Annotation involves taking data and presenting it in a way that will allow for the extraction of information by adding further knowledge to the data (meta-data). Manual annotation is undertaken by humans, and although this is more labour intensive than automatic annotation, it is generally considered to be more precise.
In Ensembl: Manual annotation of the human, mouse and zebrafish genomes from the VEGA/Havana group refers to genes that have been individually analysed by a team of experts in order to determine the transcript set at each locus.
In UniProt: Manual annotation (or curation) consists of a critical review of experimental and predicted data for each protein as well as manual verification of each protein sequence. Curation methods applied to UniProtKB/Swiss-Prot include manual extraction and structuring of information from the literature, manual verification of results from computational analyses, mining and integration of large-scale data sets, and continuous updating as new information becomes available.
Examples of manually annotated databses include UniProtKB/Swiss-Prot and IntAct.