Graphing kavalactone concentration

[kavadude]

We throw around chemotypes a lot when talking about the effects we perceive a particular cultivar to have. This has always bothered me, largely because chemotypes only tell you the relative concentration of kavalactones. Now that @Deleted User is giving us absolute and not relative numbers for the kava we drink, there really ought to be a visual representation for kavalactone content.

Here is my first go, this is generated by a little javascript program. Once I get a representation that I like, I will release it in such a way that you can type in concentrations and it will give you a chart.

Right off the bat you can see the big difference in dihydrokavain and yangonin between these samples of Nene and Moi.

I'm also thinking of going back and performing word counts on particular kavas reviews (e.g. how much 'heady' occurs in reviews for a particular kava) and seeing if and how that relates to kavalactone content.

 

[Blippy5]

I love it! And I like the idea of a horizontal bar graph too. Maybe starting form highest values to lowest?

 

[kavadude]

Very possible. I'm sick of this graphing library for now, but I'll have a go at it tomorrow. One other thing I was hoping to represent with the donuts was getting the overall KL% and changing the overall size so that e.g. a kava with 20% would be much larger next to a 10% kava's chart.

 

[Steve Mariotti]

I love this. I was just the other day thinking about "kavagrams" which would visually show the chemotype fingerprint of a given kava in an easily understood format. That's exactly this.

 

[kavadude]

Yeah, that's what I'm going for. I want it to be easy to look at a kava's chart and eyeball whether it's something you might be interested in.

 

[Sean D.]

Very nice.

 

[Steve Mariotti]

Strength could be represented by height of the bar.

My thought was a six axis polar plot.

Here's a quick mockup:

Composed in Excel, then colored in Paint.NET

 

[Blippy5]

That's the most confusing graph I have ever seen

 

[sɥɐʞɐs]

 

[Steve Mariotti]

That's the most confusing graph I have ever seen

Then perhaps you've not seen this famous graph prepared by The Pentagon here in the states:

Great way to hide your information in plain sight.

 

[Blippy5]

@shakas - you just made that up!
@Steve Mariotti - that doesn't surprise me at all. That is totally an over enthusiastic desk jockeyjockey with very basic ms Word skills, and not enough real work to do.

 

[verticity]

Personally, I'd like to see the kavalactones in order, 1-6 on the bar, with horizontal length being percentage. Then you could stack them to compare.

I think it actually might be more informative to order the KL's by "nobility", so basically 423165
Noble kavas would appear skewed left, and tudeis skewed right, the difference being apparent at a glance.

 

[sɥɐʞɐs]

I like the hexagon chart, it combines the pie chart with the bar chart. then you could list the numerical chemotype along with it.

im not sure about arranging things by nobility, cuz i don't think that exists. nobility isn't a scale where some nobles are more noble than others. some nobles have a 245xxx chemotype and some tudeis have the same. similarly, there are some tudeis that appear to have a noble chemotype. there is a scale though, to how adulterated a noble has been with tudei...

the scale of nobility could hypothetically exist if every kava was tested for FK-B percentage....and/or any other mystery chemical that turns out to be a true culprit.

 

[verticity]

im not sure about arranging things by nobility, cuz i don't think that exists. nobility isn't a scale where some nobles are more noble than others. some nobles have a 245xxx chemotype and some tudeis have the same. similarly, there are some tudeis that appear to have a noble chemotype. there is a scale though, to how adulterated a noble has been with tudei...

the scale of nobility could hypothetically exist if every kava was tested for FK-B percentage....and/or any other mystery chemical that turns out to be a true culprit.

Strictly speaking, you are right. There is overlap between noble and tudei chemotypes. But statistically, some KLs (kavain and DHK) are correlated with nobility, and others (DHM and Methysticin) are inversely correlated. It's not an exact thing by any means though.

 

[sɥɐʞɐs]

Strictly speaking, you are right. There is overlap between noble and tudei chemotypes. But statistically, some KLs (kavain and DHK) are correlated with nobility, and others (DHM and Methysticin) are inversely correlated. It's not an exact thing by any means though.

definitely, strong correlations with certain chemotypes.
the main thing i wanna get across to anyone reading this, is that...as of yet, there is no reason to believe one noble is 'more' noble than the next. if you look at Deleted User's acetone color scale, the one with the lightest yellow hue is no more noble than the noble that's six spots down and a bit darker.
but if we tested all those kavas for FKB and the FKB %'s also matched up with the color order...then i think we could start saying there might be nobles that are more noble than the next. but the variation in the FKB content of noble kavas is probably so small it would make no real difference. ...perhaps one day we'll see

 

[verticity]

definitely, strong correlations with certain chemotypes.
the main thing i wanna get across to anyone reading this, is that...as of yet, there is no reason to believe one noble is 'more' noble than the next. if you look at Deleted User's acetone color scale, the one with the lightest yellow hue is no more noble than the noble that's six spots down and a bit darker.
but if we tested all those kavas for FKB and the FKB %'s also matched up with the color order...then i think we could start saying there might be nobles that are more noble than the next. but the variation in the FKB content of noble kavas is probably so small it would make no real difference. ...perhaps one day we'll see

True, as far as I know, the people of Vanuatu, who invented the concept of noble kava, don't distinguish "grades" of nobility. But the effects of noble kavas do differ based on chemotype. 245--- is going to be more sedating than 432--- for example. So maybe it would be better to call it the "heady-heavy" axis.

 

[sɥɐʞɐs]

I'm on board with the noble "heady-heavy" axis.
Heavy noble should be thought of as being equally as noble as light noble...until we have evidence that might prove otherwise.