Skip to main content
. Author manuscript; available in PMC: 2009 Sep 22.
Published in final edited form as: IEEE/ACM Trans Comput Biol Bioinform. 2006 Apr–Jun;3(2):114–125. doi: 10.1109/TCBB.2006.22

Table 5.

Double-Blind Composite Classifier Accuracies

Train Test Accuracy Correlation
K1 X1-X24 (Pre) 53.5% −0.09
K1-K5* X1-X5* 46.7% 0.17
K1-K24* X1-X24* 63.4% 0.00
K20-K24* X20-X24* 85.7% 0.69

K1 and X1-X24 are the initial sets of known and unknown mutants (respectively). K1, X1-X24 (Pre) is a double-blind prediction using K1 to predict X1-X24. Ky-Kz*, Xy-Xz* are iterated predictions using sets Ky-Kz to predict iterations Xy-Xz (respectively).