IS9: Advanced data visualization using GGplot2 library
Aqsa Khalid
modified by Najeeha Talat Iqbal on 23-04-2026
Loading and Cleaning Data After we learn how to load a dataset, we need to also create an object which might contain multiple types of data - categorical, numerical, strings, etc. Before we can visualize this data, we have to be able to separate these types of data into separate types, assigning them into the right variables. 1. Load data 2. Create object for the dataset 3. Check for data being loaded and type of data 4. Prepare a data frame or a matrix of data we will analyze Load data:
#First, we will import our data and store it an an object or data frame
df <- read.table('gene-expression_dataset.txt', sep ='\t', header = TRUE, row.names=1)Check whether dataset get loaded or not; we can do it by checking the dimensions of data (with the dim() function) or with the head() function
#Write the first 10 rows of the dataset:
head(df)#Check dimension of data (how many rows and columns)
dim(df)## [1] 200 41Indicating we have 200 rows and 41 columns in our data.
#Check type of data str(df)
str(df)## 'data.frame': 200 obs. of 41 variables:
## $ case_1 : num 1.098 0.66 0.432 0.624 1.551 ...
## $ case_2 : num 5.7501 2.695 0.0623 9.8672 0.2589 ...
## $ case_3 : num 2.74 4.34 4.78 1.18 1.11 ...
## $ case_4 : num 0.925 0.844 0.304 0.01 0.255 ...
## $ case_5 : num 0.231 0.8823 0.0312 0.35 0.4966 ...
## $ case_6 : num 0.514 1.142 0.01 3.03 0.551 ...
## $ case_7 : num 1.96 5.505 3.372 3.956 0.393 ...
## $ case_8 : num 1.88 3.13 1.21 1.86 1.8 ...
## $ case_9 : num 1.0633 0.8273 0.0864 0.3832 0.6816 ...
## $ case_10 : num 0.3845 1.2559 0.0384 1.6795 0.5199 ...
## $ case_11 : num 2.383 5.314 3.712 4.845 0.811 ...
## $ case_12 : num 1.267 2.727 1.305 0.413 1.52 ...
## $ case_13 : num 0.564 2.214 0.45 0.879 0.896 ...
## $ case_14 : num 4.381 4.958 0.292 1.01 0.586 ...
## $ case_15 : num 0.792 1.779 0.438 2.354 1.369 ...
## $ case_16 : num 1.249 2.258 0.72 2.099 0.998 ...
## $ case_17 : num 1.24 2.25 5.5 1.72 0.98 ...
## $ control_1 : num 0.169 1.354 0.108 0.488 0.289 ...
## $ control_2 : num 0.287 0.979 0.01 0.767 0.449 ...
## $ control_3 : num 0.154 0.928 0.651 0.63 0.4 ...
## $ control_4 : num 0.199 0.99 0.122 0.561 0.193 ...
## $ control_5 : num 0.0974 0.532 0.01 0.01 0.5647 ...
## $ control_6 : num 0.01 0.6776 0.01 0.3753 0.0933 ...
## $ control_7 : num 0.479 1.588 0.1 0.954 0.356 ...
## $ control_8 : num 0.0279 0.6418 0.1714 1.2309 0.659 ...
## $ control_9 : num 0.207 0.933 0.117 0.01 0.641 ...
## $ control_10: num 0.592 0.691 0.561 0.768 0.167 ...
## $ control_11: num 0.0971 0.5196 0.1487 1.6399 0.4261 ...
## $ control_12: num 0.0522 0.1815 0.0353 2.2553 0.2997 ...
## $ control_13: num 0.0318 0.5115 0.0208 0.2628 0.69 ...
## $ control_14: num 0.234 0.615 0.177 0.01 0.638 ...
## $ control_15: num 1.112 0.48 0.114 0.281 0.15 ...
## $ control_16: num 0.0265 0.7368 0.0337 0.01 0.283 ...
## $ control_17: num 0.0286 0.7881 0.1215 5.0598 0.2015 ...
## $ control_18: num 0.24 0.545 0.115 0.01 0.187 ...
## $ control_19: num 0.227 0.8761 0.0692 0.01 0.4001 ...
## $ control_20: num 0.567 2.335 0.321 1.332 0.233 ...
## $ control_21: num 0.789 2.946 0.289 0.659 0.57 ...
## $ control_22: num 0.515 1.8517 0.0428 0.6374 0.5628 ...
## $ control_23: num 0.4241 2.0924 0.0507 0.01 0.6483 ...
## $ control_24: num 0.406 0.854 0.154 0.17 0.345 ...We can also extract the samples names (header, which are present in columns) from our data:
#Extract the sample names (column names)
Samples <- colnames(df)
print(Samples)## [1] "case_1" "case_2" "case_3" "case_4" "case_5"
## [6] "case_6" "case_7" "case_8" "case_9" "case_10"
## [11] "case_11" "case_12" "case_13" "case_14" "case_15"
## [16] "case_16" "case_17" "control_1" "control_2" "control_3"
## [21] "control_4" "control_5" "control_6" "control_7" "control_8"
## [26] "control_9" "control_10" "control_11" "control_12" "control_13"
## [31] "control_14" "control_15" "control_16" "control_17" "control_18"
## [36] "control_19" "control_20" "control_21" "control_22" "control_23"
## [41] "control_24"Extract feature names (gene ids)
#We can extract row ids from data
Genes <- rownames(df)
print(Genes)## [1] "FGFBP1" "PCSK9" "CHI3L1" "RP11-1035H13.3"
## [5] "TNIP3" "SLC6A14" "FCGR3B" "RNF182"
## [9] "CXCL3" "CCR10" "IFNG" "JPH1"
## [13] "TNFRSF9" "SAA2" "LRP8" "SAA1"
## [17] "KCNN3" "ACPP" "MYEOV" "NEUROG3"
## [21] "CEACAM7" "ERCC6L" "MCM10" "ALDH1A2"
## [25] "GSG2" "ALDH4A1" "FAM72D" "SLC9A7"
## [29] "VWA2" "AGT" "AUNIP" "HLA-DOB"
## [33] "HGF" "FCRL5" "FAM72B" "CENPI"
## [37] "ASPHD2" "PTPRO" "TMED7-TICAM2" "LAIR2"
## [41] "PDPN" "GINS4" "CLSPN" "F2RL2"
## [45] "ARNTL2" "DUSP4" "SPAG4" "LRG1"
## [49] "FCGR1A" "FIBIN" "C4BPB" "TIGIT"
## [53] "BMP7" "FAM83D" "CLIC6" "HSPA4L"
## [57] "SKA1" "MMP3" "DTL" "SPA17"
## [61] "KIF4A" "MMP12" "PARPBP" "TRIP13"
## [65] "CLCN2" "GPT2" "CENPA" "HK2"
## [69] "MYL4" "PSAT1" "FGF7" "PHLDA2"
## [73] "GZMB" "GALNT5" "RPL39L" "NCAPH"
## [77] "NETO2" "BUB1" "PRR11" "DKK4"
## [81] "MELK" "KIF2C" "HMMR" "DLGAP5"
## [85] "E2F5" "GBP5" "CHRNA5" "TNFSF13B"
## [89] "CEP55" "PBK" "TCHP" "CXCL11"
## [93] "LPL" "KIAA0101" "FCGR3A" "GGH"
## [97] "CENPW" "CCL13" "MAD2L1" "CDKN3"
## [101] "SLC6A13" "HEATR4" "PRCD" "FTCD"
## [105] "FAM129C" "CLRN1" "KLKB1" "ABCA10"
## [109] "SEC31B" "ADAMTS10" "BMP8A" "L3MBTL1"
## [113] "RNF207" "HPX" "SYT8" "TSSK6"
## [117] "OXT" "CES4A" "BEGAIN" "WNT9B"
## [121] "SPATA25" "ZBTB16" "APOH" "LCN12"
## [125] "GNRH1" "LRRC29" "PROCA1" "PPP1R32"
## [129] "SLC25A27" "LMNTD2" "AIFM3" "VPS37D"
## [133] "HSF4" "CEACAM19" "PTH2R" "NKD2"
## [137] "MKRN3" "ANGPTL6" "REG3G" "RASSF10"
## [141] "AGBL2" "SYCP2" "CYP3A7" "NEURL2"
## [145] "NPIPB11" "CNGA1" "PCSK4" "CDH16"
## [149] "SH2D6" "GABBR1" "CBS" "BCO2"
## [153] "CHST13" "ZNF257" "KCNK15" "CNTFR"
## [157] "POU5F1" "KLHDC1" "ABCA3" "ZNF354B"
## [161] "AGER" "ZNF334" "VSTM2A" "DRD4"
## [165] "BTNL9" "C17orf67" "ADHFE1" "CLDND2"
## [169] "VSIG10L" "ASPG" "C8A" "CRIP3"
## [173] "TMEM74B" "KLRC1" "TGFBR3L" "IGF2BP1"
## [177] "KLK13" "CD207" "KISS1" "RHBDL1"
## [181] "RP11-347C12.3" "ANKAR" "CDK3" "LIN7B"
## [185] "MT1M" "ENHO" "LSMEM1" "EGFL8"
## [189] "ZNF418" "TXK" "SH2D1B" "LBX2"
## [193] "CD40LG" "DEFB131" "ZNF415" "CCDC152"
## [197] "CATSPERG" "KCNG1" "AC017081.1" "SLC34A3"To clean the data, we can remove the missing values by using the na.omit() or na.exclude() functions.
#Remove NA values
df1 <- na.exclude(df)
#To check the dimension of data
dim(df1)## [1] 200 41This output indicates that we are left with 200 rows and 41 columns in our data and no NAs are present, hence the dimension will remain same. Descriptive Statistics and Visualization Descriptive statistics helps us understand statistical properties of the full dataset so that we can spot unwanted variation and decide whether we have to transform the data scaling before analysis or visualization:
Box Plot Now let’s visualize the statistical properties of each sample in the dataset using a box plot (also called box and whisker plot)
#Draw boxplot for all samples (for FPKM values)
boxplot(df1) 
#Add the main title, axis titles, and color
boxplot(df1, main="Boxplot for Case and control data", xlab="", ylab="Gene expression", col="red", las=2, cex.axis = 0.9)
Note: To make the box plot visually informative, we can add the title main=””, colors and to make sure the axis labels are fully visible. To do so, we need to position the axis label vertically “las=2”. To change the orientation back to horizontal, use las=1 or omit this parameter altogether. To define the font size of labels, cex.axis is used. This box plot represents the descriptive statistics of various samples in our dataset visually. X-axis indicates the sample ID and the Y-axis represents the values (gene expression level). The dots in the plot represent the outliers (which are either above or below the maximum and minimum gene expression values for the sample). From the above box plot, it is difficult to interpret the differences between key statistical properties of each sample. Most of the data we can see are outliers. This can be resolved by transforming the data to a more “normal” distribution by changing the scale of the Y axis. One of such methods is log-scale transformation. Before transforming values to the log scale, it is also important to add 1 to each value. Otherwise, there will be a non-numeric log value for the original zero value present in data.
Perform log scale transformation of the data to normalize data and remove outliers
log_df <- log(df1+1)
#Check the dimension of data
dim(log_df)## [1] 200 41Compute descriptive statistics of data:
#Let’s check the summary statistics of data after log transformation
summ_log <- summary(log_df)
#Print summary statistics of data
summ_log## case_1 case_2 case_3 case_4
## Min. :0.1021 Min. :0.00995 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.7012 1st Qu.:0.48448 1st Qu.:0.31289 1st Qu.:0.52283
## Median :1.0566 Median :0.74213 Median :0.76419 Median :0.77402
## Mean :1.1632 Mean :0.93618 Mean :1.08045 Mean :0.95483
## 3rd Qu.:1.4565 3rd Qu.:1.22682 3rd Qu.:1.68880 3rd Qu.:1.12247
## Max. :3.0866 Max. :3.27927 Max. :3.98563 Max. :3.24872
## case_5 case_6 case_7 case_8
## Min. :0.00995 Min. :0.00995 Min. :0.00995 Min. :0.0215
## 1st Qu.:0.36055 1st Qu.:0.45252 1st Qu.:0.26387 1st Qu.:0.3190
## Median :0.62723 Median :0.66700 Median :0.68136 Median :0.7668
## Mean :0.77187 Mean :0.83973 Mean :0.98644 Mean :1.0154
## 3rd Qu.:0.97017 3rd Qu.:1.18222 3rd Qu.:1.47293 3rd Qu.:1.5335
## Max. :2.98263 Max. :2.66822 Max. :3.75759 Max. :3.7431
## case_9 case_10 case_11 case_12
## Min. :0.00995 Min. :0.00995 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.37879 1st Qu.:0.39738 1st Qu.:0.39875 1st Qu.:0.41088
## Median :0.62046 Median :0.66798 Median :0.78239 Median :0.82234
## Mean :0.87305 Mean :0.78991 Mean :1.04767 Mean :1.08024
## 3rd Qu.:1.23354 3rd Qu.:1.07485 3rd Qu.:1.60510 3rd Qu.:1.48472
## Max. :3.16577 Max. :2.84819 Max. :3.47888 Max. :3.89966
## case_13 case_14 case_15 case_16
## Min. :0.0188 Min. :0.009961 Min. :0.06568 Min. :0.0425
## 1st Qu.:0.3497 1st Qu.:0.494055 1st Qu.:0.40246 1st Qu.:0.4960
## Median :0.6555 Median :0.810678 Median :0.68710 Median :0.7681
## Mean :0.9703 Mean :1.038355 Mean :0.98115 Mean :1.0202
## 3rd Qu.:1.3945 3rd Qu.:1.386104 3rd Qu.:1.34705 3rd Qu.:1.3764
## Max. :3.6326 Max. :3.484378 Max. :3.31386 Max. :3.0589
## case_17 control_1 control_2 control_3
## Min. :0.00995 Min. :0.00995 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.23893 1st Qu.:0.76040 1st Qu.:0.60700 1st Qu.:0.62365
## Median :0.64196 Median :1.00295 Median :0.91851 Median :0.93882
## Mean :0.99434 Mean :1.14146 Mean :0.98907 Mean :1.02624
## 3rd Qu.:1.47370 3rd Qu.:1.48400 3rd Qu.:1.26720 3rd Qu.:1.32297
## Max. :4.07530 Max. :2.99904 Max. :3.22713 Max. :3.13754
## control_4 control_5 control_6 control_7
## Min. :0.04869 Min. :0.00995 Min. :0.00995 Min. :0.02733
## 1st Qu.:0.67781 1st Qu.:0.68902 1st Qu.:0.60025 1st Qu.:0.60708
## Median :0.94026 Median :0.96610 Median :0.86881 Median :0.92000
## Mean :1.01200 Mean :1.04576 Mean :0.97668 Mean :0.98905
## 3rd Qu.:1.32909 3rd Qu.:1.36720 3rd Qu.:1.34743 3rd Qu.:1.31723
## Max. :2.79492 Max. :2.96105 Max. :2.72988 Max. :2.63393
## control_8 control_9 control_10 control_11
## Min. :0.01527 Min. :0.00995 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.66030 1st Qu.:0.69109 1st Qu.:0.60079 1st Qu.:0.63055
## Median :1.03166 Median :1.03359 Median :0.93264 Median :0.94185
## Mean :1.04745 Mean :1.10851 Mean :1.02744 Mean :1.00481
## 3rd Qu.:1.35648 3rd Qu.:1.46385 3rd Qu.:1.37440 3rd Qu.:1.37322
## Max. :2.87036 Max. :2.84382 Max. :3.01716 Max. :2.71023
## control_12 control_13 control_14 control_15
## Min. :0.00995 Min. :0.02056 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.33687 1st Qu.:0.51336 1st Qu.:0.44277 1st Qu.:0.36141
## Median :0.63271 Median :0.76820 Median :0.67664 Median :0.63005
## Mean :0.72857 Mean :0.92460 Mean :0.79166 Mean :0.80805
## 3rd Qu.:1.06069 3rd Qu.:1.22473 3rd Qu.:1.01207 3rd Qu.:1.02713
## Max. :2.36610 Max. :2.51067 Max. :2.65814 Max. :3.03619
## control_16 control_17 control_18 control_19
## Min. :0.00995 Min. :0.00995 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.47752 1st Qu.:0.45143 1st Qu.:0.38516 1st Qu.:0.51261
## Median :0.77803 Median :0.72242 Median :0.67966 Median :0.80335
## Mean :0.87858 Mean :0.85490 Mean :0.79467 Mean :0.91087
## 3rd Qu.:1.15452 3rd Qu.:1.13024 3rd Qu.:1.07376 3rd Qu.:1.14325
## Max. :2.89364 Max. :2.85765 Max. :2.61133 Max. :2.97833
## control_20 control_21 control_22 control_23
## Min. :0.00995 Min. :0.05922 Min. :0.00995 Min. :0.00995
## 1st Qu.:0.50972 1st Qu.:0.70841 1st Qu.:0.58280 1st Qu.:0.63435
## Median :0.79130 Median :0.96650 Median :0.80122 Median :0.93775
## Mean :0.93261 Mean :1.07358 Mean :0.96453 Mean :1.06107
## 3rd Qu.:1.10132 3rd Qu.:1.34238 3rd Qu.:1.21742 3rd Qu.:1.33757
## Max. :3.15393 Max. :3.04136 Max. :3.00967 Max. :3.21967
## control_24
## Min. :0.0391
## 1st Qu.:0.5607
## Median :0.8004
## Mean :0.9943
## 3rd Qu.:1.2219
## Max. :3.4710#Write summary statistics to a file and export it
write.table(summ_log, file="stat_sum-log-data.txt", col.names=TRUE, sep="\t")Draw Box plot for data
# Draw boxplot for all samples
boxplot(log_df, main="Boxplot for log-transformed Data", xlab="", ylab="Gene expression (log[FPKM])", col="red", las=2, cex.axis = 0.9)
Draw box plot for groups of samples (case only)
Group1 <- log_df[,1:17]Or select group using below command of grep
Group1 <- log_df[grep('^case', names(log_df))]Extract the specific samples from the data with their name
#Extract specific samples from data with their name
case <- as.data.frame(log_df[grep('^case', names(log_df))])
cont <- as.data.frame(log_df[grep('^control', names(log_df))])Compute the mean of rows of data
#Compute the mean for each gene within each group
case <- rowMeans(case)
cont <- rowMeans(cont)Bind groups together in the single data frame name as “group”
#Bind groups together into 2 columns
group <- cbind(case, cont)Provide new column names
#Provide the column names to both groups
colnames(group) <- c("case", "control")Draw a box plot
#Draw boxplot for groups of samples
boxplot(group, main="Boxplot for groups", xlab="Groups", ylab="Gene expression(log[FPKM])", col=c('cyan', 'pink'))
Scatter Plot Scatterplot between 2 samples (with FPKM values) in the same condition: geom_point() → for relationships between two variables Use geom_dotplot() → for distribution of one variable
#Extract samples of same group
sample1 <- df1[,1]
sample2 <- df1[,2]
#Combine them column-wise
s1_2 <- cbind(sample1,sample2)
#Draw Scatterplot
plot(s1_2)
ggplot(s1_2, aes(x = sample1, y = sample2)) +
geom_point()
Scatterplot between 2 samples (with FPKM values) from different conditions:
# Extract samples of different groups
sample3 <- df1[,1]
sample4 <- df1[,19]
# Combine them column-wise
s3_4 <- cbind(sample3,sample4)
# Draw Scatterplot
plot(s3_4)
ggplot(s3_4, aes(x = sample3, y = sample4)) +
geom_point()
Whole data
# draw scatter plot for whole data (with FPKM values)
G1 <- df1 [1:2]
G2 <- df1[18:20]
G <- cbind(G1,G2)
plot(G)
Bar Plot Now let’s see how we can visualize gene expression for a single sample. We can do this using a bar plot.
#Take the 2nd sample
sample1 <- log_df[,2]
#Make a bar plot
barplot(sample1, main="Barplot of Sample", xlab="Genes", ylab="Gene expression (log[FPKM])")
Correlation plot between samples using corrplot. logdf will be susubsetted for case in 1:17 columns. colorbrewer library will change the color
### correlation plot between genes
logdf_case <- log_df[, 1:17]
cor<- cor(logdf_case)
corrplot(cor, method = c("circle"), type = c("lower"), sig.level = 0.05)
Col<- (brewer.pal(8,"BrBG")) ### adding 8 layers of colors in BrBG
corrplot(cor, method = c("circle"), type = c("lower"), sig.level = 0.05,
col = Col, # Apply RColorBrewer palette
order = "hclust", # Reorder based on clustering
addCoef.col = "black", # Add correlation coefficients
tl.col = "black", # Text label color
tl.srt = 45, # Text label rotation
number.cex = 0.5) # Coefficient font size
#Extract 2nd sample with first 20 genes
sample_20genes <- log_df[1:20,2] ### first 20 genes
# Add colors to the barplot
barplot(sample_20genes, main="Barplot of Sample for 20 genes", xlab="Genes", ylab="Gene expression (log[FPKM])", col = rainbow(length(sample_20genes)))
Next, let’s see how we can visualize gene expression for a single gene among multiple samples. We can do this using a bar plot.
#Extract 135th gene from the data
gene_135 <- log_df[135, ]
#Convert this data frame into a matrix
gene_135_m <- as.matrix(gene_135)
#Extract gene ids list from the data
Genes <- rownames(log_df)
#Extract the gene id for 135th gene from list
gene_id <- Genes[135]
#Draw barplot for the selected gene
barplot(gene_135_m, main= gene_id, xlab="", ylab="Gene expression (log[FPKM])", las = 2, cex.names = 0.7, col = "red")
Creating volcano plot creating a function for t-test to find the p-value
#function()
ttest <- function(df, grp1, grp2) {
x = df[grp1]
y = df[grp2]
x = as.numeric(x)
y = as.numeric(y)
results = t.test(x, y)
results$p.value
}#Calculating P values
# P value < 0.01 and fold change of > 1.5 (log2FC of 0.58)
#calculating p-value
rawpvalue = apply(df, 1, ttest, grp1 = c(1:17), grp2 = c(18:41))
####transform our data into log2 base.
log_df2 = log2(df)
##calculate the mean of each gene per control group
control = apply(log_df2[,18:41], 1, mean)
#calculate the mean of each gene per case group
case = apply(log_df2[, 1:17], 1, mean)
#confirming that we have a vector of numbers
class(control)## [1] "numeric"class(case)## [1] "numeric"#because our data is already log2 transformed, we can take the difference between the means.
##And this is our log2 Fold Change or log2 Ratio == log2(control / test)##
foldchange <- case - control# Transform the p-value (-1*log(p-value)) and create a volcano plot using ggplot2
results = cbind(foldchange, rawpvalue)
results = as.data.frame(results)
results$genename <- rownames(results)
results$delabel[1:100] <- "up"
results$delabel[101:200] <- "down"#Plotting
library(ggplot2)
ggplot(results, aes(x=foldchange, y=-1*log10(rawpvalue))) +
geom_point(aes(colour = delabel), alpha=0.5, size=4) +
scale_colour_manual(values = c("up"= "red", "down"="blue"))
#Saving list of Up-regualated and down-regulated genes
##saving list of upregulated genes
write.table(results$genename[1:100], file = "uregulated_genes.txt", sep = "\t", dec = ".", row.names = FALSE, col.names = TRUE, quote=FALSE)
##saving list of downregulated genes
write.table(results$genename[101:200], file = "downregulated_genes.txt", sep = "\t", dec = ".", row.names = FALSE, col.names = TRUE, quote=FALSE)Heatmap
#Heatmap for all the genes
M1 <- as.matrix(log_df2)
#install.packages("pheatmap")
library(pheatmap)## Warning: package 'pheatmap' was built under R version 4.5.3#ploting heatmap
# Heat map
pdf(file = "Heatmap_all_genes_test.pdf", # The directory you want to save the file in
width = 60, # The width of the plot in inches
height = 60)
pheatmap(M1,
cellnote = M1, # Display cell values as text
cellnote.fontsize = 5, # Set the font size
fontsize = 9, # Set the general font size
main = "Heatmap for all genes") # Add a main title
#dev.off()Upreglated genes
#Heatmap for all upregulated genes
M1_up <-as.matrix(log_df2[1:20,])
Matrix_up <- as.matrix(log_df[1:20,])
pdf(file = "Heatmap_upregulated_counts_log_df_2.pdf", # The directory you want to save the file in
width = 50, # The width of the plot in inches
height = 50)
pheatmap(
M1_up,
color = colorRampPalette(c("navy", "white", "firebrick3"))(100), # Custom color palette
clustering_method = "complete", # Hierarchical clustering method
show_rownames = TRUE, # Hide row names
show_colnames = TRUE, # Show column names
fontsize_row = 20, # Font size for row labels
fontsize_col = 20, # Font size for column labels
)
#dev.off()Plot of upregulated genes
# Heat map
pdf(file = "Plot_upregulated_genes_1.pdf", # The directory you want to save the file in
width = 30, # The width of the plot in inches
height = 30)
pheatmap(M1_up,
cellnote = M1_up, # Display cell values as text
cellnote.fontsize = 5, # Set the font size
fontsize = 9, # Set the general font size
main = "Heatmap for upregulated genes") # Add a main title
#dev.off()Downregulated genes
#Heatmap for all downregulated genes
M1_down <-as.matrix(log_df2[101:200,])
# Heat map
pdf(file = "Plot_downregulated_genes_1.pdf", # The directory you want to save the file in
width = 70, # The width of the plot in inches
height = 70)
pheatmap(M1_down,
cellnote = M1_down, # Display cell values as text
cellnote.fontsize = 5, # Set the font size
fontsize = 9, # Set the general font size
main = "Heatmap for upregulated genes") # Add a main title
#dev.off()End of Session