Skip to main content
. 2018 Sep 8;34(17):i944–i953. doi: 10.1093/bioinformatics/bty570

Fig. 4.

Fig. 4.

The panels give the average percentage of clusters enriched with respect to the total number of clusters for yeast–complexes (left) and yeast–pathways (right), the standard deviation is not represented to avoid overcrowding the panels. The colours represent the models from which the clustering is obtained: HG in blue and PPI in orange. The type of line represents the category of GO annotations: BP are full lines, MF are dashed lines and CC are dotted line. The black vertical lines signal the number of clusters selected from the set of NMI and SSE curves according to the procedure described in Section 3.4.1