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.3 HbF
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$HbF, main="HbF (%)", xlab="HbF")
plot(density(na.omit(bt_pheno$HbF)), main="HbF")
polygon(density(na.omit(bt_pheno$HbF)), col = "coral")
# removing outliers
abline(v=0, lwd=1, lty=2)
bt_pheno$HbF <- as.numeric(ifelse(bt_pheno$HbF == 0, "NA", bt_pheno$HbF))
plot(density(na.omit(bt_pheno$HbF)), main="HbF (0) outlier pruned")
polygon(density(na.omit(bt_pheno$HbF)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$HbF)
bt_pheno$onHbF <- BNphen$x.t
plot(density(na.omit(bt_pheno$onHbF)), main="Normalized HbF")
polygon(density(na.omit(bt_pheno$onHbF)), col = "lightblue")
# CUBIC ROOT TRANSFORMATION
bt_pheno$nHbF <- (bt_pheno$HbF)^(1/3)
plot(density(na.omit(bt_pheno$nHbF)), main="Cubic root transformed HbF")
polygon(density(na.omit(bt_pheno$nHbF)), col = "lightblue")
# fit smooth spline for Age against Hb
HbF_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","HbF"))])
HbF_plot <- qplot(AGE, HbF, data = HbF_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
HbF_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.4 MCV
par(mar = c(4, 4, 2, 2))
hist(pheno$MCV, main="MCV (fL)")
plot(density(na.omit(bt_pheno$MCV)), main="MCV (fL)")
polygon(density(na.omit(bt_pheno$MCV)), col = "coral")
# removing outliers
#abline(v=120, lwd=1, lty=2)
#bt_pheno$MCV <- as.numeric(ifelse(bt_pheno$MCV > 120, "NA", bt_pheno$MCV))
#plot(density(na.omit(bt_pheno$MCV)), main="MCV (>120) outlier pruned")
#polygon(density(na.omit(bt_pheno$MCV)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$MCV)
bt_pheno$onMCV <- BNphen$x.t
plot(density(na.omit(bt_pheno$onMCV)), main="Normalized (orderNorm) MCV")
polygon(density(na.omit(bt_pheno$onMCV)), col = "lightblue")
# fit smooth spline for Age against Hb
mcv_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","MCV"))])
mcv_plot <- qplot(AGE, MCV, data = mcv_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
mcv_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.5 HCT
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$HCT, main="HCT")
plot(density(na.omit(bt_pheno$HCT)), main="HCT")
polygon(density(na.omit(bt_pheno$HCT)), col = "coral")
# removing ourliers
#abline(v=43, lwd=1, lty=2)
#bt_pheno$HCT <- as.numeric(ifelse(bt_pheno$HCT > 43, "NA", bt_pheno$HCT))
#plot(density(na.omit(bt_pheno$HCT)), main="HCT (>43) outlier pruned")
#polygon(density(na.omit(bt_pheno$HCT)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$HCT)
bt_pheno$nHCT <- BNphen$x.t
plot(density(na.omit(bt_pheno$nHCT)), main="Normalized HCT")
polygon(density(na.omit(bt_pheno$nHCT)), col = "lightblue")
# fit smooth spline for Age against Hb
htc_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","HCT"))])
htc_plot <- qplot(AGE, HCT, data = htc_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
htc_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.6 MCHC
#Mean Corpuscular Hemoglobin Concentration
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$MCHC, main="MCHC (g/dL)")
plot(density(na.omit(bt_pheno$MCHC)), main="MCHC (g/dL)")
polygon(density(na.omit(bt_pheno$MCHC)), col = "coral")
# removing
abline(v=c(10, 60), lwd=1, lty=2)
bt_pheno$MCHC <- as.numeric(
ifelse(bt_pheno$MCHC > 60, "NA",
#bt_pheno$MCHC
ifelse(bt_pheno$MCHC < 10, "NA", bt_pheno$MCHC)
)
)
plot(density(na.omit(bt_pheno$MCHC)), main="MCHC (>60 and <10) outlier pruned")
polygon(density(na.omit(bt_pheno$MCHC)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$MCHC)
bt_pheno$nMCHC <- BNphen$x.t
plot(density(na.omit(bt_pheno$nMCHC)), main="Normalized MCHC")
polygon(density(na.omit(bt_pheno$nMCHC)), col = "lightblue")
# fit smooth spline for Age against Hb
mchc_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","MCHC"))])
mchc_plot <- qplot(AGE, MCHC, data = mchc_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
mchc_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.7 MCH
#The Mean Corpuscular Hemoglobin
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$MCH, main="MCH (pg/cell)")
plot(density(na.omit(bt_pheno$MCH)), main="MCH (pg/cell)")
polygon(density(na.omit(bt_pheno$MCH)), col = "coral")
# removing outliers
abline(v=10, lwd=1, lty=2)
bt_pheno$MCH <- as.numeric(
ifelse(bt_pheno$MCH < 10, "NA",
bt_pheno$MCH
#ifelse(bt_pheno$MCH < 10, "NA", bt_pheno$MCH)
)
)
plot(density(na.omit(bt_pheno$MCH)), main="MCH (<10) outlier pruned")
polygon(density(na.omit(bt_pheno$MCH)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$MCH)
bt_pheno$nMCH <- BNphen$x.t
plot(density(na.omit(bt_pheno$nMCH)), main="Normalized MCH")
polygon(density(na.omit(bt_pheno$nMCH)), col = "lightblue")
# fit smooth spline for Age against Hb
MCH_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","MCH"))])
MCH_plot <- qplot(AGE, MCH, data = MCH_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
MCH_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.8 WBC
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$WBC, main="WBC (x10^9 cells/L)")
plot(density(na.omit(bt_pheno$WBC)), main="WBC (x10^9 cells/L)")
polygon(density(na.omit(bt_pheno$WBC)), col = "coral")
# removing
abline(v=60, lwd=1, lty=2)
bt_pheno$WBC <- as.numeric(
ifelse(bt_pheno$WBC > 60, "NA",
bt_pheno$WBC
#ifelse(bt_pheno$WBC < 10, "NA", bt_pheno$WBC)
)
)
plot(density(na.omit(bt_pheno$WBC)), main="WBC (>60) outlier pruned")
polygon(density(na.omit(bt_pheno$WBC)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$WBC)
bt_pheno$nWBC <- BNphen$x.t
plot(density(na.omit(bt_pheno$nWBC)), main="Normalized WBC")
polygon(density(na.omit(bt_pheno$nWBC)), col = "lightblue")
# fit smooth spline for Age against Hb
wbc_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","WBC"))])
wbc_plot <- qplot(AGE, WBC, data = wbc_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
wbc_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.9 LYMPHOCYTE
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$LYMPHOCYTE, main="LYMPHOCYTE (x10^9 cells/L)")
plot(density(na.omit(bt_pheno$LYMPHOCYTE)), main="LYMPHOCYTE (x10^9 cells/L)")
polygon(density(na.omit(bt_pheno$LYMPHOCYTE)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$LYMPHOCYTE)
bt_pheno$nLYMPHOCYTE <- BNphen$x.t
plot(density(na.omit(bt_pheno$nLYMPHOCYTE)), main="Normalized LYMPHOCYTE")
polygon(density(na.omit(bt_pheno$nLYMPHOCYTE)), col = "lightblue")
# fit smooth spline for Age against Hb
lym_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","LYMPHOCYTE"))])
lym_plot <- qplot(AGE, LYMPHOCYTE, data = lym_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
lym_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.10 MONOCYTE
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$MONOCYTE, main="MONOCYTE (x10^9 cells/L)")
plot(density(na.omit(bt_pheno$MONOCYTE)), main="MONOCYTE (x10^9 cells/L)")
polygon(density(na.omit(bt_pheno$MONOCYTE)), col = "coral")
# removing
abline(v=0, lwd=1, lty=2)
bt_pheno$MONOCYTE <- as.numeric(
ifelse(bt_pheno$MONOCYTE == 0, "NA",
bt_pheno$MONOCYTE
#ifelse(bt_pheno$MONOCYTE == 0, "NA", bt_pheno$MONOCYTE)
)
)
plot(density(na.omit(bt_pheno$MONOCYTE)), main="MONOCYTE (=0) outlier pruned")
polygon(density(na.omit(bt_pheno$MONOCYTE)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$MONOCYTE)
bt_pheno$nMONOCYTE <- BNphen$x.t
plot(density(na.omit(bt_pheno$nMONOCYTE)), main="Normalized MONOCYTE")
polygon(density(na.omit(bt_pheno$nMONOCYTE)), col = "lightblue")
# fit smooth spline for Age against Hb
mon_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","MONOCYTE"))])
mon_plot <- qplot(AGE, MONOCYTE, data = mon_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
mon_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.11 NEUTROPHIL
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$NEUTROPHIL, main="NEUTROPHIL (g/L)")
plot(density(na.omit(bt_pheno$NEUTROPHIL)), main="NEUTROPHIL (g/L)")
polygon(density(na.omit(bt_pheno$NEUTROPHIL)), col = "coral")
# removing
abline(v=0, lwd=1, lty=2)
bt_pheno$NEUTROPHIL <- as.numeric(
ifelse(bt_pheno$NEUTROPHIL == 0, "NA",
bt_pheno$NEUTROPHIL
#ifelse(bt_pheno$NEUTROPHIL < 10, "NA", bt_pheno$NEUTROPHIL)
)
)
plot(density(na.omit(bt_pheno$NEUTROPHIL)), main="NEUTROPHIL (=0) outlier pruned")
polygon(density(na.omit(bt_pheno$NEUTROPHIL)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$NEUTROPHIL)
bt_pheno$nNEUTROPHIL <- BNphen$x.t
plot(density(na.omit(bt_pheno$nNEUTROPHIL)), main="Normalized NEUTROPHIL")
polygon(density(na.omit(bt_pheno$nNEUTROPHIL)), col = "lightblue")
# fit smooth spline for Age against NEUTROPHIL
neu_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","NEUTROPHIL"))])
neu_plot <- qplot(AGE, NEUTROPHIL, data = neu_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
neu_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.12 EOSINOPHIL
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$EOSINOPHIL, main="EOSINOPHIL (x10^9 cells/L)")
plot(density(na.omit(bt_pheno$EOSINOPHIL)), main="EOSINOPHIL (x10^9 cells/L)")
polygon(density(na.omit(bt_pheno$EOSINOPHIL)), col = "coral")
# removing
# abline(v=2, lwd=1, lty=2)
# bt_pheno$EOSINOPHIL <- as.numeric(
# ifelse(bt_pheno$EOSINOPHIL > 2, "NA",
# bt_pheno$EOSINOPHIL
# #ifelse(bt_pheno$EOSINOPHIL < 10, "NA", bt_pheno$EOSINOPHIL)
# )
# )
#
# plot(density(na.omit(bt_pheno$EOSINOPHIL)), main="EOSINOPHIL (>2) outlier pruned")
# polygon(density(na.omit(bt_pheno$EOSINOPHIL)), col = "coral")
BNphen <- bestNormalize(bt_pheno$EOSINOPHIL)
bt_pheno$nEOSINOPHIL <- BNphen$x.t
plot(density(na.omit(bt_pheno$nEOSINOPHIL)), main="Normalized EOSINOPHIL")
polygon(density(na.omit(bt_pheno$nEOSINOPHIL)), col = "lightblue")
# fit smooth spline for Age against EOSINOPHIL
eu_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","EOSINOPHIL"))])
eu_plot <- qplot(AGE, EOSINOPHIL, data = eu_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
eu_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.13 BASOPHIL
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$BASOPHIL, main="BASOPHIL")
plot(density(na.omit(bt_pheno$BASOPHIL)), main="BASOPHIL")
polygon(density(na.omit(bt_pheno$BASOPHIL)), col = "coral")
# removing
abline(v=1, lwd=1, lty=2)
bt_pheno$BASOPHIL <- as.numeric(
ifelse(bt_pheno$BASOPHIL > 1, "NA",
bt_pheno$BASOPHIL
#ifelse(bt_pheno$BASOPHIL < 10, "NA", bt_pheno$BASOPHIL)
)
)
plot(density(na.omit(bt_pheno$BASOPHIL)), main="BASOPHIL (>1) outlier pruned")
polygon(density(na.omit(bt_pheno$BASOPHIL)), col = "coral")
BNphen <- bestNormalize(bt_pheno$BASOPHIL)
bt_pheno$nBASOPHIL <- BNphen$x.t
plot(density(na.omit(bt_pheno$nBASOPHIL)), main="Normalized BASOPHIL")
polygon(density(na.omit(bt_pheno$nBASOPHIL)), col = "lightblue")
# binarize basophil after removing outliers (1-control, 2-case)
# manual (10.1016/j.ajhg.2021.08.007)
bt_pheno$bBASOPHIL <- as.numeric(
ifelse(bt_pheno$BASOPHIL < 0.05, 1,
#bt_pheno$BASOPHIL
ifelse(bt_pheno$BASOPHIL >= 0.05, 2, "NA")
)
)
bt_pheno$cBASOPHIL <- as.numeric(
ifelse(bt_pheno$BASOPHIL < 0.05, 0,
#bt_pheno$BASOPHIL
ifelse(bt_pheno$BASOPHIL >= 0.05, 1, "NA")
)
)
hist(bt_pheno$bBASOPHIL, main="binarized")
# fit smooth spline for Age against BASOPHIL
bas_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","BASOPHIL"))])
bas_plot <- qplot(AGE, BASOPHIL, data = bas_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
bas_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.14 PLATELET
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$PLATELET, main="PLATELET (x10^9/L)")
plot(density(na.omit(bt_pheno$PLATELET)), main="PLATELET (x10^9/L)")
polygon(density(na.omit(bt_pheno$PLATELET)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$PLATELET)
bt_pheno$nPLATELET <- BNphen$x.t
plot(density(na.omit(bt_pheno$nPLATELET)), main="Normalized PLATELET")
polygon(density(na.omit(bt_pheno$nPLATELET)), col = "lightblue")
# fit smooth spline for Age against PLATELET
pla_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","PLATELET"))])
pla_plot <- qplot(AGE, PLATELET, data = pla_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
pla_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.15 RETICULOCYTES
# RETICULOCYTE PER 1000 RBCs
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$RETICULOCYTE, main="RETICULOCYTE")
plot(density(na.omit(bt_pheno$RETICULOCYTE)), main="RETICULOCYTE")
polygon(density(na.omit(bt_pheno$RETICULOCYTE)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$RETICULOCYTE)
bt_pheno$nRETICULOCYTE <- BNphen$x.t
plot(density(na.omit(bt_pheno$nRETICULOCYTE)), main="Normalized RETICULOCYTE")
polygon(density(na.omit(bt_pheno$nRETICULOCYTE)), col = "lightblue")
# fit smooth spline for Age against PLATELET
ret_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","RETICULOCYTE"))])
ret_plot <- qplot(AGE, RETICULOCYTE, data = ret_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
ret_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.16 RETICULOCYTE
# RETICULOCYTE /uL
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$RETICULOCYTES, main="RETICULOCYTES /uL")
plot(density(na.omit(bt_pheno$RETICULOCYTES)), main="RETICULOCYTES /uL")
polygon(density(na.omit(bt_pheno$RETICULOCYTES)), col = "coral")
# removing outliers
abline(v=1000000, lwd=1, lty=2)
bt_pheno$RETICULOCYTES <- as.numeric(
ifelse(bt_pheno$RETICULOCYTES > 1000000, "NA",
bt_pheno$RETICULOCYTES
#ifelse(bt_pheno$RETICULOCYTES < 10, "NA", bt_pheno$RETICULOCYTES)
)
)
plot(density(na.omit(bt_pheno$RETICULOCYTES)), main="RETICULOCYTES (>1000000) outlier pruned")
polygon(density(na.omit(bt_pheno$RETICULOCYTES)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$RETICULOCYTES)
bt_pheno$nRETICULOCYTES <- BNphen$x.t
plot(density(na.omit(bt_pheno$nRETICULOCYTES)), main="Normalized RETICULOCYTES")
polygon(density(na.omit(bt_pheno$nRETICULOCYTES)), col = "lightblue")
# fit smooth spline for Age against RETICULOCYTES
rets_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","RETICULOCYTES"))])
rets_plot <- qplot(AGE, RETICULOCYTES, data = rets_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
rets_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.17 SBP
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$SBP, main="SBP")
plot(density(na.omit(bt_pheno$SBP)), main="SBP")
polygon(density(na.omit(bt_pheno$SBP)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$SBP)
bt_pheno$nSBP <- BNphen$x.t
plot(density(na.omit(bt_pheno$nSBP)), main="Normalized SBP")
polygon(density(na.omit(bt_pheno$nSBP)), col = "lightblue")
# fit smooth spline for Age against RETICS
sbp_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","SBP"))])
sbp_plot <- qplot(AGE, SBP, data = sbp_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
sbp_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.18 DBP
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$DBP, main="DBP")
plot(density(na.omit(bt_pheno$DBP)), main="DBP")
polygon(density(na.omit(bt_pheno$DBP)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$DBP)
bt_pheno$nDBP <- BNphen$x.t
plot(density(na.omit(bt_pheno$nDBP)), main="Normalized DBP")
polygon(density(na.omit(bt_pheno$nDBP)), col = "lightblue")
# fit smooth spline for Age against DBP
dbp_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","DBP"))])
dbp_plot <- qplot(AGE, DBP, data = dbp_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
dbp_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
4.19 BMI
par(mar = c(4, 4, 2, 2))
hist(bt_pheno$BMI, main="BMI")
plot(density(na.omit(bt_pheno$BMI)), main="BMI")
polygon(density(na.omit(bt_pheno$BMI)), col = "coral")
# removing outliers
abline(v=40, lwd=1, lty=2)
bt_pheno$BMI <- as.numeric(
ifelse(bt_pheno$BMI > 40, "NA",
bt_pheno$BMI
#ifelse(bt_pheno$BMI < 10, "NA", bt_pheno$BMI)
)
)
plot(density(na.omit(bt_pheno$BMI)), main="BMI (>40) outlier pruned")
polygon(density(na.omit(bt_pheno$BMI)), col = "coral")
BNphen <- bestNormalize::orderNorm(bt_pheno$BMI)
bt_pheno$nBMI <- BNphen$x.t
plot(density(na.omit(bt_pheno$nBMI)), main="Normalized BMI")
polygon(density(na.omit(bt_pheno$nBMI)), col = "lightblue")
# fit smooth spline for Age against BMI
bmi_pheno <- na.omit(bt_pheno[, which(names(bt_pheno) %in% c("FID","GENDER","AGE","BMI"))])
bmi_plot <- qplot(AGE, BMI, data = bmi_pheno, geom='smooth', span =0.5, color = GENDER) + theme_bw()
bmi_plot## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
