Annotation Assigned by

From GO Wiki
Jump to navigation Jump to search

Draft for Assigned_by Documentation

Introduction

It is mandatory for every GO annotation to supply a value in the 'assigned_by' field column 15.

The value in this field should acknowledge the group which made the decision to link the GO term to a DB Object ID. For manual annotations this field will display the name of the team with whom the curator who created the GO annotation is affiliated. Similarly, for electronic annotations it is the group that made the original decision to link GO identifier to a DB Object ID that should be acknowledged in this field.

All annotation groups that should be used in the 'assigned_by' field should have an entry in the GO Database Abbreviations file.

Assigned_by Value for External2GO annotations

Where an external2go file has been applied to produce GO annotation, it is the name of the group that created the external annotation set that should be acknowledged in the assigned_by field (and in many cases this will also be the group that is responsible for the external2go file that should be applied over the external annotation set), as they can be considered the ‘intellectual owners’ of the GO annotations. A third party that uses the external2go and external annotation resources to generate the GO annotation set does not need to be acknowledged in the assigned_by field, as in many instances they will have been directly following the annotation decisions contained in the external2GO file and external annotations. Only where a curation group subsequently carries out a review and reinterpretation of the produced data should they be acknowledged as the creators of the annotations.

InterPro2GO

If you use the InterPro2GO mapping developed and maintained by InterPro group, and use it in the manner described by the InterPro2GO reference GO_REF:0000002 to generate IEA annotations, then the resultant annotations are those that the InterPro curators have suggested for your sequences based on matches from the InterPro signatures. Therefore the assigned_by field for these annotations should always be ‘InterPro’.

For InterPro2GO, please also consider using the GO_REF: 0000002 in the DB:Reference field (column 6)


Swiss-Prot Keyword to GO (SPKW2GO)

Swiss-Prot Subcellular Location to GO (SPSL2GO)

HAMAP2GO

For SPKW2GO, SPSL2GO and HAMAP2GO mappings, the assigned_by field should always display 'UniProtKB'. Both SPKW2GO and SPSL2GO mappings were created specifically by UniProtKB to translate from UniProtKB keyword definitions to GO terms, therefore can only be correctly ever be used to extract GO annotations from UniProtKB entries. In a similar manner, UniProtKB proteins are manually assigned to HAMAP family rules by the HAMAP team in UniProtKB, who also generate the HAMAP2GO mapping.

For SPKW2GO, please also consider using the GO_REF: 0000004 in the DB:Reference field (column 6)

For SPSL2GO, please also consider using the GO_REF: 0000023 in the DB:Reference field (column 6)

For HAMAP2GO, please also consider using the GO_REF: 0000020 in the DB:Reference field (column 6)


EC2GO

EC2GO is created directly from the GO OBO file, however, it can be used to generate GO annotations from a number of different external data sets. Ideally a GO annotation would acknowledge both the translation work carried out by curators of the EC2GO and also the external database group that made the original annotation decision to apply an EC number to one of their sequences. However, as only one value is currently allowed in column 15, then it should equate to the external database from where the external annotations originated, as the reference applied to the annotation set (column 6), will have identified the generator of the external2go file. For example, if the EC2GO file was used in combination with the UniProtKB dataset to generate annotations, the assigned_by field should be ‘UniProtKB’. In this manner, the origin of the annotation set can be accurately determined.

Please also consider using the GO_REF: 0000003 in the DB:Reference field (column 6)


Comments