Computational Systems Biology Methods and Protocols.7z

(nextflipdebug5) #1

  1. Heatmap. In the gene expression heatmap (Fig.8b), each row
    represents a gene, and each column represents a sample. Rows
    and columns are hierarchically clustered. Gene sets with specific
    expression patterns can be identified from the heatmap. The
    default genes used in the heatmap are the DEGs called by the
    “DEG calling” module; user can also upload other genes
    through the left menu to plot the heatmap.

  2. Principal component analysis (PCA). PCA projects high-
    dimensional data points onto a low-dimensional space for visu-
    alization. The orthogonal axes of the space are names PC1
    (principal component 1), PC2, and so on. They are chosen in
    such a way that the projected data points have the largest
    variance in the direction of PC1 and the second largest in
    PC2. The overall relationship or clustering of data points in
    the original high dimension can be visualized and identified in
    the low dimension more easily. In this example, the embryonic
    and adult samples are separated into two clusters, which are
    similar to the result from hierarchical clustering (Fig.8c).


Fig. 8(a) The expression levels of Mobp gene in all samples. The expression levels in adult are much higher
than that in embryonic samples. (b) Heatmap for DEGs between embryonic and adult samples. (c) Principal
component analysis shows that the adult and embryonic samples are separated into two clusters, indicating
they are in two different conditions


178 Chao Zhang et al.

Free download pdf