Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Phylogenetic tree information aids supervised learning for predicting protein-protein interaction based on distance matrices

Figure 3

Schematic illustration of TreeSec method to derive super phylogenetic vector from a distance matrix for a given protein from the distance matrix of its orthologous proteins A, B, ..., H. Section 1 across the tree leads to four clusters of the orthologous proteins: α = {A, B}, β = {C}, δ = {D, E, F}, and γ = {G, H}. The distances among these clusters are calculated by Eqs. (4–6), resulting in an intermediate matrix. The procedure is repeated for all sections, producing more intermediate matrices. The upper triangles of the matrices are transformed into vectors and concatenated (denoted by the symbol ⊕) into a super phylogenetic vector. In this way, the phylogenetic vector is extended with extra "bits" that encode the topological information of the protein tree with reference to the species tree.

Back to article page