Table 2.
Paired comparison results between automated annotation services
| IGS vs IMG | IGS vs JCVI | IGS vs RAST | IMG vs JCVI | IMG vs RAST | JCVI vs RAST | |
|---|---|---|---|---|---|---|
| Baseline |
62% |
51% |
39% |
46% |
40% |
32% |
| One HP another non-HP |
1% |
1% |
1% |
2% |
2% |
1% |
| Annotation and database term |
23% |
20% |
11% |
27% |
14% |
30% |
| Gene symbol entry |
1% |
2% |
5% |
1% |
4% |
5% |
| Pfam entry |
3% |
7% |
11% |
6% |
10% |
8% |
| Orthologue |
1% |
2% |
7% |
3% |
7% |
6% |
| Vector space model |
1% |
2% |
3% |
1% |
2% |
2% |
| Matching |
1% |
3% |
6% |
1% |
4% |
1% |
| Overall result | 93% | 88% | 83% | 87% | 83% | 85% |
Baseline: baseline annotation comparison result; One HP another non-HP: one annotation is a ‘hypothetical protein’ and another one has a characterised function; Annotation and database term: two annotations have the same annotation text or database terms; Gene symbol entry: annotations are derived from the same gene symbol entry; Pfam entry: annotations are derived from the same Pfam entry; Orthologue: annotations are derived from the same OGA; Vector space model: vector space model comparison; Matching: matching between comparison results; Overall result: total number of gene annotations compared.