METHOD | Run 1 | Run 2 | Run 3 | Run 4 | Run 5 | Avg. |
|---|
MOPHY | 5.28 | 5.85 | 8.82 | 5.82 | 8.15 | 6.79 |
RDL | 14.40 | 15.50 | 19.80 | 13.12 | 14.83 | 15.53 |
Random Modules | 14.81 | 9.33 | 11.54 | 8.11 | 12.6 | 11.29 |
Only Protein Similarity | 11.72 | 11.56 | 13.96 | 11.22 | 10.75 | 11.84 |
Random Homolog Selection | 15.01 | 13.60 | 17.73 | 22.72 | 17.61 | 17.33 |
- For the simulated networks, we compare the performance of MOPHY with the random module method, random homolog selection method, RDL and a method that only uses protein similarity in the networks. Nodal distance for five simulation instances as well as the average values of these five runs are shown in the table. For MOPHY, the result used is the best performance achieved with coverage 0.60 and diameter 2, by using the most specific modules. Similarly for the random module method, the best performance is achieved for the instance with coverage 0.40 and diameter 4, with the most comprehensive modules. For the random homolog selection method, the best result is achieved with coverage 0.60 and diameter 3, by using the most specific modules. As clearly seen, MOPHY outperforms the other methods in terms of capturing the evolutionary distances between species.
