Heatmap visualization in Rstudio _ high resoultion figure

Hello,

Heatmap is one of the most frequently visualized method on SCI papers.

This is the method how to heatmap visualize in Rstudio.  

 

 

 

 

 

 

1. Install Packages

 

# Install Library 

install.packages("stats")

install.packages("gplots")

install.packages("RColorBrewer")

 

 

 

 

 

 

2. Run Packages

 

# Loads Library 

library(stats)

library(gplots)

library(RColorBrewer)

 

 

 

 

 

 

 

3. Set the Heatmap color

 

# Heatmap color 

scalegreenbluered=colorRampPalette(colors=c("white","chartreuse4","darkgreen","black"))(256)

colors=c("white", "chartreuse4", "darkgreen", "black")

My case, i selected white, green, darkgreen, and black color.

Fill in the low value from left to high value at right.

 

(256)

The number "(256)" is higheset value. This means how many times will you divide colors. 

See the below divided colors from 10 to 256.

 

Now, i belive you understood until here. 

 

 

 

 

 

 

 

 

4. Input Data

 

4.1 Check data format

 

This is the data format below. 

You can check column and row data format by run below command.

 

# Data forma 

B = matrix(

  

  # Taking sequence of elements

  c("bacteria1", "5", "43", 4, 5, "bacteria2", "43", "21", 9, 10, "bacteria3", "53", "32", 14, 15, "bacteria4", "51", "21", 19, 20, "bacteria5", "21", "21", 24, 25),

  

  # No of rows

  nrow = 5,

  

  # No of columns

  ncol = 5,

  

  # By default matrices are in column-wise order

  # So this parameter decides how to arrange the matrix

  byrow = TRUE

)

 

# Naming rows

colnames(B) = c("variable1", "variable1", "variable1", "variable2", "variable2")

 

# Naming columns

 

print(B)

The result of data format is in below.

 

    variable1   variable1 variable1 variable2 variable2

[1,] "bacteria1" "5"       "43"      "4"       "5"      

[2,] "bacteria2" "43"      "21"      "9"       "10"     

[3,] "bacteria3" "53"      "32"      "14"      "15"     

[4,] "bacteria4" "51"      "21"      "19"      "20"     

[5,] "bacteria5" "21"      "21"      "24"      "25"   

 

 

 

4.2 Input data and normalization

 

 

# Heatmap Calculation 

dat1 <- read.delim("clipboard", row.names=1) #Data input

 

asinTransform <- function(dat1) { asin(sqrt(dat1)) }

pAsin <- asinTransform(dat1) #Normalization

 

orderheat <-as.matrix(pAsin)

dat1 <- read.delim("clipboard", row.names=1) 

This command will input yourdata from copied dataframe from excel.

 

When you visualize data values, sometimes normalization is necessary to well understand the differentiation.

I used to Arcsin normalization method.

 

Left is normalized and right is not normalized below.

This is same data, but you can notice different pattern can color expressions.

 

 

 

 

 

5. Calculate Row and Column dendrogram

 

 

# Calculation 

distance.row = dist(as.matrix(orderheat), method = "euclidean") #Calculation

cluster.row = hclust(distance.row, method = "ward.D")#Calculation

distance.col = dist(t(as.matrix(orderheat)), method = "euclidean") #Calculation

cluster.col = hclust(distance.col, method = "ward.D") #Calculation

 

 

 

 

 

 

6. Visualization Plotting

 

6.1 Only Row dendrogram

 

# Row dendrogram Plot 

heatmap.2(orderheat, dendrogram="row", trace="none", keysize=1, tracecol = "#303030", cexCol = 0.5, cexRow = 0.7, col=scalegreenbluered, scale="none", margins=c(4,10), sepcolor="white", colsep=c(7, 14, 21, 28, 41, 54), sepwidth=c(0.2), key.title='Heatmap scale', Colv=FALSE)

dendrogram = 

You can select here ther dendrogram patterns; only row or both.

 

cexCol = 

Change text size of column

 

cexRow = 

Change text size of row

 

margins = 

Change graph size

 

sepcolor = 

Set division line color of column

 

colsep = 

Set division line location of column

 

sepwidth = 

Change division line size

 

 

Below is example result 

(The label of Column and Row is erased by default, because of SCI data that published yet).

 

 

 

 

 

6.2 Column and Row dendrogram

 

# Column & Row dendrogram plot

heatmap.2(orderheat, dendrogram="both", trace="none", keysize=1, Rowv=as.dendrogram(cluster.row), Colv=as.dendrogram(cluster.col), tracecol = "#303030", cexCol = 0.35, col=scalegreenbluered, scale="none", margins=c(4,6), key.title='Heatmap scale')

Command explain skip. Same with above.

 

Below is example result.

 

 

 

 

 

7. Create High resolution figure

 

7.1 Set working directory

# Set working directory 

getwd()

setwd("F:/SRA/Output_Figure")

getwd()

 

 

 

 

7.2 Create High resolution figure Output

# Get Figures with high resolution 

tiff(file="heatmap.2.tiff", width = 10, height = 6, units = "in", res = 600)

heatmap.2(orderheat, dendrogram="row", trace="none", keysize=1, tracecol = "#303030", cexCol = 0.5, cexRow = 0.7, col=scalegreenbluered, scale="none", margins=c(4,10), sepcolor="white", colsep=c(7, 14, 21, 28, 41, 54), sepwidth=c(0.2), key.title='Heatmap scale', Colv=FALSE)

dev.off()