在 R 地图上绘制来自 csv 文件的数据，突出显示它们的不同来源
Plotting data from a csv file on a R map, highlighting their different origin
我是 R 的新手，我发现绘制这些数据非常困难
我已经获得了我的地图（使用 rworldmap）。这是我使用的两行代码：
library(rworldmap)
newmap <- getMap(resolution = "low")
plot(newmap)
plot(newmap, xlim = c(-180, 180), ylim = c(-90, 90), asp = 1, fill=TRUE, col="white", bg="lightblue",)
然后我做了如下操作以显示我的.csv文件的内容
coordinates <- read.csv("seqmap.csv", header=T, fill=T, col.names=c("ID CODE", "latitude", "longitude", "sample nature", "seq.methods"), row.names=NULL)
head(coordinates)
现在问题来了：
我想要多边形：
SQUARE 样品性质=沉积物
CIRCLE 样品性质=水
并作为所选多边形的填充颜色
1) seq.methods = Illumina
的红色
2) seq.methods = 454
的蓝色
3) seq.methods 为绿色 = Ion torrent
4) 黄色代表 seq.methods = Sanger
举个例子：
对于使用 Illumina 测序的沉积物样本：以红色作为填充颜色的正方形
对于序列为 454 的水样：以蓝色作为填充色的圆圈
我被困在这一点上，据我所知，我觉得我不能再进一步了:(
在下面 link 你可以找到 .csv 文件
https://www.dropbox.com/s/t812lf6xgc0d6kf/seqmap.csv?dl=0
提前感谢您的帮助，对于任何 grammar/spelling 错误，我深表歉意，英语不是我的母语 :)
如果你可以接受字符而不是多边形，你可能想像这样尝试：
library(rworldmap)
newmap <- getMap(resolution = "low")
download.file("https://www.dropbox.com/s/t812lf6xgc0d6kf/seqmap.csv?dl=1", csv <- tempfile(fileext = ".csv"))
coordinates <- read.csv(csv, header=T, fill=T, sep=";", col.names=c("ID CODE", "latitude", "longitude", "sample nature", "seq.methods"), row.names=NULL)
pdf(pdffile <- tempfile(fileext = ".pdf"), width = 80, height = 40)
plot(newmap, xlim = c(-180, 180), ylim = c(-90, 90), asp = 1, fill=TRUE, col="white", bg="lightblue",)
with(coordinates,
points(x=longitude, y=latitude, cex=.5, # reduce symbol size a bit
col=adjustcolor(c("Illumina"="red", "454"="blue", "ion torrent"="green", "Sanger"="yellow")[as.character(seq.methods)], alpha.f = .5),
pch=c("sediment"=15, "water"=19)[as.character(sample.nature)]
)
)
dev.off()
shell.exec(pdffile) # open pdf doc on windows to zoom/pan easily...
由于您的数据点遍布世界各地，当您在一张地图上绘制所有内容时，很难区分地图上的点。作为替代方案，您可以创建一个带有世界多个地方特写镜头的多面图。我建议采取以下步骤：
1: 阅读数据并确保他们有正确的 class:
coordinates <- read.csv2("seqmap.csv", header=T, fill=T,
col.names=c("ID.CODE", "latitude", "longitude", "sample.nature", "seq.methods"),
row.names=NULL)
coordinates$latitude <- as.numeric(as.character(coordinates$latitude))
coordinates$longitude <- as.numeric(as.character(coordinates$longitude))
2:加载需要的库：
library(ggplot2)
library(ggmap)
library(grid)
library(gridExtra)
3: 创建点簇并为点分配簇值：
km <- kmeans(coordinates[,c("longitude","latitude")], centers = 6)
coordinates$cluster <- km$cluster
4: 获取不同集群的映射：
> km$centers
longitude latitude
1 -156.7000 71.37000
2 -88.3875 28.82875
3 -39.0700 -17.65200
4 139.3400 35.88667
5 94.4150 -67.29500
6 -64.5000 32.17000
map1 <- get_map(location = c(lon = -156.7, lat = 71.37), zoom = 7)
map2 <- get_map(location = c(lon = -88.3875, lat = 28.82875), zoom = 9)
map3 <- get_map(location = c(lon = -39.07, lat = -17.652), zoom = 9)
map4 <- get_map(location = c(lon = 139.34, lat = 35.88667), zoom = 8)
map5 <- get_map(location = c(lon = 94.415, lat = -67.295), zoom = 4)
map6 <- get_map(location = c(lon = -64.5, lat = 32.17), zoom = 10)
5: 创建地图并将它们存储为对象：
p1 <- ggmap(map1) +
geom_point(data=coordinates[coordinates$cluster==1,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21,"water"=22)) +
scale_fill_manual(values=c("Illumina"="red")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Alaska")
p2 <- ggmap(map2) +
geom_point(data=coordinates[coordinates$cluster==2,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21,"water"=22)) +
scale_fill_manual(values=c("Illumina"="red","ion torrent"="green")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("New Orleans")
p3 <- ggmap(map3) +
geom_point(data=coordinates[coordinates$cluster==3,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("454"="blue")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Brazil")
p4 <- ggmap(map4) +
geom_point(data=coordinates[coordinates$cluster==4,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21)) +
scale_fill_manual(values=c("454"="blue")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Japan")
p5 <- ggmap(map5) +
geom_point(data=coordinates[coordinates$cluster==5,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("Sanger"="yellow")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Antartica")
p6 <- ggmap(map6) +
geom_point(data=coordinates[coordinates$cluster==6,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("454"="blue","Illumina"="red")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Bermuda")
6: 创建一个单独的图例：
p0 <- ggplot(data=coordinates, aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods, color=seq.methods)) +
geom_point(size=4) +
scale_shape_manual("Sample:",values=c("sediment"=21,"water"=22)) +
scale_fill_manual("Method:",values=c("454"="blue","Illumina"="red","ion torrent"="green","Sanger"="yellow")) +
scale_color_manual("Method:",values=c("454"="blue","Illumina"="red","ion torrent"="green","Sanger"="yellow")) +
theme(legend.key=element_rect(fill=NA))
# function to extract the legend (borrowed from: https://github.com/hadley/ggplot2/wiki/Share-a-legend-between-two-ggplot2-graphs )
g_legend<-function(a.gplot){
tmp <- ggplot_gtable(ggplot_build(a.gplot))
leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
legend <- tmp$grobs[[leg]]
return(legend)}
legend <- g_legend(p0)
lwidth <- sum(legend$width)
7: 创建最终图：
grid.arrange(arrangeGrob(p1, p2, p3, p4, p5, p6, ncol=2),
legend, widths=unit.c(unit(1, "npc") - lwidth, lwidth), nrow=1)
结果如下：
我是 R 的新手，我发现绘制这些数据非常困难 我已经获得了我的地图（使用 rworldmap）。这是我使用的两行代码：
library(rworldmap)
newmap <- getMap(resolution = "low")
plot(newmap)
plot(newmap, xlim = c(-180, 180), ylim = c(-90, 90), asp = 1, fill=TRUE, col="white", bg="lightblue",)
然后我做了如下操作以显示我的.csv文件的内容
coordinates <- read.csv("seqmap.csv", header=T, fill=T, col.names=c("ID CODE", "latitude", "longitude", "sample nature", "seq.methods"), row.names=NULL)
head(coordinates)
现在问题来了： 我想要多边形：
SQUARE 样品性质=沉积物
CIRCLE 样品性质=水
并作为所选多边形的填充颜色
1) seq.methods = Illumina
的红色2) seq.methods = 454
的蓝色3) seq.methods 为绿色 = Ion torrent
4) 黄色代表 seq.methods = Sanger
举个例子： 对于使用 Illumina 测序的沉积物样本：以红色作为填充颜色的正方形
对于序列为 454 的水样：以蓝色作为填充色的圆圈
我被困在这一点上，据我所知，我觉得我不能再进一步了:( 在下面 link 你可以找到 .csv 文件 https://www.dropbox.com/s/t812lf6xgc0d6kf/seqmap.csv?dl=0
提前感谢您的帮助，对于任何 grammar/spelling 错误，我深表歉意，英语不是我的母语 :)
如果你可以接受字符而不是多边形，你可能想像这样尝试：
library(rworldmap)
newmap <- getMap(resolution = "low")
download.file("https://www.dropbox.com/s/t812lf6xgc0d6kf/seqmap.csv?dl=1", csv <- tempfile(fileext = ".csv"))
coordinates <- read.csv(csv, header=T, fill=T, sep=";", col.names=c("ID CODE", "latitude", "longitude", "sample nature", "seq.methods"), row.names=NULL)
pdf(pdffile <- tempfile(fileext = ".pdf"), width = 80, height = 40)
plot(newmap, xlim = c(-180, 180), ylim = c(-90, 90), asp = 1, fill=TRUE, col="white", bg="lightblue",)
with(coordinates,
points(x=longitude, y=latitude, cex=.5, # reduce symbol size a bit
col=adjustcolor(c("Illumina"="red", "454"="blue", "ion torrent"="green", "Sanger"="yellow")[as.character(seq.methods)], alpha.f = .5),
pch=c("sediment"=15, "water"=19)[as.character(sample.nature)]
)
)
dev.off()
shell.exec(pdffile) # open pdf doc on windows to zoom/pan easily...
由于您的数据点遍布世界各地，当您在一张地图上绘制所有内容时，很难区分地图上的点。作为替代方案，您可以创建一个带有世界多个地方特写镜头的多面图。我建议采取以下步骤：
1: 阅读数据并确保他们有正确的 class:
coordinates <- read.csv2("seqmap.csv", header=T, fill=T,
col.names=c("ID.CODE", "latitude", "longitude", "sample.nature", "seq.methods"),
row.names=NULL)
coordinates$latitude <- as.numeric(as.character(coordinates$latitude))
coordinates$longitude <- as.numeric(as.character(coordinates$longitude))
2:加载需要的库：
library(ggplot2)
library(ggmap)
library(grid)
library(gridExtra)
3: 创建点簇并为点分配簇值：
km <- kmeans(coordinates[,c("longitude","latitude")], centers = 6)
coordinates$cluster <- km$cluster
4: 获取不同集群的映射：
> km$centers
longitude latitude
1 -156.7000 71.37000
2 -88.3875 28.82875
3 -39.0700 -17.65200
4 139.3400 35.88667
5 94.4150 -67.29500
6 -64.5000 32.17000
map1 <- get_map(location = c(lon = -156.7, lat = 71.37), zoom = 7)
map2 <- get_map(location = c(lon = -88.3875, lat = 28.82875), zoom = 9)
map3 <- get_map(location = c(lon = -39.07, lat = -17.652), zoom = 9)
map4 <- get_map(location = c(lon = 139.34, lat = 35.88667), zoom = 8)
map5 <- get_map(location = c(lon = 94.415, lat = -67.295), zoom = 4)
map6 <- get_map(location = c(lon = -64.5, lat = 32.17), zoom = 10)
5: 创建地图并将它们存储为对象：
p1 <- ggmap(map1) +
geom_point(data=coordinates[coordinates$cluster==1,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21,"water"=22)) +
scale_fill_manual(values=c("Illumina"="red")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Alaska")
p2 <- ggmap(map2) +
geom_point(data=coordinates[coordinates$cluster==2,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21,"water"=22)) +
scale_fill_manual(values=c("Illumina"="red","ion torrent"="green")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("New Orleans")
p3 <- ggmap(map3) +
geom_point(data=coordinates[coordinates$cluster==3,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("454"="blue")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Brazil")
p4 <- ggmap(map4) +
geom_point(data=coordinates[coordinates$cluster==4,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("sediment"=21)) +
scale_fill_manual(values=c("454"="blue")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Japan")
p5 <- ggmap(map5) +
geom_point(data=coordinates[coordinates$cluster==5,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("Sanger"="yellow")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Antartica")
p6 <- ggmap(map6) +
geom_point(data=coordinates[coordinates$cluster==6,], aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods), size=3) +
scale_shape_manual(values=c("water"=22)) +
scale_fill_manual(values=c("454"="blue","Illumina"="red")) +
guides(shape=FALSE, fill=FALSE) +
ggtitle("Bermuda")
6: 创建一个单独的图例：
p0 <- ggplot(data=coordinates, aes(x=longitude, y=latitude, shape=sample.nature, fill=seq.methods, color=seq.methods)) +
geom_point(size=4) +
scale_shape_manual("Sample:",values=c("sediment"=21,"water"=22)) +
scale_fill_manual("Method:",values=c("454"="blue","Illumina"="red","ion torrent"="green","Sanger"="yellow")) +
scale_color_manual("Method:",values=c("454"="blue","Illumina"="red","ion torrent"="green","Sanger"="yellow")) +
theme(legend.key=element_rect(fill=NA))
# function to extract the legend (borrowed from: https://github.com/hadley/ggplot2/wiki/Share-a-legend-between-two-ggplot2-graphs )
g_legend<-function(a.gplot){
tmp <- ggplot_gtable(ggplot_build(a.gplot))
leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
legend <- tmp$grobs[[leg]]
return(legend)}
legend <- g_legend(p0)
lwidth <- sum(legend$width)
7: 创建最终图：
grid.arrange(arrangeGrob(p1, p2, p3, p4, p5, p6, ncol=2),
legend, widths=unit.c(unit(1, "npc") - lwidth, lwidth), nrow=1)
结果如下：