data on flavor molecules from OpenAg experiments
Paper published at https://doi.org/10.1371/journal.pone.0213918
we're manipulating environments and measuring the effects of those manipulations on flavor molecules.
These data files are measurements of volatile molecules as collected, measured, and identified by headspace-solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS, or GC-MS for short). Individual data files contain raw relative abundances of each measured molecule, which is correlated, but not equivalent to, concentration in plant tissue. The sample masses and treatment conditions are in the "Master list of data files" sheet: https://github.com/OpenAgInitiative/flavor-data/blob/master/Master%20List%20of%20Data%20Files.xlsx.
you have to think of flavor in two minds—one, what molecules we can smell (since smell is the most important part of flavor), but also (two) where those molecules come from, how they were made, and why they were made.
in the case of plant smells, volatile molecules are part of the plant's specialized metabolism, evolved to increase that plant's fitness under selection from a particular aspect of their environment,
Floral smells are an adaptation to attract pollinators.
fruit smells are an adaptation to attract organisms that will eat the fruit and disperse the seed inside.
leafy and herbal smells and flavors, on the other hand, are mostly adaptations to stresses: -UV light -water loss -insect predators -harmful microorganisms -heat -drought etc.
applying these stresses, in plants that are adapted to them, will increase production of volatiles associated with that stress.