ValWood
commented
6 years ago @CuzickA @Hammond-Kosack
Could you have a look at the above, we can discuss the options on a future call.
It might be possible to do some of these using a more specific evidence: For example cell growth assay using RNAi knockdown (this does not exist, but you can request new evidence types that combine 2 existing evidence codes)
Note that the evidence code selection that you use will be configured based on the major types of experiments performed in your community. You will want to keep this to a fairly manageable list to begin with, as it is in a dropdown.
Future options include i) different evidence code sets configures for different species ii) Full evidence code ontology (we decided to keep it simple but you might want to make more options avaiable)
If you have any examples where the allele construction method is important, for the experimental outcome but does NOT affect the expression level (rather the allele type), we will handle these in a different way so it is a separate discussion.
PHI base would like to be able to specify (in some cases), the methodology of the allele creation, as it can affect the expression level The example provided was "promoter knockdown" vs. RNAi. This is sort of a special case.
At present we have only
"knockdown"
One possibility would be to allow an extension of this menu by configuration
"promoter switch-off" "RNAi knockdown "
Note that you can also do this:
So you can label the construct as an RNAi construct, even though it is stored as a WT allele with reduced expression.
If they perform a control experiment to show absence of RNA, you can do so like this
It might be better to have an option to report if the Knockdown" is known to be "partial" or "complete" in the interface when we capture "knockdown" instead.