Fig. 2

Relationships between transcriptome samples. A Correlation matrix heat map of transcript expression across all samples. Cluster dendrogram and spearman correlation coefficient heatmap are based on normalized TPM (transcripts per million mapped reads) values of expressed transcripts. The spearman correlation coefficients between different biological repetitions were calculated by the cor function in R software. Red indicates higher correlation; blue indicates lower correlation. The legend is added in the top right corner. B Scree plot of PCA. The first three principal components can explain more than 43% of the variability among the samples. C, D Principal component analysis (PCA) for the 28 samples. BML1234 in light the red circle. L2010-3 in the light blue triangle. The explained variances are shown in brackets. The cos2 of variables on all the dimensions are shown in different shape size. A high cos2 indicates a good representation of the variable on the principal component. PCA was performed using the R function “prcomp” based on the normalized read counts. The correlation heatmap and PCA diagram were drawn by the pheatmap package and factoextra package in R software, respectively