Create experiment table

[petermr]

Some papers describe several profiles, often wrapped as separate tables. The structure of the table should be preserved or we lose critical information. Here is an example (Thanks to Shruthi):
#55 (comment)

Table 1. Percentage composition of yellow-orange Algerian Lantana camara leaf essential oils during 2007.
No. Componenta RIb RIc Feb. Apr. June Aug. Oct. Dec. Mean SD
1 2-Hexanal 857 846 tr tr tr tr tr tr – –
2 α-Thujene 921 924 tr tr tr tr tr tr – –
3 α-Pinene 926 932 0.8 0.3 0.8 0.2 0.6 0.1 0.5 0.3

Here we have 7 profiles (Feb. Apr. June Aug. Oct. Dec. Mean SD) [we might miss Mean and SD]. The other 5 only really make sense if they are linked. So this is an experiment (I suppose it could be a ProfileCollection but I think Experiment is better.

Here's another in the same paper:

Table 2. Percentage composition of yellow-orange Algerian Lantana camara leaf essential oils at March over 3 years.
No. Component b RI a RIc 2007 2008 2009 Mean SD
1 2-Hexanal 857 846 tr tr tr – –
2 α-Thujene 921 924 tr tr tr – –
3 α-Pinene 926 932 0.2 0.1 0.2 0.2 0.0
4 Camphene 932 946 tr tr tr – –

here there are 3 main profiles, all with the same metadata except date.

PROPOSAL:
experiment would become a new table and profile with one-to-many profiles. Should every profile have an experiment (which would hold the experimental data)??

a profile would know which experiment if belonged to.

We'll need to think about this.

[petermr]

We need some samples of papers with multiple profiles to see how they can be represented.
How often are multiple profiles reported? What are the reasons?