Skip to main content

Table 2 Accuracy of predicting IgVH mutational status with individual / combined potential biomarkers.

From: Using gene co-expression network analysis to predict biomarkers for chronic lymphocytic leukemia

Genes Prediction Accuracy Sub-cellular location
SH2D1A 57.32% cytoplasmic
IL2RB 68.84% membrane
KLRK1 63.67% membrane
CD247 66.03% membrane
GZMB 57.13% secreted
CD3G 62.52% membrane
CD3D 64.27% membrane
GZMK 57.58% secreted
CD8A 68.31% membrane
NKG7 64.94% membrane
ZAP70 68.46% cytoplasmic
LAG3 59.53% membrane
ZAP70+IL2RB 73.22% -
ZAP70+IL2RB+CD8A 74.62% -
  1. The names with bold letter indicate predicting accuracy above 65%. A linear classifier was used. The cross-validation was carried out with 20% holdout. Each test was carried out independently.