Table 3.
SGP automatic annotation statistics.
| Database | % Annotated (automatic annotation) | ||
| Contigs and singlets | Contigs | Singlets | |
| GO-GOA | 30.5 | 48.5 | 22.0 |
| pdb or swiss-prot | 32.6 | 51.1 | 23.9 |
| pdb | 17.3 | 31.4 | 10.7 |
| swiss-prot | 31.7 | 50.1 | 23 |
| nr | 41.1 | 60.3 | 32.1 |
| nt | 36.7 | 53.7 | 28.6 |
| Distribution between databases | |||
| pdb | 17.3 | 31.4 | 10.7 |
| swiss-prot | 15.3 | 19.7 | 13.2 |
| nr | 8.9 | 9.6 | 8.6 |
| nt | 8.8 | 7.7 | 9.3 |
| any database (pdb + swiss-prot + nr + nt) | 50.3 | 68.4 | 41.8 |
| no hits | 49.7 | 31.6 | 58.2 |
Detailed SGP dataset annotation statistics is available at SGP data resource > Data and results > Annotations > SGP full annotation, statistics.
Databases. NCBI databases – pdb: RCSB-PDB; swiss-prot: SWISS-PROT protein sequence database; nr: all non-redundant GenBank CDS translations + RefSeq Proteins + PDB + SwissProt + PIR + PRF; nt (nucleotide sequences): all GenBank + RefSeq Nucleotides + EMBL + DDBJ + PDB sequences (excluding HTGS0,1,2, EST, GSS, STS, PAT, WGS), no longer "non-redundant". GO-GOA: Gene Ontology assignments for the UNIPROT database produced by the GOA project.
% Annotated. Calculated as the number of sequences with successful annotations for a given subset (i.e. contigs and singlets, contigs, singlets) in a given database to the total number of sequences of this subset where annotation was attempted; GO annotation statistics was calculated separately from other databases as GO hits/GO no-hits. BLAST threshold E-values of 10-10 for PDB and 10-15 for other databases were used.
Distribution between databases. Only one successful annotation per sequence was counted, in the following ranking order: pdb OR swiss prot OR nr OR nt.
SGP dataset. 20019 contig consensus and singlet sequences.