Hi everyone,

I have an rna-seq dataset of three biological replicates for a control and three for the treatment. I wanted to do some data mining, including heirachial clustering and heatmaps, to see how genes are clustering with the respective treatments. I was working on this in R but I am getting confused about the output and I was hoping someone could check that I am taking the right path. In R:

1) First the organization of the data: the rows = samples and cols=genes

the dataframe is called: L.normalized.counts.up

2) in R what I have done:

#convert dataframe into matrix, and the transpose for the rows = genes, cols=samples matrix

m_matrix <- data.matrix(L.normalized.counts.up)
t_matrix <- t(m_matrix)

#HC of genes

corr <- 1 - (cor(m_matrix, method = "pearson"))
disr <-dist(corr)
hr <- hclust(disr, method = "average")
dendro_hr <- as.dendrogram(hr)

#HC of samples

cort <- 1 - cor(t(m_matrix))
distt <- dist(cort)
hc <- hclust(distt, method = "average")
dendro_hc <- as.dendrogram(hc)

#heatmap of HC of samples

result1<-heatmap3(t_matrix, Colv= dendro_hc, cexRow = 1, cexCol = 1, labRow = "", balanceCol = T)

#heatmap of HC of genes

result2<-heatmap3(t_matrix, Rowv= dendro_hr, cexRow = 1, cexCol = 1, labRow = "", balanceCol = T)

I understand the HC clustering and the use of the distance matrix. But, I don't understand how to interpret the output with regard to the coloration and how I have specific the heatmaps to be generated. Along the y-axis is the HC of the genes (with a dendrogram), and along the x-axis are the samples (also with a dendrogram). Would anyone be able to clarify?

Thanks in advance

The above image is the plot of result2 heatmap.