Genomic Data Visualization using R
Jubayer Hossain
23 February, 2021
What is ggplot2?
- ggplot2 is a data visualization package for the statistical programming language R.
- Created by Hadley Wickham in 2005, ggplot2 is an implementation of Leland Wilkinson’s Grammar of Graphics—a general scheme for data visualization which breaks up graphs into semantic components such as scales and layers.
Types of Visualization
In statistics, we generally have two kinds of visualization:
- Exploratory data visualization: Exploring the data visually to find patterns among the data entities.
- Explanatory data visualization: Showcasing the identified patterns using simple graphs.
Why Visualization?
“A picture is worth a thousand words”
- Data visualizations make big and small data easier for the human brain to understand, and visualization also makes it easier to detect patterns, trends, and outliers in groups of data.
- Good data visualizations should place meaning into complicated datasets so that their message is clear and concise.
Grammar of Graphics
- Data - The dataset being plotted
- Aesthetics- The scales onto which we plot our data
- Geometry - The visual elements used for our data
- Facet -Groups by which we divide the data
Load Packages and Set Theme
library(tidyverse)
library(gridExtra)
theme_set(theme_gray(base_size = 18))Load Data
data <- read.csv("variants_from_assembly.bed", sep="\t", stringsAsFactors = TRUE, header = FALSE, quote = '')Exploring Data
# examine first few rows
head(data) V1 V2 V3 V4 V5 V6 V7 V8 V9
1 6 103832058 103832059 SV1 185 + Insertion 0 185
2 6 102958468 102958469 SV2 317 + Insertion -14 303
3 6 102741692 102741693 SV3 130 + Deletion 130 0
4 6 102283759 102283760 SV4 1271 + Insertion -12 1259
5 6 101194032 101194033 SV5 2864 + Insertion -13 2851
6 6 101056644 101056645 SV6 265 + Insertion 0 265# Colnames
names(data) <- c("chrom", "start", "stop", "name", "size", "strand", "type", "ref.dist", "query.dist")head(data) chrom start stop name size strand type ref.dist query.dist
1 6 103832058 103832059 SV1 185 + Insertion 0 185
2 6 102958468 102958469 SV2 317 + Insertion -14 303
3 6 102741692 102741693 SV3 130 + Deletion 130 0
4 6 102283759 102283760 SV4 1271 + Insertion -12 1259
5 6 101194032 101194033 SV5 2864 + Insertion -13 2851
6 6 101056644 101056645 SV6 265 + Insertion 0 265# Data Structures
glimpse(data)Rows: 9,556
Columns: 9
$ chrom <fct> 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6…
$ start <int> 103832058, 102958468, 102741692, 102283759, 101194032, 101…
$ stop <int> 103832059, 102958469, 102741693, 102283760, 101194033, 101…
$ name <fct> SV1, SV2, SV3, SV4, SV5, SV6, SV7, SV8, SV9, SV10, SV11, S…
$ size <int> 185, 317, 130, 1271, 2864, 265, 334, 248, 362, 31, 499, 16…
$ strand <fct> +, +, +, +, +, +, +, +, +, +, +, +, +, +, +, +, +, +, +, +…
$ type <fct> Insertion, Insertion, Deletion, Insertion, Insertion, Inse…
$ ref.dist <int> 0, -14, 130, -12, -13, 0, 0, 8915, 0, 1939, -6, -1141, -2,…
$ query.dist <int> 185, 303, 0, 1259, 2851, 265, 334, 8667, 362, 1908, 493, 4…# Summary
summary(data) chrom start stop name
1 : 779 Min. : 55378 Min. : 55379 SV1 : 1
2 : 702 1st Qu.: 28750831 1st Qu.: 28750832 SV10 : 1
7 : 596 Median : 67213567 Median : 67213568 SV100 : 1
6 : 568 Mean : 77646932 Mean : 77646933 SV1000 : 1
3 : 565 3rd Qu.:116282726 3rd Qu.:116282726 SV10000: 1
4 : 530 Max. :249205989 Max. :249205990 SV10001: 1
(Other):5816 (Other):9550
size strand type ref.dist
Min. : 1.0 +:9556 Contraction:1677 Min. :-11736
1st Qu.: 136.0 Deletion :2257 1st Qu.: -9
Median : 307.0 Expansion :2479 Median : 3
Mean : 681.0 Insertion :3143 Mean : 1077
3rd Qu.: 580.2 3rd Qu.: 348
Max. :9951.0 Max. :157902
query.dist
Min. :-18243.0
1st Qu.: 0.0
Median : 161.0
Mean : 1236.4
3rd Qu.: 758.2
Max. :163991.0
Bar Chart
ggplot(data, aes(x=chrom))+
geom_bar()
# Filtering Data
data <- filter(data, chrom %in% c(seq(1,22), "X", "Y", "MT"))
ggplot(data, aes(x=chrom))+
geom_bar()
# Ordering Chromosomes
data$chrom <- factor(data$chrom, levels=c(seq(1,22),"X","Y","MT"))
ggplot(data, aes(x=chrom))+
geom_bar()
# Xlab, Ylab, Title
ggplot(data, aes(x=chrom))+
geom_bar()+
labs(title = "Chromosomes Distribution",
x="Chromosomes",
y= "Counts")
# Centering title
ggplot(data, aes(x=chrom))+
geom_bar()+
labs(title = "Chromosomes Distribution",
x="Chromosomes",
y= "Counts")+
theme(plot.title = element_text(hjust = 0.5))
# Color by categorical variable
ggplot(data, aes(x=chrom, fill=type))+
geom_bar()+
labs(title = "Chromosomes Distributions",
x="Chromosomes",
y= "Counts")+
theme(plot.title = element_text(hjust = 0.5))
# Ordering type
data$type <- factor(data$type, levels = c("Insertion","Deletion", "Expansion","Contraction"))# Color by categorical variable
ggplot(data, aes(x=chrom, fill=type))+
geom_bar()+
labs(title = "Chromosomes Distributions",
x="Chromosomes",
y= "Counts")+
theme(plot.title = element_text(hjust = 0.5))
# Color by categorical variable
ggplot(data, aes(x=chrom, fill=type))+
geom_bar()+
labs(title = "Chromosomes Distributions",
x="Chromosomes",
y= "Counts")+
theme(plot.title = element_text(hjust = 0.5))+
guides(fill=guide_legend(title = "Type"))
fig <- function(width, heigth){
options(repr.plot.width = width, repr.plot.height = heigth)
}# Facet Wrap
ggplot(data, aes(x=chrom, fill=type))+
geom_bar()+
labs(title = "Chromosomes Distributions",
x="Chromosomes",
y= "Counts")+
guides(fill=guide_legend(title = "Type"))+
facet_wrap(~type)+
fig(20,6)
# Facet Grid
ggplot(data, aes(x=chrom, fill=type))+
geom_bar()+
labs(title = "Chromosomes Distributions",
x="Chromosomes",
y= "Counts")+
guides(fill=guide_legend(title = "Type"))+
facet_grid(type ~ .)+
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5))
Scatter Plot
ggplot(data, aes(x=ref.dist, y=query.dist))+
geom_point()
ggplot(data, aes(x=ref.dist, y=query.dist, color=type))+
geom_point()+
xlim(-500, 500)+
ylim(-500, 500)Warning: Removed 3920 rows containing missing values (geom_point).
Boxplot
ggplot(data, aes(x=type, y=size))+
geom_boxplot()
ggplot(data, aes(x=type, y=size))+
geom_boxplot()+
coord_flip()
Violin Plot
ggplot(data, aes(x=type, y=size))+
geom_violin()
ggplot(data, aes(x=type, y=size))+
geom_violin()+
coord_flip()
ggplot(data, aes(x=type, y=size))+
geom_violin(adjust=0.2)+
scale_y_log10()+
coord_flip()
Desnsity Plot
ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)Warning: Removed 2660 rows containing non-finite values (stat_density).
ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)+
facet_wrap(~type)Warning: Removed 2660 rows containing non-finite values (stat_density).
ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)+
facet_grid(type ~ .)Warning: Removed 2660 rows containing non-finite values (stat_density).
ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)+
facet_grid(. ~type)Warning: Removed 2660 rows containing non-finite values (stat_density).
ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)+
facet_grid(type ~chrom)Warning: Removed 2660 rows containing non-finite values (stat_density).
Pie Chart
type_counts <- summary(data$type)
pie(type_counts)
Arrange Plot
f1 <- ggplot(data, aes(x=chrom,fill=type))+
geom_bar()+
labs(title = "Distribution of Chromosomes",
x="Chromosomes", y="Counts",
tag = "A")+
guides(fill=guide_legend(title = "Type"))
f2 <- ggplot(data, aes(x=size, fill=type))+
geom_density(alpha=0.5)+
xlim(0,500)+
labs(title = "Density Plot of Size",
x = "Size",
y = "Probability",
tag = "B")+
guides(fill=guide_legend(title = "Type"))
grid.arrange(f1, f2, nrow=2, ncol=1)Warning: Removed 2660 rows containing non-finite values (stat_density).