使用真棒ggplot包，我想要一个条形图，其中fill美学被映射到一个连续变量，实际上是值，在它上面是一个纹理，实际上是条纹和交叉影线。
颜色渐变很重要，因为它代表重要性，而纹理将显示类别“A”、“B”以及它们的重叠。因此，对角线分别以一种方式、相反方式和交叉剖面线。 我知道维恩图可以完成这项工作，但我们有 35 个样本，我认为比较会更容易看到。
The fill is trivial, however, the texture one is tricky. On SO, thanks to @baptise https://stackoverflow.com/users/471093/baptiste (see here https://stackoverflow.com/questions/26110160/how-to-apply-cross-hatching-to-a-polygon-using-the-grid-graphical-system/26110400#26110400 and here https://stackoverflow.com/questions/25014492/geom-bar-pictograms-how-to) I managed to get this output with dummy data:
问题是透明度十字架的背景。如果可能的话得到alpha在背景下工作grid.patternFill()功能那就太好了。不幸的是，在我手中它不起作用。
任何帮助都感激不尽。
虚拟数据可以在此处加载：
dfso <- structure(list(Sample = c("S1", "S2", "S1", "S2", "S1", "S2"),
qvalue = c(14.704287341, 8.1682824035, 13.5471896224, 6.71158432425,
12.3900919038, 5.254886245), type = structure(c(1L, 1L, 2L,
2L, 3L, 3L), .Label = c("A", "overlap", "B"), class = "factor"),
value = c(897L, 1082L, 503L, 219L, 388L, 165L)), class = c("tbl_df",
"tbl", "data.frame"), row.names = c(NA, -6L), .Names = c("Sample",
"qvalue", "type", "value"))
代码在这里：
library("ggplot2")
library("gridSVG")
library("gridExtra")
library("dplyr")
library("RColorBrewer")
cols <- brewer.pal(7,"YlOrRd")
pso <- ggplot(dfso)+
geom_bar(aes(x = Sample, y = value, fill = qvalue, linetype = type), width = .8, colour = "black", stat = "identity", position = "stack", alpha = 1)+
theme_classic(18)+
theme( panel.grid.major = element_line(colour = "grey80"),
panel.grid.major.x = element_blank(),
panel.grid.minor = element_blank(),
legend.key = element_blank(),
axis.text.x = element_text(angle = 90, vjust = 0.5))+
guides(linetype = FALSE) +
ylab("Count")+
scale_fill_gradientn("-log10(qvalue)", colours = cols, limits = c(0, 20))+
scale_linetype_manual(values = c("dotted", "solid", "dotted"))+
scale_y_continuous(expand = c(0, 0), limits = c(0, 2000))
# gridSVG
pat1 <- pattern(linesGrob(gp = gpar(col="black", lwd = 1)),
width = unit(5, "mm"), height = unit(5, "mm"),
dev.width = 1, dev.height = 1)
pat2 <- pattern(linesGrob(x = unit(0:1, "npc"), y = unit(1:0, "npc"),
gp = gpar(col="black", lwd = 1)),
width = unit(5, "mm"), height = unit(5, "mm"),
dev.width = 1, dev.height = 1)
crossGrob <- gTree(children = gList(linesGrob(gp = gpar(col="black", lwd = 1)), linesGrob(x = unit(0:1, "npc"), y = unit(1:0, "npc"), gp = gpar(col="black", lwd = 1))))
registerPatternFill("hash1", pat1)
registerPatternFill("hash2", pat2)
registerPatternFill("cross", grob = crossGrob, dev.width = 1, dev.height = 1, width = unit(5, "mm"), height = unit(5, "mm"))
gridsvg("crossbars.svg", width = 10)
print(pso)
grid.force()
grid.patternFill("geom_rect.rect", alpha = 0.2, grep = TRUE, group = FALSE,
label = rep(c("hash1", "cross", "hash2"), 1))
dev.off()
In grid.patternFill the alpha据我了解，参数应该提供透明度。但是，它没有任何效果，并且颜色丢失。我只填充了第一个条的图案，以便您看到对比。
Edit: the alpha工作正常，但它作用于图案本身，i.e线条。这就解释了为什么线条显得如此苍白。问题更多在于背景被假定为白色且不透明。
The linetype映射是试图突出显示重叠部分，但这并不好。如果透明度有效gridSVG我将放弃这部分并始终保持实线。
提前谢谢了，
Aurelien
如果有任何用途，输出sessionInfo():
R version 3.2.1 (2015-06-18)
Platform: x86_64-apple-darwin14.3.0 (64-bit)
Running under: OS X 10.10.4 (Yosemite)
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] grid stats graphics grDevices utils datasets methods base
other attached packages:
[1] gridExtra_0.9.1 RColorBrewer_1.1-2 dplyr_0.4.1 gridSVG_1.4-3 ggplot2_1.0.1
loaded via a namespace (and not attached):
[1] Rcpp_0.11.6 XML_3.98-1.3 assertthat_0.1 digest_0.6.8 MASS_7.3-42 plyr_1.8.3 DBI_0.3.1
[8] gtable_0.1.2 magrittr_1.5 scales_0.2.5 stringi_0.5-5 reshape2_1.4.1 labeling_0.3 proto_0.3-10
[15] RJSONIO_1.3-0 tools_3.2.1 stringr_1.0.0 munsell_0.4.2 parallel_3.2.1 colorspace_1.2-6
I am answering my own question, as there is a way to fill a specific color to the pattern thank to this link https://www.stat.auckland.ac.nz/~paul/Reports/leaf/leaf.html#fig-barchart-pattern.
On the first bar, for category "A", it could look like this:
更换pat1模式通过以下代码：
pat1 <- pattern(gTree(children=gList(
rectGrob(gp=gpar(col=NA, fill=cols[4])),
linesGrob(gp=gpar(col="black", lwd = 5)))),
width = unit(5, "mm"), height = unit(5, "mm"),
dev.width = 1, dev.height = 1)
For geom_bar颜色很少，它会起作用，但对于我的问题来说fill颜色映射到热图比例，这会很乏味。