Figure 3

Rank aggregation of gene lists using the GA algorithm. Plots created by the plot( ) function for the GA rank aggregation of the gene lists. We can see that algorithm stabilized after roughly 500 iterations. The distribution of the population in the final generation is concentrated just to the right of the optimal solution. The bottom plot clearly shows why the solution makes sense. Genes ranked high in the final list usually come from several individual lists as indicated by the presence of multiple lines intersecting. The genes at the end of the final list are the ones included in a single list but somewhere close to the top. The rank of 26 is artificial in our procedure and it simply indicates that that particular gene is not present in the individual list.