Resin is what breeding is all about, the other traits are all secondary... now I believe the genetic makeup of resin must be very complex, so it may be impossible to get a daughter plant with exactly the same resin (high) as the mother plant.
Yep exactly what you define as a trait will determine whether it is going to follow basic mendelien genetics.
As you said resin production is not a simple trait, it's not just about presence/absence of THC, it's the amount not only of THC but other structurally related compounds, all of which will be quantitative traits (ie exist as a continuum). I'm not sure there has been a lot if any research done on the cannabinoid biosynthetic pathway.
Depending on how the pathways is (constructed) there is going to be a huge number of potential ways a difference at the DNA level could manifest itself in the functioning of the pathway(s) and hence expression of the "trait". These could Be:
*mutations in the biosynthetic gene coding region, producing enzymes with differential activity whose combination would produce changes in the flux through the pathway(s) and hence a change in the balance of metabolites within the pathway(s) (ie milieu of cannabinoids)
* mutations that change the abundance of any of these enzymes (activity tends to be proportional to abundance). These could be within promoter/enhancer/terminator regions in the up/down stream of the gene, in the 3/5` UTR of the resulting mRNA transcript or even within the coding region itself due to codon usage effect.
* mutation in genes that control the production of any of these enzyme ie transcritpion factors, mRNA processing, chromatin remodelling, translation, processing sorting enzymes.
*mutation in proteins that post-translationally modify the biosynthetic enzymes, changing activity/localisation/protein-protein interaction etc...
All we need to do is sequence the cannabis genome, construct a SNP chip, perform a mass breeding program, use HPLC to determine metabolite levels, MS-MS to measure protein abundance/modification, mRNA microarray chips etc... and we could sort it out ;)
