at and its use can generate a spreadsheet shown in Table 6.3.

del=topTreat(treat(sam.model),

number=nrow(X))

op.model)

Table 6.3. The spreadsheet generated by topTreat.

logFC

AveExpr

t

P.Value

adj.P.Val

3.292589

3.290646

302.4542

2.610402e-23

2.965485e-20

1

3.314729

3.293327

299.0579

2.963821e-23

2.965485e-20

3.247577

3.244655

297.7635

3.093526e-23

2.965485e-20

3.334391

3.330026

297.4771

3.148082e-23

2.965485e-20

3.360363

3.352813

296.3017

3.296045e-23

2.965485e-20

3.371104

3.379636

294.4377

3.537942e-23

2.965485e-20

The heatmap of 20 top DEGs for the prostate cancer data. ‘P’ stands for the

mour and ‘M’ stands for the metastasis tumour. The DEGs were detected using

ological significance and the statistical significance.

ow how DEGs contribute to the discrimination power between

es of tumours of the prostate cancer data, the heatmap for top 20

as generated and is shown in Figure 6.10. The DEGs were

ed based on both the biological significance and the statistical

nce. It can be seen that two classes of tumours were well separated