Skip to main content
. 2016 Mar 21;12(3):e1004801. doi: 10.1371/journal.pcbi.1004801

Table 2. Vector class prediction efficiency.

RFM model predictions*
Vectors Material Strong Weak
LV_1 DC 100 0
LV_2 DC 100 0
A AP205_3 DC 3 97
MLV_2 DC 5 95
AP205_1 Spleen 99 1
MVA_1 Spleen 100 0
B rAd_1 Spleen 98 2
MLV_1 Spleen 9 91
MPT_1 Spleen 0 100
AP205_1 PBMC 73 27
MVA_1 PBMC 90 10
Qb_1 PBMC 99 1
Qb_2 PBMC 95 5
C Qb_3 PBMC 100 0
Qb_4 PBMC 100 0
Qb_5 PBMC 98 2
MLV_1 PBMC 0 100
MPT_1 PBMC 0 100
rAd_1_6 Spleen 98 2
D rAd_1_48 Spleen 0 100
rAd_1_72 Spleen 2 98

*Number of the 100 bootstrapped datasets predicted as “Strong” or “Weak”.