Overlaps between two top deregulated gene lists by nDGE in non-smoker samples of the different data sets. The S2, S5 and S6 represent the data sets Smoker2, Smoker5 and Smoker6. “#” and “*” indicate the number of overlap is significantly larger than that in permutation background with P < 0.00005 and P < 0.00001, respectively. A to C show overlaps of top 10, 50 and 100 up regulated genes between the different data sets. D to F show overlaps of top 10, 50 and 100 downregulated genes between the different data sets.