2019-12-11用ggcor画相关性图
2019-12-11 本文已影响0人
iColors
节选自http://houyun.xyz/post/2019/10/26/ggcor/
安装绘图R包(还处在开发阶段)
if(!require(devtools))
install.packages("devtools")
if(!require(ggcor))
devtools::install_github("houyunhuang/ggcor", force = TRUE)
绘制基本图，如同corrplot包一样方便
library(ggcor)
## Loading required package: ggplot2
## ggcor可以调整的参数比较多，并且处于开发阶段，主要的参数是以下几个：
## function(
## x,
## y = NULL,
## type = c("full", "upper", "lower"),
## show.diag = FALSE,
## cor.test = FALSE,
## cor.test.alt = "two.sided",
## cor.test.method = "pearson",
## cluster.type = c("none", "all", "row", "col"),
## cluster.method = "HC",
## cluster.absolute = FALSE,
## keep.name = FALSE,
## ...)
## ggcor提供了定制的geom_square()、geom_circle2()、geom_ellipse2()、geom_pie2()、geom_colour()、geom_confbox()、geom_num()、geom_mark()、geom_cross()9个ggplot2图层函数，可以根据需要进行叠加。除了ggplot2中一般化的参数（x、y、fill、colour、size等）最常用参数r、p、low、upp、num、r0、sig.thres、sig.level、mark等。
ggcor(mtcars) + geom_square()
image.png
ggcor(mtcars, type = "upper") + geom_circle2()
image.png
ggcor(mtcars, type = "lower", show.diag = TRUE) + geom_ellipse2()
image.png
ggcor(mtcars, type = "full", cluster.type = "all") + geom_pie2()
## Registered S3 method overwritten by 'gclus':
## method from
## reorder.hclust vegan
## Registered S3 method overwritten by 'dendextend':
## method from
## rev.hclust vegan
## Registered S3 method overwritten by 'seriation':
## method from
## reorder.hclust gclus
image.png
ggcor(mtcars, cluster.type = "all") +
geom_colour() +
geom_num(aes(num = r), colour = "grey90", size = 3.5)#可以标记相关系数的图层
image.png
ggcor(mtcars, type = "full", cor.test = TRUE) + geom_confbox()
image.png
ggcor(mtcars, type = "full", cor.test = TRUE, cluster.type = "all") +
geom_colour() +
geom_cross()
image.png
ggcor(mtcars, type = "full", cor.test = TRUE, cluster.type = "all") +
geom_raster() +
geom_mark(sig.thres = 0.05, size = 3, colour = "grey90")#过滤掉不显著的值
image.png
ggcor(mtcars, type = "full", cor.test = TRUE, cluster.type = "all") +
geom_raster() +
geom_mark(r = NA, sig.thres = 0.05, size = 5, colour = "grey90")#不标记系数，只标记显著符号
image.png
ggcor(mtcars, type = "full", cor.test = TRUE, cluster.type = "all") +
geom_raster() +
geom_mark(r = NA, sig.level = 0.05, mark = "*", vjust = 0.65, size = 6, colour = "grey90")
image.png
绘制非对称相关系数图，这个功能非常独有
library(vegan) # 使用vegan包所带的数据集
## Loading required package: permute
## Loading required package: lattice
## This is vegan 2.5-6
data(varechem)
data(varespec)
df03 <- fortify_cor(x = varechem, y = varespec[ , 1:30], cluster.type = "col")#调用fortify_cor()函数来处理数据
ggcor(df03) + geom_colour()
image.png
df04 <- fortify_cor(x = varespec[ , 1:30], y = varechem, cor.test = TRUE)
ggcor(df04) + geom_square() + geom_cross(size = 0.2)
image.png
绘制上下三角不一样相关系数图
df05 <- fortify_cor(x = varechem, cor.test = TRUE, cluster.type = "all")
ggcor(df05) + geom_circle2()
image.png
df05_lower <- get_lower_data(df05, show.diag = FALSE)
ggcor(df05_lower) + geom_circle2()
image.png
ggcor(df05) +
geom_pie2(data = get_data(type = "upper", show.diag = FALSE)) +
geom_ellipse2(data = get_data(type = "lower", show.diag = TRUE))
image.png
ggcor(df05) +
geom_segment(aes(x = x - 0.5, y = y + 0.5, xend = x + 0.5, yend = y - 0.5),
data = get_data(type = "diag"), size = 0.5, colour = "grey60") +
geom_colour(data = get_data(type = "upper", show.diag = FALSE)) +
geom_mark(data = get_data(type = "upper", show.diag = FALSE), size = 3) +
geom_circle2(data = get_data(r >= 0.5, type = "lower", show.diag = FALSE),
r = 0.8, fill = "#66C2A5") +
geom_num(aes(num = r), data = get_data(type = "lower",
show.diag = FALSE), size = 3)
image.png
列名放在对角线
ggcor(mtcars, cor.test = TRUE, cluster.type = "all") +
geom_confbox(data = get_data(type = "upper", show.diag = FALSE)) +
geom_num(aes(num = r), data = get_data(type = "lower", show.diag = FALSE), size = 3.5) +
add_diaglab(size = 4.56) + remove_axis()
image.png
mantel 检验组合图,这个是最酷的功能
ggcor提供了mantel检验的封装函数fortify_mantel()，支持vegan包中的mantel()、mantel.partial()和ade4包中的mantel.randtest()、mantel.rtest()函数，差别上说mantel.partial()是偏mantel检验（有控制变量），其它三个是mantel检验.
mantel <- fortify_mantel(varespec, varechem, spec.select = list(spec01 = 22:25,
spec02 = 1:4,
spec03 = 38:43,
spec04 = 15:20))
df06 <- as_cor_tbl(mantel)#转换成ggcor可用的数据格式
ggcor(df06) + geom_pie2() + geom_cross()
image.png
corr <- fortify_cor(varechem, type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
mantel <- fortify_mantel(varespec, varechem,
spec.select = list(spec01 = 22:25,
spec02 = 1:4,
spec03 = 38:43,
spec04 = 15:20),
mantel.fun = "mantel.randtest")
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(mantel, diag.label = TRUE) +
add_diaglab(angle = 45) +
geom_square() + remove_axis("y")
image.png
corr <- fortify_cor(varechem, type = "upper", show.diag = FALSE,
cor.test = TRUE, cluster.type = "all")
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(mantel, diag.label = TRUE) +
add_diaglab(angle = 45) +
geom_pie2() + remove_axis("y")
image.png
corr <- fortify_cor(varechem, type = "lower", show.diag = FALSE,
cor.test = TRUE, cluster.type = "all")
ggcor(corr, xlim = c(0.5, 20)) +
add_link(mantel, diag.label = TRUE) +
add_diaglab(angle = 45) +
geom_ellipse2() + remove_axis("y")
image.png
corr <- fortify_cor(varechem, varechem[ , 1:7], type = "full", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
mantel <- fortify_mantel(varespec, varechem,
spec.select = list(spec01 = 22:25,
spec02 = 1:4,
spec03 = 38:43,
spec04 = 15:20),
mantel.fun = "mantel.randtest", nrepet = 2000)
extra.params <- extra_params(group.label = text_params(size = 6),
link.params = link_params(group.point.hjust = 2))
ggcor(corr, axis.y.position = "left", legend.position = "left", xlim = c(0.5, 14.5)) +
add_link(mantel, extra.params = extra.params) +
geom_circle2()
image.png
group <- rep(LETTERS[1:3], 8)
corr <- fortify_cor(varechem, type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
mantel <- fortify_mantel(varespec[ , 38:43], varechem,
spec.group = group,
env.group = group,
is.pair = TRUE,
mantel.fun = "mantel.randtest")
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(mantel, diag.label = TRUE) +
add_diaglab(angle = 45) +
geom_colour() + geom_shade(sign = -1) +
remove_axis("y")
image.png
corr <- fortify_cor(varechem, type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
mantel <- fortify_mantel(varespec, varechem,
spec.select = list(spec01 = 22:25,
spec02 = 1:4,
spec03 = 38:43,
spec04 = 15:20),
mantel.fun = "mantel.randtest")
mantel <- dplyr::filter(mantel, p <= 0.05)
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(mantel, diag.label = TRUE, legend.drop = TRUE) +
add_diaglab(angle = 45) +
geom_square() + remove_axis("y")
image.png
corr <- fortify_cor(varechem, type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
corr01 <- fortify_cor(varechem, varespec[ , 38:39], type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
## Warning: 'type=upper' just supports for symmetric correlation matrix.
mantel <- fortify_mantel(varespec, varechem,
spec.select = list(spec01 = 22:25,
spec02 = 1:4,
spec03 = 38:43,
spec04 = 15:20),
mantel.fun = "mantel.randtest")
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(x = corr01, diag.label = TRUE,
link.line.colours = c("#E31A1C", "#33A02C")) +
add_diaglab(angle = 45) +
geom_square() + remove_axis("y")
image.png
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(x = corr01, mapping = aes(size = abs(r)), diag.label = TRUE) +
add_diaglab(angle = 45) +
geom_square() +
scale_size_continuous(limits = c(0, 1), range = c(0.25, 3)) +
guides(size = guide_legend(title = "abs r", override.aes = list(colour = "grey35"),
order = 1)) +
remove_axis("y")
image.png
corr <- fortify_cor(varechem, type = "upper", show.diag = TRUE,
cor.test = TRUE, cluster.type = "all")
df <- data.frame(x = rep(LETTERS[1:3], 14),
y = rep(cor_tbl_yname(corr), 3),
r = runif(42, -1, 1),
p = runif(42, 0, 0.5), stringsAsFactors = FALSE)
ggcor(corr, xlim = c(-5, 14.5)) +
add_link(df, diag.label = TRUE, colour = "red") +
add_diaglab(angle = 45) +
geom_square() + remove_axis("y")
image.png
