Similarity values obtained by G-SESAME
| | GTT2 | GTT1 | GLR1 | GPX1 | HYR1 | GPX2 |
| GTT2 | | 1 | 0.186 | 0.234 | 0.234 | 0.234 |
| GTT1 | | | 0.186 | 0.234 | 0.234 | 0.234 |
| GLR1 | | | | 0.379 | 0.379 | 0.379 |
| GPX1 | | | | | 0.955 | 1 |
| HYR1 | | | | | | 0.955 |
| GPX2 | | | | | | |
Clustering results obtained by G-SESAME
| Threshold | Initial | 1.000 | 0.950 | 0.370 | 0.230 |
Clustering Result |
|
|
|
|
GPX2 GPX1 HYR1 GLR1 GTT1 GTT2
|
|
Similarity values obtained by Resnik's method
| | GTT2 | GTT1 | GLR1 | GPX1 | HYR1 | GPX2 |
| GTT2 | | 5.496 | 0.281 | 0.281 | 0.281 | 0.281 |
| GTT1 | | | 0.281 | 0.281 | 0.281 | 0.281 |
| GLR1 | | | | 0.281 | 0.281 | 0.281 |
| GPX1 | | | | | 6.098 | 6.098 |
| HYR1 | | | | | | 6.036 |
| GPX2 | | | | | | |
Clustering results obtained by Resnik's method
| Threshold | Initial | 6.098 | 5.398 | 0.198 |
Clustering Result |
|
|
|
GTT1 GTT2 GPX2 HYR1 GPX1 GLR1
|
|
Analysis:
GPX1 and GPX2 are annotated by the same two molecular function terms in SGD database. They are GO:0004602: glutathione peroxidase
activity and GO:0047066: phospholipid-hydroperoxide glutathione peroxidase activity.
GTT1 and GTT2 are annotated by the same molecular function term GO:0004364: glutathione transferase activity.
Therefore, GPX1 and GPX2 have the same molecular functions. GTT1 and GTT2 have the same molecular function.
The similarity values and clustering results obtained by G-SESAME are consistent with the human perception.
However, ResnikĄ¯s method indicates the similarity value between GPX1 and GPX2 is larger than the similarity value between GTT1 and GTT2,
in consistent with human perception.
On the other hand, HYR1 is annotated by three molecular function terms: GO:0004602: glutathione peroxidase activity, GO:0047066: phospholipid-hydroperoxide
glutathione peroxidase activity, and GO:0008379: thioredoxin peroxidase activity.
Therefore, Therefore, HYR1 is slightly different from GPX1 and GPX2. This is also revealed by G-SESAME measurement.
However, the similarity value between HYR1 and GPX1 is larger than the similarity value between GTT1 and GTT2, which are annotated by an identical term.
This is inconsistent with human perspectives.
Conclusion:
G-SESAME is better.