Closed ValWood closed 1 year ago
Assigning to @Antonialock to assign the right person, and check if this should be an annotation check
The process is automatic so the only way of getting rid of this interaction at the IntAct end would be to add a filter so all cross-species interactions are removed from the IntAct2GO (and UniProt) exports. This has been discussed in the past, but would mean (a) all host-pathogen interactions and (b) all interactions designed to show sequence conservation are lost so it was always agreed the loss outweighed the gain. A data import filter at your end would be the other alternative.
Hmm it would be a shame to use pathogen-host interactions from GO, as these are real interactions.
Is it possible to make a way to distinguish these ? It seems a shame that the pathogen-host interactions and the sequence conservation examples are not distinguishable. MAybe they could have a specific "qualifier" to distinguish them.
This would also enable the pathogen-host interaction pairs to be accessed as a set. I suspect PHI-base will be interested in this set as they are likely to be represent proteins that should be included in include . Is it possible to do a query to retrieve all interspecies physical interactions?
We can filter them locally from PomBase.
Can this be closed now?
Yes I guess (note to self add filter to PomBase)
Could this annotation be deleted, it records a physical interaction between a pombe protein and human protein (these shouldn't get through to GO should they)?
INTA | P04910 | hta2 | enables | GO:0005515 | protein binding | ECO:0000353(IPI) | ECO:0000353 | (IPI) | | PMID:21633354 | | UniProtKB:Q15003
ECO:0000353 | (IPI)