The goal of plotGenStruct is to assist with plotting clustered genetic STRUCTURE datasets using the ggplot2 package. Data is automatically sorted into clusters and several palettes can be easily chosen.
At present the package can only be installed from github with:
# install.packages("devtools")
devtools::install_github("aj2duncan/plotGenStruct")Existing STRUCTURE data can be plotted using the function
struc_plot(). If you don’t have any data to hand then some random data
can be generated using the create_sruc_data() function
This is a basic example which shows you how to solve a common problem:
library(plotGenStruct)
create_struc_data()
# A tibble: 5,000 x 3
Individual Cluster q
<chr> <int> <dbl>
1 Ex101 1 0.063
2 Ex101 2 0.014
3 Ex101 3 0.0048
4 Ex101 4 0.016
5 Ex101 5 0.183
6 Ex101 6 0.022
7 Ex101 7 0.007
8 Ex101 8 0.006
9 Ex101 9 0.008
10 Ex101 10 0.676
# ... with 4,990 more rowsTo see what the reordered data looks like we can use struc_reorder().
struc_reorder(create_struc_data())
# A tibble: 5,000 x 4
Individual Cluster q main_cluster
<chr> <int> <dbl> <int>
1 Ex154 1 0.999 1
2 Ex219 1 0.997 1
3 Ex403 1 0.997 1
4 Ex537 1 0.987 1
5 Ex581 1 0.987 1
6 Ex394 1 0.986 1
7 Ex316 1 0.980 1
8 Ex234 1 0.957 1
9 Ex184 1 0.955 1
10 Ex152 1 0.953 1
# ... with 4,990 more rowsFinally we can plot the data.
struc_plot(create_struc_data())
You can also specify different palettes which is just a short cut to
several of the scale_fill_*() functions from
ggplot2.
struc_plot(create_struc_data(), fill_pal = "viridis")
This includes a custom palette. Here we also specify only two clusters.
# custom palette
struc_plot(create_struc_data(num_clusters = 2),
fill_pal = "custom", values = c("blue", "red"))
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent