Figure 1.
Visualization of complementarity score and illustration of triangle types.
(A) Constitution of the two TOR sub-complexes TORC1 and TORC2 complex. (B) Phylogenetic profiles (as an example) of TOR and Rptor and Rictor and the formed triangle. The nodes represent proteins and the lines indicate a pairwise interaction. (C) Visualization of the complementary score and the different triangle types used throughout the study. We consider “open” and “closed” triangles and closed triangles with 2 inparalogs. The ellipses symbolize the different complexes.
Table 1.
Some statistics of the three interaction data sets.
Figure 2.
Distribution of the profile relationships of (A) co-complex protein pairs in the two complex data sets and (B) yeast interactions.
The 4 differently colored slices correspond to the types of similarities between profiles as described in Methods.
Figure 3.
Frequency of open triangles as a function of the triplet score for (A) co-complex protein pairs and (B) yeast interactions.
Each bar corresponds to an equally-sized bin, with the right-most bar belonging to the best scoring triplets. The three small plots on the right side of the figures displays ratios between different categories.
Figure 4.
Distribution of profile-pair types among co-complex proteins and yeast interactions.
The pairs are divided into the the same categories as in the pie charts of Figure 2. The dark part of each bar corresponds to the number of co-complex protein pairs or yeast interactions that are part of a positive-scoring triplet.
Figure 5.
Heat maps showing the enrichment of (A) co-complex pairs or (B) yeast interactions as a function of pairwise profile similarity scores and triplet-based complementarity scores.
The color intensity corresponds to the log2-enrichment of protein pairs that are part of the same complex or interact, among all protein pairs inside a pair- and triplet-based bin.
Table 2.
Examples of triplet profiles.
Table 3.
Percentages of duplicated genes among different categories.