如何自定义两个因素的X轴和基于两个因素的颜色条 [英] How to customize the X axis of two factors and colour bars based on two factors
问题描述
我已经制作了一个图表，其中的数据显示了基于两个主要因素(季节和曝光)在两个不同地点发生的位移，这两个因素都在X轴上进行了标记.我想按季节编辑X轴，所以它不是按字母顺序排列的，而是从春季-冬季(而不是秋季到冬季)标记的.在第二行显示曝光度"；有两套暴露标签；我可以看到，这是因为它在季节"标签的每三个标签中居中，但似乎无法纠正它.
I've produced a plot with data showing dislodgement at two different sites based on two main factors (Season and Exposure) which are both labelled in the X axis. I'd like to edit the X axis for seasons so it is not in alphabetical order but labelled from Spring - Winter (instead of Autumn to Winter). On the second row showing Exposure; the are two sets of Exposed labels; I can see that this is because it is being centered in one every three labels of the Season labels but can't seem to correct it.
如果可能的话，我想根据季节更改条形的颜色，第二个站点使用相同颜色的浅色.理想情况下，春季为绿色，夏季为黄色，秋季为棕色，冬季为灰色，并在季节之间放置刻度，在两次曝光之间放置较长的刻度
If possible too, I would like to change the colours of my bars based on season with a lighter shade of the same colour for the second site. Ideally, green for Spring, Yellow for Summer, Brown for Autumn and Grey for Winter, and place ticks in between the seasons and a longer tick in between exposure
我尝试使用此代码；
Season <- as.character(data$Season)
#Then turn it back into a factor with the levels in the correct order
Season <- factor(data$Season), levels=unique(data$Season)
要更正Seasons的字母顺序，即使在我更正了csv中的标签之后，它也无济于事.归档到正确的顺序.
I've tried using this code;
Season <- as.character(data$Season)
#Then turn it back into a factor with the levels in the correct order
Season <- factor(data$Season), levels=unique(data$Season)
To correct the alphabetical order for Seasons but it does nothing, even after I have corrected the labels in my csv. file to the correct order.
这是我目前正在使用的完整代码，由堆栈溢出用户针对上一个问题为我提供.
This is my full code that I'm using at the moment that was kindly provided for me by a stack overflow user on a previous question.
library(ggplot2)
library(gtable)
library(grid)
Season <- (data$Season)
Site <- (data$Site)
Exposure <- (data$Exposure)
Average <- data$Average
SEM <- data$SEM
gg <- ggplot(aes(x=as.factor(Site), y=Average, fill=as.factor(Site)), Season, data = data)
gg <- gg + geom_bar(stat = 'identity')
gg <- gg + scale_fill_discrete(guide_legend(title = 'Site'))
gg <- gg + scale_fill_manual(values=c('black', 'grey85'), guide_legend(title = 'Site'))
gg <- gg + theme_classic()
gg <- gg + geom_errorbar(aes(ymin=Average-SEM, ymax=Average+SEM), width=.3)
gg <- gg + facet_wrap(~Exposure*Season, strip.position=c('bottom'), nrow=1, drop=F)
gg <- gg + scale_y_continuous(expand = expand_scale(mult = c(0, .05)))
gg <- gg + theme(axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.title.x = element_blank(),
axis.line = element_line(color='black'),
strip.placement = 'outside',
panel.spacing.x=unit(0, "lines"),
panel.grid.major.x = element_blank(),
panel.grid = element_blank(),
panel.background = element_rect(fill='white'),
strip.background = element_rect(fill='white', color='white')
)
print(gg)
season.levels <- levels(data$Season)
exposure.levels <- levels(data$Exposure)
g <- ggplotGrob(gg)
grob.numbers <- grep("strip-b", g$layout$name)
b.strips <- gtable_filter(g, "strip-b", trim = FALSE)
season.left.panels <- seq(1, by=length(levels(data$Exposure)), length.out = length(season.levels))
season.right.panels <- seq(length(exposure.levels), by=length(exposure.levels), length.out = length(season.levels))
left <- b.strips$layout$l[season.left.panels]
right <- b.strips$layout$r[season.right.panels]
top <- b.strips$layout$t[3]
bottom <- b.strips$layout$b[3]
mat <- matrix(vector("list", length = 10), nrow = 2)
mat[] <- list(zeroGrob())
for (i in 1:length(season.levels)) {
res <- gtable_matrix("season.strip", mat, unit(c(1, 0, 1, 0, 1), "null"), unit(c(1, 1), "null"))
season.left <- season.left.panels[i]
res <- gtable_add_grob(res, g$grobs[[grob.numbers[season.left]]]$grobs[[3]], 2, 1, 2, 5)
for (j in 0:2) {
exposure.x <- season.left+j
res$grobs[[c(1, 5, 9)[j+1]]] <- g$grobs[[grob.numbers[exposure.x]]]$grobs[[3]]
}
new.grob.name <- paste0(levels(data$Season)[i], '-strip')
g <- gtable_add_grob(g, res, t = top, l = left[i], b = top, r = right[i], name = c(new.grob.name))
new.grob.no <- grep(new.grob.name, g$layout$name)[3]
g$grobs[[new.grob.no]]$grobs[[nrow(g$grobs[[new.grob.no]]$layout)]]$children[[3]]$children[[3]]$gp <- gpar(fontface='bold')
}
grid.newpage()
grid.draw(g)
推荐答案
您可以使用alpha美感为每个Site获取不同的颜色阴影，然后手动分配所需的颜色:
You can use the alpha aesthetic to get different shades of the color per Site and then manually assign your desired colors:
data$Season <- factor(data$Season, levels=c('Spring', 'Summer', 'Autumn', 'Winter'))
data$Site <- as.factor(data$Site)
gg <- ggplot(aes(x=Site, y=Average, fill=Season), data=data)
gg <- gg + geom_bar(stat = 'identity', aes(alpha=Site))
gg <- gg + scale_alpha_manual(values=c(1, .7), guide_legend(title = 'Site'))
gg <- gg + scale_fill_manual(values=c('green', 'yellow', 'brown', 'grey'), guide_legend(title = 'Season')) # to get bars desired colors instead of ggplot's default colors
gg <- gg + theme_classic() # get white background and black axis.line for x- and y-axis
gg <- gg + geom_errorbar(aes(ymin=Average-SEM, ymax=Average+SEM), width=.3)
gg <- gg + facet_wrap(~Season*Exposure, strip.position=c('bottom'), nrow=1, drop=F)
gg <- gg + scale_y_continuous(expand = expand_scale(mult = c(0, .05))) # remove space below zero
gg <- gg + theme(axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.title.x = element_blank(),
axis.line = element_line(color='black'),
strip.placement = 'outside', # place x-axis above (factor-label-) strips
panel.spacing.x=unit(0, "lines"), # remove space between facets (for continuous x-axis)
panel.grid.major.x = element_blank(), # remove vertical grid lines
# panel.grid = element_blank(), # remove all grid lines
# panel.background = element_rect(fill='white'), # choose background color for plot area
strip.background = element_rect(fill='white', color='white') # choose background for factor labels, color just matters for theme_classic()
)
print(gg)
这篇关于如何自定义两个因素的X轴和基于两个因素的颜色条的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
