The solutions given below are just one way of obtaining the desired plots! There are probably several different ways you could code to get the same plots.
For the exercises in the first two lab sessions: R introduction and the ggplot basics 1, all you need to do is to replace the filename raw_counts.txt with each of the different files to look at the differences between different normalization methods.
Task Plot 1:
gc_raw <- read.table(file = "data/counts_raw.txt", sep = "\t", header = T)
gc_filt <- read.table(file = "data/counts_filtered.txt", sep = "\t", header = T)
gc_vst <- read.table(file = "data/counts_vst.txt", sep = "\t", header = T)
gc_deseq <- read.table(file = "data/counts_deseq2.txt", sep = "\t", header = T)
md <- read.table("data/metadata_raw.csv", header = T, sep = ";")
gene_counts_all <-
gc_raw %>% gather(Sample_ID, Raw, -Gene) %>%
full_join(gc_filt %>% gather(Sample_ID, Filtered, -Gene), by = c("Gene", "Sample_ID")) %>%
full_join(gc_vst %>% gather(Sample_ID, VST, -Gene), by = c("Gene", "Sample_ID")) %>%
full_join(gc_deseq %>% gather(Sample_ID, DESeq2, -Gene), by = c("Gene", "Sample_ID")) %>%
gather(Method, count, Raw:DESeq2) %>%
filter(!is.na(count)) %>%
full_join(md, by = "Sample_ID")
gene_counts_all$Time <- factor(gene_counts_all$Time, levels = c("t0","t2","t6","t24"))
gene_counts_all$Replicate <- factor(gene_counts_all$Replicate, levels = c("A","B","C"))
gene_counts_all$Method <- factor(gene_counts_all$Method, levels = c("Raw","Filtered","DESeq2","VST"))
gene_counts_all %>%
group_by(Time, Replicate, Method) %>%
summarise(mean=mean(log10(count +1)),se=se(log10(count +1))) %>%
ggplot(aes(x= Time, y= mean, fill = Replicate)) +
geom_bar(position = position_dodge2(), stat = "identity") +
geom_errorbar(aes(ymin=mean-se, ymax=mean+se), position = position_dodge2(.9, padding = .6)) +
facet_wrap(~Method, scales = "free")Task Plot 2:
gc_long %>%
group_by(Time, Replicate) %>%
summarise(mean=mean(log10(count +1)),se=se(log10(count +1))) %>%
ggplot(aes(x=Time, y=mean, color = Replicate)) +
facet_wrap(~Replicate) +
geom_line(aes(group=1), stat= "identity", size = 2) +
scale_x_discrete(limits= c("t0", "t2", "t24")) +
scale_y_continuous(limits = c(0.4,0.8), breaks = seq(0.4,0.8,0.05)) +
guides(color="none") +
ylab(label = "mean(log10(count + 1))") +
theme_light() +
theme(axis.text = element_text(face="bold", size=12),
axis.title = element_text(face="bold", color = "#C84DF9", size=14),
axis.ticks = element_blank())p4 <- ggplot(data=iris,mapping=aes(x=Sepal.Length, y = Sepal.Width, color = Species)) +
geom_point(size = 3, alpha = 0.6) +
theme_classic(base_size = 12) +
border()
d1 <- ggplot(data=iris,mapping=aes(Sepal.Length, fill = Species)) +
geom_density(alpha = 0.6) +
theme_classic() +
clean_theme() +
theme(legend.position = "none")
d2 <- ggplot(data=iris,mapping=aes(Sepal.Width, fill = Species)) +
geom_density(alpha = 0.6) +
theme_classic() +
clean_theme() +
theme(legend.position = "none") +
rotate()
ggarrange(d1, NULL, p4, d2,
ncol = 2, nrow = 2, align = "hv",
widths = c(3, 1), heights = c(1, 3),
common.legend = TRUE)Eigenvalues <- gc_mds$eig
Variance <- Eigenvalues / sum(Eigenvalues)
Variance1 <- 100 * signif(Variance[1], 3)
Variance2 <- 100 * signif(Variance[2], 3)
gc_mds_long <- gc_mds$points %>%
as.data.frame() %>%
rownames_to_column("Sample_ID") %>%
full_join(md, by = "Sample_ID")
gc_mds_long$Sample_Name <- factor(gc_mds_long$Sample_Name, levels = c("t0_A","t0_B","t0_C","t2_A","t2_B","t2_C","t6_A","t6_B","t6_C","t24_A","t24_B","t24_C"))
gc_mds_long$Time <- factor(gc_mds_long$Time, levels = c("t0","t2","t6","t24"))
gc_mds_long$Replicate <- factor(gc_mds_long$Replicate, levels = c("A","B","C"))
ggplot(gc_mds_long, aes(x=V1, y=V2, color = Time)) +
geom_point(size = 3, aes(shape = Replicate)) +
xlab(paste("PCO1: ", Variance1, "%")) +
ylab(paste("PCO2: ", Variance2, "%")) +
geom_vline(xintercept = 0, linetype=2) +
geom_hline(yintercept = 0, linetype=2) +
theme_bw() +
theme(panel.grid.major = element_blank(),
panel.grid.minor = element_blank()) End of document