Table 1 GNA algorithm evaluation results. Three GNA algorithms were evaluated based on their ability to correctly align predicted ortholog proteins (51) and identify interologs (22). Bold is used to indicate the best value for each metric. All algorithms failed to align the networks using topological information exclusively and combination with biological information proved necessary for the correct alignment of protein orthologs.
Alignment based on topological information | Alignment based on topological and biological information | ||||
|---|---|---|---|---|---|
MAGNA++ | CytoGEDEVO | NETAL | MAGNA++ | CytoGEDEVO | |
Correct node alignments | 0 | 3 | 0 | 33 | 24 |
Incorrect node alignments | 51 | 48 | 51 | 18 | 27 |
Aligned interactions | 56 | 46 | 40 | 31 | 39 |
Interologs identified | 0 | 0 | 0 | 15 | 12 |