Table 3

Validation of clusters: PAM analysis generating classifiers for clusters validated in independent dataset GSE2658

CD1 clusterCD2 clusterMF clusterMS clusterPR clusterHY clusterMyeloid cluster
Samples in training set* 13 34 32 33 15 77 39 
Samples in validation set (GSE2568) 28 60 37 68 47 116 145 
No. of probes in classifier 106 45 35 86 56 87 
PPV, % 57.60 80.60 97.10 94.30 61.70 76.70 58.50 
NPV, % 97.60 98.30 98.90 99.40 95.10 94.60 81.70 
Sn, % 67.90 90.00 89.20 97.10 61.70 84.80 42.80 
Sp, % 96.40 96.30 99.70 98.80 95.10 91.30 89.40 
CD1 clusterCD2 clusterMF clusterMS clusterPR clusterHY clusterMyeloid cluster
Samples in training set* 13 34 32 33 15 77 39 
Samples in validation set (GSE2568) 28 60 37 68 47 116 145 
No. of probes in classifier 106 45 35 86 56 87 
PPV, % 57.60 80.60 97.10 94.30 61.70 76.70 58.50 
NPV, % 97.60 98.30 98.90 99.40 95.10 94.60 81.70 
Sn, % 67.90 90.00 89.20 97.10 61.70 84.80 42.80 
Sp, % 96.40 96.30 99.70 98.80 95.10 91.30 89.40 

Per cluster, the number of samples and total number of samples in the training set and test set are shown, the number of probe sets in the classifier, and the PPV, NPV, Sn, and Sp.

NPV indicates negative predictive value; PAM, prediction analysis of microarrays; PPV, positive predictive value; PR, proliferation-associated genes; Sn, sensitivity; and Sp, specificity.

*

The total of all samples in the training set was 320.

The total of all samples in the validation set was 414 except in the Myeloid cluster, where the total was 559.

Close Modal

or Create an Account

Close Modal
Close Modal