Viral term WebEx meeting august 21 2009
2pm GMT/8am CT/9am ET
WebEx and Conference Call:
from US only: 888 727 6732
from Europe: 719 867 3417
Present: Michelle, Candace, Marcus, Fiona, Jane, Alex
We discussed whether we needed to have - as well as a general term for a host-virus process - separate terms for the virus and host contributions to the process. For example:
viral replication within host cell [GENERAL] ---[i] replication of viral symbiont within host cell by virus [VIRUS] ---[i] replication of viral symbiont within host cell by host [HOST]
See also Candace's email 21 July 2009
The advantage of having the child terms would be that it would allow us to be able to distinguish those gps which a virus contributes to a process versus those the host contributes. We found the problem with the child terms was that it was difficult to find sensible names for what we were trying to describe, which was essentially the role the organism contributing a given gp plays in a process. Other term names we considered were 'reproduction of virus within host cell via viral protein' but this was rejected because it refers to the protein being annotated.
The alternative would be to use just the general term and to annotate both the host and viral genes to this process. The limitation of this strategy is that there would be no automated means of distinguishing annotated host genes from viral genes. There was some discussion of how this might be done using taxon ids but this would be rather imprecise given the large range of possible host taxa.
We felt that the problem we were having was that we were trying to including information in term names regarding the relationship between a gp, the organism encoding it and the process it was being annotated to. This is related to a mailing list discussion we've been having with Chris about designating relationships between gps and terms in annotations. From Chris's mail:
I would prefer we move to a system whereby annotations are sentences; e.g. [gene product] [relation] [function] [gene product] [relation] [process] [gene product] [relation] [component]
So in this case, we might have:
[gene product] [is encoded by host organism and is involved in] [reproduction of virus]
(could probably think of a better name for that relation!)
We decided to take this to the next GOC meeting as part of a wider discussion about relations between gps and terms in annotations, and for this stage in term development create only the GENERAL terms as described above.
2. Reproduction v/s growth v/s development
GO has distinct high-level terms for reproduction, growth and development, and we recently split the MOP terms in line with this. For viruses we decided that the most sensible strategy would be to use only 'reproduction' and to not have growth or development terms at all.
There was some further discussion about the structure of the node for the viral terms. We decided that 'viral reproduction' would encompass all viral processes (alternatively 'viral life-cycle'?). There was some concern about potential confusion between 'viral reproduction' and 'viral replication' which would be a child - 'viral replication' is just the replication of the particle itself. We decided to call the term 'replication of viral components' to avoid this confusion. Proposed graph with some additional terms we discussed:
viral reproduction ---[i] replication of viral components ------[i] replication of viral nucleic acid ---[i] viral assembly
We also decided that for the production of viral proteins we would create a term as a child of 'gene expression' that would also be part of viral replication, e.g.
gene expression ---[i] expression of viral gene syn: viral protein production
replication of viral components ---[p] expression of viral gene syn: viral protein production
(note: 'expression of viral gene' may or may not include transcription - would need to make this clear in the definition)
We also discussed the processing of viral proteins and whether the general GO term 'protein processing' could be used. We decided it couldn't and we would need a new term e.g. 'processing of viral proteins'.
3. Symbiont granularity
Is this a good split for levels of granularity for the symbiont
non-cellular (acellular?) symbiont unicellular symbiont multicellular symbiont
replication of non-cellular symbiont within host
We decided to stick with just non-cellular v/s cellular for now, until we have any multicellular symbionts being annotated in which case we can make a judgement.