Related
I'd like to place this data onto a bar plot using ggplot2
where the column "Clades" would be placed on the X axis and the values from each column (such as the values of 19A, for example) would be place on Y axis
I'm trying something like this:
cols = as.vector(names(snv_data)[2:19])
ggplot(df, aes(x=cols)) + geom_bar()
But I keep getting this:
I'm new to ggplot2 so any help is very welcome!
I'm doing this to try and get 7 plots (one for each column such as 19A, 20A, 20B, etc) where each plot would have the Clades on the X-axis and each value from each column as the "counts" on the Y-axis
dput:
structure(list(Clades = c("C.T", "A.G", "G.A", "G.C", "T.C",
"C.A", "G.T", "A.T", "T.A", "T.G", "A.C", "C.G", "A.del", "TAT.del",
"TCTGGTTTT.del", "TACATG.del", "AGTTCA.del", "GATTTC.del"), `19A` = c(413,
93, 21, 0, 49, 9, 238, 13, 3, 1, 0, 4, 1, 0, 0, 0, 0, 0), `20A` = c(7929,
1920, 1100, 419, 1025, 124, 3730, 124, 22, 45, 64, 17, 8, 19,
23, 39, 0, 0), `20B` = c(5283, 1447, 2325, 1106, 336, 117, 946,
137, 35, 53, 123, 11, 9, 10, 21, 1, 0, 0), `20E (EU1)` = c(13086,
1927, 650, 1337, 1864, 96, 2967, 243, 69, 92, 115, 1486, 27,
5, 0, 1, 0, 0), `20I (Alpha, V1)` = c(71142, 12966, 12047, 15587,
14935, 15382, 11270, 12211, 5284, 4273, 430, 99, 5674, 4536,
4974, 4592, 0, 0), `20J (Gamma, V3)` = c(2822, 654, 883, 409,
501, 213, 843, 399, 203, 27, 429, 198, 1, 0, 197, 0, 0, 0), `21J (Delta)` = c(166003,
49195, 26713, 1399, 25824, 15644, 95967, 2011, 329, 11034, 716,
21087, 10532, 198, 0, 14, 9809, 10503)), class = "data.frame", row.names = c("C.T",
"A.G", "G.A", "G.C", "T.C", "C.A", "G.T", "A.T", "T.A", "T.G",
"A.C", "C.G", "A.del", "TAT.del", "TCTGGTTTT.del", "TACATG.del",
"AGTTCA.del", "GATTTC.del"))
To add to the previous answer, here is how you can get 7 plots (1 for each Clade, which is how I interpreted the question) using facet_wrap():
df <- df %>%
pivot_longer(-Clades)
ggplot(data = df,
aes(x = Clades,
y = value)) +
geom_bar(aes(fill = Clades),
stat = 'identity') +
facet_wrap(~name, scales = 'free_y') +
theme(axis.text.x = element_blank())
As cazman said in the comments, you need to get your data in long form for it to work with ggplot2 (efficiently).
First, use pivot_longer(), and then use ggplot2:
library(tidyverse)
dat %>%
pivot_longer(-Clades) %>%
ggplot(aes(x=Clades, y=value, fill=name)) +
geom_col()
I am trying to cut a dendrogram into three classes using the rect.hclust function, but when I export the graph, it cuts off the borders at the bottom of the graph. In addition, I would like to increase the weight of the borders, but I am not sure how to do this, as the lwd argument doesn't seem to exist for this function. What can I do to fix these parameters?
Data:
cluster <- data.frame(plot=c(1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20),
meas1 = c(443,
836,
903,
684,
94,
125,
733,
846,
625,
234,
437,
775,
269,
774,
17,
502,
80,
51,
523,
229),
meas2 = c(735,
574,
793,
261,
961,
136,
404,
138,
45,
935,
698,
675,
594,
497,
152,
153,
30,
667,
547,
745),
meas3 = c(23,
526,
36,
93,
708,
970,
399,
111,
456,
439,
569,
503,
337,
213,
399,
850,
614,
491,
28,
452))
Code for hierarchical clustering and dendrogram generation:
#Generate distance matrix
dist_mat <- dist(cluster, method = 'euclidean')
#Hierarchical clustering
hclust_avg <- hclust(dist_mat, method = 'average')
#Cut into 3 classes
cut_avg <- cutree(hclust_avg, k = 3)
#Plot dendrogram
plot(hclust_avg, xlab = "", ylab = "Euclidean Distance", sub = "",
main = "Cluster Dendrogram", lwd = 2)
#Generate borders around each group
rect.hclust(hclust_avg, k=3, border = 2:6)
Exporting the graph looks as follows:
I need the red-green-blue borders to be thicker (higher border weights), and I also need them to not be cut off at the bottom of the graph.
I bypassed the problem of the missing lower edge of the rectangle by setting the figure margins to zero before drawing the rectangle.
The line width of the rectangles can be set by setting par(lwd), e.g. par(lwd=4), as in the example below:
cluster <- data.frame(plot=1:20,
meas1 = c(443,
836,
903,
684,
94,
125,
733,
846,
625,
234,
437,
775,
269,
774,
17,
502,
80,
51,
523,
229),
meas2 = c(735,
574,
793,
261,
961,
136,
404,
138,
45,
935,
698,
675,
594,
497,
152,
153,
30,
667,
547,
745),
meas3 = c(23,
526,
36,
93,
708,
970,
399,
111,
456,
439,
569,
503,
337,
213,
399,
850,
614,
491,
28,
452))
#Generate distance matrix
dist_mat <- dist(cluster, method = 'euclidean')
#Hierarchical clustering
hclust_avg <- hclust(dist_mat, method = 'average')
#Cut into 3 classes
cut_avg <- cutree(hclust_avg, k = 3)
pars <- par()
#Plot dendrogram
plot(hclust_avg, xlab = "", ylab = "Euclidean Distance", sub = "",
main = "Cluster Dendrogram", lwd = 2)
par(lwd=4, mar=c(0,0,0,0))
#Generate borders around each group
rect.hclust(hclust_avg, k=3, border = 2:6)
# reset par
par(lwd=pars$lwd, mar=pars$mar)
Created on 2020-06-30 by the reprex package (v0.3.0)
I am working with a dataset consisting of different plant genotypes, rates of fertilizer applications, and 5 different measurements. I am using ggplot2 to produce multiple bar graphs, and then using the gridExtra package to combine multiple graphs onto a single page. The trouble I am having involves moving and resizing the scale so that there is only one scale for each of my graphs, and I would like to move it to the lower right corner of the graph. The data/code below should better explain what I mean.
Packages/Dataset
#Open packages
library(dplyr)
library(ggplot2)
library(gridExtra)
#Dataset
plantdata <- data.frame(genotype = c(1,
1,
1,
1,
2,
2,
2,
2,
3,
3,
3,
3,
2,
2,
1,
3,
3,
3,
1,
3,
2,
2,
1,
1,
1,
2,
2,
1,
3,
3,
3,
3,
2,
1,
2,
1),
rate=c(1,
2,
3,
4,
1,
2,
3,
4,
1,
2,
3,
4,
2,
4,
1,
1,
3,
2,
3,
4,
1,
3,
4,
2,
2,
1,
3,
3,
4,
1,
3,
2,
4,
1,
2,
4),
measure1=c(958,
309,
750,
43,
20,
868,
905,
674,
64,
151,
677,
144,
803,
485,
707,
881,
684,
222,
399,
507,
4,
690,
831,
574,
104,
238,
378,
897,
63,
154,
582,
641,
750,
855,
194,
55),
measure2=c(359,
728,
180,
614,
241,
989,
117,
101,
95,
156,
227,
355,
597,
50,
636,
912,
149,
862,
897,
601,
176,
7,
182,
214,
453,
569,
124,
113,
969,
781,
713,
613,
800,
334,
435,
748),
measure3=c(639,
304,
891,
317,
869,
901,
723,
267,
837,
923,
171,
991,
107,
309,
733,
705,
819,
807,
346,
447,
891,
957,
359,
323,
846,
944,
400,
548,
327,
48,
677,
624,
564,
854,
658,
343),
measure4=c(805,
24,
624,
675,
261,
437,
601,
129,
733,
172,
746,
586,
142,
243,
103,
779,
612,
870,
84,
881,
850,
456,
255,
52,
228,
492,
556,
66,
670,
682,
736,
178,
568,
501,
229,
500),
measure5=c(667,
105,
565,
724,
238,
861,
299,
13,
171,
759,
755,
557,
739,
228,
870,
595,
793,
790,
572,
590,
365,
974,
550,
766,
441,
265,
245,
909,
150,
88,
473,
245,
340,
378,
998,
121))
Function for standard error of the mean:
sem <- function(x) sd(x)/sqrt(length(x))
Generating the graphs:
#Measurement 1 graph
meas1 <- select(plantdata, genotype, rate, measure1)
#Aggregating data frame
meas1_mean <- aggregate(meas1, by=list(meas1$genotype, meas1$rate), mean)
meas1_sem <- aggregate(meas1, by=list(meas1$genotype, meas1$rate), sem)
g1 <- ggplot(meas1_mean, aes(x=Group.1, y=measure1, fill=factor(Group.2)))+
geom_bar(stat="identity",width=0.6, position="dodge", col="black")+
scale_fill_discrete(name= 'rate', labels=c("1","2","3","4"))+
xlab("Genotype")+ylab("Measurement")+
geom_errorbar(aes(ymin= meas1_mean$measure1, ymax=meas1_mean$measure1+meas1_sem$measure1), width=0.2, position = position_dodge(0.6))+
ggtitle("Plant Measurement 1")+
scale_fill_brewer(palette='PRGn', name= 'rate', labels=c("1","2","3","4"))+
theme(plot.title = element_text(hjust=0.5))
####################################################################
#Measurement 2 graph
meas2 <- select(plantdata, genotype, rate, measure2)
#Aggregating dataframe
meas2_mean <- aggregate(meas2, by=list(meas2$genotype, meas2$rate), mean)
meas2_sem <- aggregate(meas2, by=list(meas2$genotype, meas2$rate), sem)
#Generating graph
g2 <- ggplot(meas2_mean, aes(x=Group.1, y=measure2, fill=factor(Group.2)))+
geom_bar(stat="identity",width=0.6, position="dodge", col="black")+
scale_fill_discrete(name= 'rate', labels=c("1","2","3","4"))+
xlab("Genotype")+ylab("Measurement")+
geom_errorbar(aes(ymin= meas2_mean$measure2, ymax=meas2_mean$measure2+meas2_sem$measure2), width=0.2, position = position_dodge(0.6))+
ggtitle("Plant Measurement 2")+
scale_fill_brewer(palette='PRGn', name= 'rate', labels=c("1","2","3","4"))+
theme(plot.title = element_text(hjust=0.5))
####################################################################
#Measurement 3 graph
meas3 <- select(plantdata, genotype, rate, measure3)
#Aggregating dataframe
meas3_mean <- aggregate(meas3, by=list(meas3$genotype, meas3$rate), mean)
meas3_sem <- aggregate(meas3, by=list(meas3$genotype, meas3$rate), sem)
#Graph
g3 <- ggplot(meas3_mean, aes(x=Group.1, y=measure3, fill=factor(Group.2)))+
geom_bar(stat="identity",width=0.6, position="dodge", col="black")+
scale_fill_discrete(name= 'rate', labels=c("1","2","3","4"))+
xlab("Genotype")+ylab("Measurement")+
geom_errorbar(aes(ymin= meas3_mean$measure3, ymax=meas3_mean$measure3+meas3_sem$measure3), width=0.2, position = position_dodge(0.6))+
ggtitle("Plant Measurement 3")+
scale_fill_brewer(palette='PRGn', name= 'rate', labels=c("1","2","3","4"))+
theme(plot.title = element_text(hjust=0.5))
##############################################################
#Measurement 4 graph
meas4 <- select(plantdata, genotype, rate, measure4)
#Aggregating dataframe
meas4_mean <- aggregate(meas4, by=list(meas4$genotype, meas4$rate), mean)
meas4_sem <- aggregate(meas4, by=list(meas4$genotype, meas4$rate), sem)
#Graph
g4 <- ggplot(meas4_mean, aes(x=Group.1, y=measure4, fill=factor(Group.2)))+
geom_bar(stat="identity",width=0.6, position="dodge", col="black")+
scale_fill_discrete(name= 'rate', labels=c("1","2","3","4"))+
xlab("Genotype")+ylab("Measurement")+
geom_errorbar(aes(ymin= meas4_mean$measure4, ymax=meas4_mean$measure4+meas4_sem$measure4), width=0.2, position = position_dodge(0.6))+
ggtitle("Plant Measurement 4")+
scale_fill_brewer(palette='PRGn', name= 'rate', labels=c("1","2","3","4"))+
theme(plot.title = element_text(hjust=0.5))
################################################################
#Measurement 5 graph
meas5 <- select(plantdata, genotype, rate, measure5)
#Aggregate dataframe
meas5_mean <- aggregate(meas5, by=list(meas5$genotype, meas5$rate), mean)
meas5_sem <- aggregate(meas5, by=list(meas5$genotype, meas5$rate), sem)
#Graph
g5 <- ggplot(meas5_mean, aes(x=Group.1, y=measure5, fill=factor(Group.2)))+
geom_bar(stat="identity",width=0.6, position="dodge", col="black")+
scale_fill_discrete(name= 'rate', labels=c("1","2","3","4"))+
xlab("Genotype")+ylab("Measurement")+
geom_errorbar(aes(ymin= meas5_mean$measure5, ymax=meas5_mean$measure5+meas5_sem$measure5), width=0.2, position = position_dodge(0.6))+
ggtitle("Plant Measurement 5")+
scale_fill_brewer(palette='PRGn', name= 'rate', labels=c("1","2","3","4"))+
theme(plot.title = element_text(hjust=0.5))
Then, I arranged all the graphs onto one page, as follows:
grid.arrange(g1, g2, g3, g4, g5,
nrow=2, ncol=3)
The output:
Because the requirements for this graph are that it needs to be exported in a smaller format, I would like to re-arrange the scale so that there is only one scale for the whole combined image, and I would like to increase the size of it. See below:
How can I do this? If there is a better way to generate this than the way I did it, I am open to learning how to do it.
Consider the ggpubr package and in there the ggarrange function with the argument common.legend = TRUE.
a <- data.frame(x = rnorm(100), y=rnorm(100), group = gl(5,20))
b <- data.frame(x = rnorm(100), y=rnorm(100), group = gl(5,20))
c <- data.frame(x = rnorm(100), y=rnorm(100), group = gl(5,20))
library(ggplot2)
library(ggpubr)
p1 <- ggplot(a, aes(x = x, y = y, color = group)) + geom_point()
p2 <- ggplot(b, aes(x = x, y = y, color = group)) + geom_point()
p3 <- ggplot(c, aes(x = x, y = y, color = group)) + geom_point()
ggarrange(p1, p2, p3, common.legend = TRUE)
Documentation: https://rpkgs.datanovia.com/ggpubr/reference/ggarrange.html
for visualized my data, I used gplot.
Question: Why "colour" doesn't change, and is it possible to do type = "h" like in basic plot?
print(qplot(roundpop, Observation, data=roundpopus), shape = 5, colour = "blue") # i tryed with "" and without.
And if it's possible to change type to histogram, like on second picture, can I draw a line by the top of lines?
Like that:
and maybe to write labels (states) on the top of the lines. Because I know how to give a name only for dots on basic plot.
Thank you!
Here are some options, which you may want to tweak according to your needs:
library(ggplot2)
df <- structure(list(x = c(1, 2, 2, 2, 3, 3, 3, 3, 4, 4, 5, 5, 5, 6,
6, 6, 7, 7, 7, 7, 8, 9, 10, 10, 10, 12, 13, 13, 20, 20, 27, 39
), y = c(33, 124, 45, 294, 160, 105, 276, 178, 377, 506, 176,
393, 247, 378, 221, 796, 503, 162, 801, 486, 268, 575, 828, 493,
252, 495, 836, 551, 413, 832, 1841, 1927), lab = c("i8g8Q", "oXlWk",
"NC2WO", "pYxBL", "Xfsy6", "FJcOl", "Ke98f", "K2mCW", "g4XYi",
"ICzWp", "7nqrK", "dzhlC", "JagAW", "0bObp", "8ljIW", "E8OZR",
"6Tuxz", "3Grbq", "xqsld", "BvuJT", "JXi2N", "eSDYS", "OYVWN",
"vyWzK", "6AKxk", "nCgPx", "8lHrq", "kWAGm", "E08Rd", "cmIYY",
"btoUm", "k6Iek")), .Names = c("x", "y", "lab"), row.names = c(NA,
-32L), class = "data.frame")
p <- ggplot(df, aes(x, y))
gridExtra::grid.arrange(
p + geom_point(),
p + geom_point() + geom_text(aes(label = lab), angle = 60, hjust = 0, size = 2),
p + geom_segment(aes(xend=x, yend=0)),
p + geom_segment(aes(xend=x, yend=0)) + geom_line(color = "red", size = 2) ,
p + geom_segment(aes(xend=x, yend=0)) + geom_smooth(span = .4, se = FALSE, color = "red", size = 2)
)
Im creating a ggplot with geom_vline at a specific location on the x axis. i would like the x axis to show that specific value
Following is my data + code:
dput(agg_data)
structure(list(latency = structure(c(0, 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 24, 26, 28,
29, 32, 36, 37, 40, 43, 46, 47, 48, 49, 54, 64, 71, 72, 75, 87,
88, 89, 93, 134, 151), class = "difftime", units = "days"), count = c(362,
11, 8, 5, 4, 2, 8, 6, 4, 2, 2, 1, 5, 1, 2, 2, 2, 1, 1, 1, 2,
1, 2, 1, 1, 1, 1, 1, 1, 1, 2, 2, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1,
1, 1, 1), cum_sum = c(362, 373, 381, 386, 390, 392, 400, 406,
410, 412, 414, 415, 420, 421, 423, 425, 427, 428, 429, 430, 432,
433, 435, 436, 437, 438, 439, 440, 441, 442, 444, 446, 447, 448,
449, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460), cum_dist = c(0.78695652173913,
0.810869565217391, 0.828260869565217, 0.839130434782609, 0.847826086956522,
0.852173913043478, 0.869565217391304, 0.882608695652174, 0.891304347826087,
0.895652173913044, 0.9, 0.902173913043478, 0.91304347826087,
0.915217391304348, 0.919565217391304, 0.923913043478261, 0.928260869565217,
0.930434782608696, 0.932608695652174, 0.934782608695652, 0.939130434782609,
0.941304347826087, 0.945652173913043, 0.947826086956522, 0.95,
0.952173913043478, 0.954347826086957, 0.956521739130435, 0.958695652173913,
0.960869565217391, 0.965217391304348, 0.969565217391304, 0.971739130434783,
0.973913043478261, 0.976086956521739, 0.980434782608696, 0.982608695652174,
0.984782608695652, 0.98695652173913, 0.989130434782609, 0.991304347826087,
0.993478260869565, 0.995652173913044, 0.997826086956522, 1)), .Names = c("latency",
"count", "cum_sum", "cum_dist"), row.names = c(NA, -45L), class = "data.frame")
and code:
library(ggplot2)
library(ggthemes)
russ<-ggplot(data=agg_data,aes(x=as.numeric(latency),y=cum_dist))+geom_line(size=2)
russ<-russ+ggtitle("Latency from first click to Qualified Demo:") + xlab("in Days")+ ylab("Proportion of maturity")+theme_economist()
russ<-russ+geom_vline(aes(xintercept=10), color="black", linetype="dashed")
russ
which creates the following plot:
I want the plot show the value '10' (same location as the vline) on the x-axis
I looked for some other similar answers, like in Customize axis labels
but this one re creates the x axis labels (with scale_x_discrete), and does not add a new number to the current scale, which is more of what im looking for.
thanks in advance!
In your case x scale is continuous, so you can use function scale_x_continuous() and provide breaks at positions you need.
russ + scale_x_continuous(breaks=c(0,10,50,100,150))