Standartisation of names of properties

On Mar 6, 2009, at 1:46 PM, Mojca Durjava wrote:

****
The problem is, that we don't have any agreement how we will introduce data, each one of us can determine his/her own system. Maybe it would be better to have an agreement on which properties we will use (name, unite,...), otherwise the introduced data will be difficult to retrieve.
****

The question is open to discussion.

My first remark:

1) The database will (later on) provide possibilities to AUTOMATICALLY convert units. Thus, introduce data using Units reported in the article.

2) This does not solve problem of standardization of names. First, we already have an explosion of them, e.g. 22 properties contain IC50.

LogIC50 PR
LogIC50 ERβ
LogIC50 ERα
IC50 mussel

etc.

In fact, all of them correspond to IC50, but are different with respect to the Targets (protein).

Another example is provided in the second topic (EC50 with different Species and administration routes). So, what are your suggestions? Would you like to have multiple properties, which also include in the name conditions, species, OR, should we have only few properties but with obligatory conditions?