Evidence code for 'ISS from IC evidence'
Discussions at the last GO annotation call on the 7th of January came to the conclusion that it would be desirable and reasonable to transfer IC annotation by ISS, however it was also agreed that it is important that such ISS-from IC annotation should be given their own evidence code. See minutes.
However, this does add a second aspect to ISS - not only the method for obtaining/propagating GO terms (via orthologs; ISO, sequence alignments; ISA or sequence models; ISM) but also the type of evidence supplied to the original annotation (EXP or IC)
- should we create one new code (e.g. ISC) to be a sibling or subclass of ISS? Or should we create ISA and ISO siblings that are specifc for IC-evidence propagation?
FIRST DRAFT: Inferred from Sequence Similarity, using Curator Judgement (ISC)
The ISC evidence code should only be used delete whenever when a sequence-based analysis forms the basis for an annotation and review of the evidence and annotation has been done manually. If the annotation has not been reviewed manually, the correct evidence code is IEA, even if the evidence supporting the annotation is all sequence based.
ISC evidence code has been created to support curation groups who would like to transfer IC-evidenced GO terms from a well-characterized gene product, to a homologue where it is not possible to supply the GO term by an ISS annotation method.
ISC-evidenced annotations complement the set of ISS, ISO or ISA annotations for an uncharacterized gene product which are propagated from experimentally-evidenced annotations in a homolog’s entry; allowing the recipient gene product to be supplied with annotations from high-quality annotation sources. In all respects, the procedure for creating an ISC-evidenced annotation is similar to that for ISS, ISA or ISO. Population of the ‘with’ field (column 8) with the identifier of the ‘donor’ gene product identifier is essential when using the ISC code.
However, ISC-evidenced annotations should only be created when use of the ISS, ISA or ISO evidence codes (to propagate experimentally-evidenced annotation) is possible. IC-evidenced annotations are of high-quality; generated using a combination of domain-knowledge and conservative curator judgement along with experimental-evidence for a connected function/process/component annotation. For further information on the IC evidence code please see here.
gpA is annotated to 'DNA helicase activity' by IDA, 'DNA duplex unwinding' by IC from helicase activity
gpB is similar to gpA
gpB gets 'DNA helicase activity' by ISS from EXP-evidenced annotation with gpA
gpB gets 'DNA duplex unwinding' by ISS from IC-evidenced annotation with gpA
- Should we also make IC-evidence codes that describe specifically sequence alignment or orthology data (ISA and ISO equivalents)?
Meeting consensus: no, one evidence for all types of ISS from IC is enough.