Hybrid visualization

[reivilos]

Hi all,
This is to let you know I'm working on a hybrid visualization tool for fun.
If you've got ideas, suggestions, I'm all ears.
Olivier

One of the features would be parentage display.

Below is the case of Paph. Atlantis and Aladin.

This shows what is takes to remake both...

 

[Justin]

we need Tim to comment on this one.

 

[reivilos]

This is paph. Whitemoor:

 

[Ross]

Great idea! I like the colors added for species and primary hybrids.

Perhaps color coding of any grex that is used more than once in a complex cross might yield interesting results. For instance, Acteus (Leeanum x insigne) was used as a parent six different times in the breeding lines that led to the final plant, Paph Whitemoor (Dervish X FC Puddle): once on the FC Puddle side as a grandparent, four times to produce Grace Darling, one of the parents of Dervish, and once in the production of Bordube, the other parent of Dervish. It's also interesting to note that Dervish got two of its "shots" of Acteus through Paph Florence Spencer.

 

[reivilos]

It looks like this now:

 

[Dido]

Interesting

 

[tim]

To me it's reinventing the wheel...aq plus already does this...

When I look at a pedigree I'd rather have the neatly organized tree more common of pedigree information. The other problem with a more web-like view is confusion associated with things like insigne - your chart could be said to imply the same insigne was used 4 times, which of course it was not...

That being said if you prefer to look at registrations this way, go for it. Any way to get more people to look at pedigrees is a tremendously useful tool.

I think it's funny just to see these two charts - with boxallii separate from villosum, which in many early cases it was not, and this abiding problem with cardinal mercier, whose parentage is not in fact unknown, but is a re-awarded piece of Lathamianum 'Cardinal Mercier'...so many problems with early registration.

 

[reivilos]

Thanks for your input Tim. I wasn't
aware of AQ Plus. I only had a try at
OW and wasn't satisfied by it - at least
for paph.

I'm usure what you call the usual pedigree tree. If it's the genealogy tree
OW shows I find it annoying to see
multiple occurrences of a single
individual on the tree.

We no longer have a paph breeding tradition here in Europe - and especially in France - for several reasons. Though 95% of all species are readily available,
hybrids are scarce. Want a hangianum? Buy 10 tomorrow.
With the graph representation I know which paph cannot be remade in a lifetime!

Second thing I'd like to visualize is the intermediary steps. I believe breeding steps are sometimes dictated not by a desired improvement on a flower, but by
availability at a given time.

The data used is the RHS register as is... Of course it requires post processing.

 

[Dane]

Hi all,
This is to let you know I'm working on a hybrid visualization tool for fun.
If you've got ideas, suggestions, I'm all ears.
Olivier

One of the features would be parentage display.

Below is the case of Paph. Atlantis and Aladin.

This shows what is takes to remake both...

What if you had to just do the family tree tracing back the generations, in a tree like order

 

[jjkOC]

Hi Reivilos, this is an interesting way of visualizing parentages. I wonder in the tool you've made, how does it define the position of the names? I made a small pedigree tree just to figure out the origin of some multiflorals I was interested in and found I worked from the center outward using a more linear format.

Also, in your design I like the color coding of species, primaries etc, but it seems that in a web design it is a little less easy to trace pollen and pod parents or just which are the parents as your eyes have to scan for arrowheads.

In the pedigree for multiflorals, like you I was also trying to understand the logic of breeding, for example trying to figure out whether a new cross was a backcross (BC), an F1BC, F2BC or a 3WC etc. I was trying to find patterns of when you might see improvements (given equally superior parents) or injection of different traits.

I wonder if you could add a function to display this kind of information when right clicking/hovering over the names?

Just some thoughts. Thank you for sharing!

 

[reivilos]

Maybe it's more readable like this:

 

[jjkOC]

This is definitely more readable, at least for me... I forgot to mention that I like that you have dates for each hybrid!

 

[Drorchid]

I like the last graph the best, way more readable! One more piece of information that you can add, is next to the species note the percentage that is in the background of the final hybrid (Paph. Whitemoor); In this way you can see which species have the most genetic influence.

RObert

 

[reivilos]

I added the percentages. I'll add a web interface.