How to represent a genealogical tree with complex ancestry

Discussion in 'Taxonomy' started by Secundino, Dec 11, 2015.

Slippertalk Orchid Forum

Help Support Slippertalk Orchid Forum:

  1. Dec 11, 2015 #1

    Secundino

    Secundino

    Secundino

    Adorable Stud

    Joined:
    Dec 31, 2012
    Messages:
    2,323
    Likes Received:
    0
    Location:
    Spain
    I'm trying to find out the ancestry of one of Yamamotos Dendrobiums (Dendr. Comet King) and so far have managed to find the maternal ancestors (89 times crossed with nobile ... it is really complex!) This - well documented at rhs - is the easy (but time consuming) part.

    But how on earth do you represent this, in a visual appealing but space saving way? Has anyone here done it and can help?
    Thank you!
     
    Last edited: Dec 11, 2015
  2. Dec 11, 2015 #2

    gonewild

    gonewild

    gonewild

    Grower

    Joined:
    Jun 26, 2006
    Messages:
    5,165
    Likes Received:
    1
    Location:
    Puerto Maldonado, Peru
    You have to make a Family Forest. :)
     
  3. Dec 11, 2015 #3

    Ozpaph

    Ozpaph

    Ozpaph

    Well-Known Member

    Joined:
    Jan 15, 2011
    Messages:
    14,736
    Likes Received:
    61
    Location:
    Brisbane, Australia
    why not use orchidwiz?
     
  4. Dec 11, 2015 #4

    gonewild

    gonewild

    gonewild

    Grower

    Joined:
    Jun 26, 2006
    Messages:
    5,165
    Likes Received:
    1
    Location:
    Puerto Maldonado, Peru
    Nobile used 89 times. Over howw many years?

    Nobile is a very common ancestor but unless the same nobile clone was used each time then each nobile use needs to be treated as a separate genetic entry. but then that complicates the tree with parentage of each nobile clone used. You need a big piece of paper!

    If you want to illustrate it visually you might use a 3D approach and connect each crossing together with strings of different colors.
     
  5. Dec 12, 2015 #5

    Secundino

    Secundino

    Secundino

    Adorable Stud

    Joined:
    Dec 31, 2012
    Messages:
    2,323
    Likes Received:
    0
    Location:
    Spain
    I'm on my way ... as I have no access to Orchidwiz, I'm compiling myself. That's not the problem. The difficult step is how to represent it.

    The first half is nearly done - have a look ... as to now, the oldest cross is Dend. Ainswworthii from 1874 by Dr. Ainsworth. But Veitch and Sanders have their share, too, until in the sixties Yamamoto begins an very extense/intense breeding program. No wonder his plants look like they look!

    [​IMG]
     
    Last edited: Dec 12, 2015
  6. Dec 12, 2015 #6

    Secundino

    Secundino

    Secundino

    Adorable Stud

    Joined:
    Dec 31, 2012
    Messages:
    2,323
    Likes Received:
    0
    Location:
    Spain
    And this is the culprit:
    [​IMG]
     
  7. Dec 12, 2015 #7

    SlipperFan

    SlipperFan

    SlipperFan

    Addicted

    Joined:
    Jun 10, 2006
    Messages:
    43,582
    Likes Received:
    4
    Location:
    Michigan, USA
    What a forest for that flower!
     
  8. Dec 12, 2015 #8

    naoki

    naoki

    naoki

    Well-Known Member

    Joined:
    Apr 7, 2012
    Messages:
    2,057
    Likes Received:
    14
    Location:
    Fairbanks, Alaska, USA
    Looks like that you figured out how to plot. Graphviz could be a good program if you have a huge pedigree (it is probably an overkill, though).

    http://www.graphviz.org

    If you know R, there are also several pedigree type packages (kinship2 and Genetic Analysis Package), which can make pedigree figures.
     
  9. Dec 12, 2015 #9

    Secundino

    Secundino

    Secundino

    Adorable Stud

    Joined:
    Dec 31, 2012
    Messages:
    2,323
    Likes Received:
    0
    Location:
    Spain
    Somehow - but I cannot copy and paste complex elements - 'sub-trees' like Lady Colman e.g. - that appear more than once and I'd need to extend the page on the right side, but the program just allows the left side ... my pc skills are rudimentary, and I'm afraid I can not use your links, naoki, for the same reason, but thank you all the same.

    But it has been a incredible piece of understanding; 12 generations extending to the beginnings of orchid hybridisation, six species involved (so far - still on the first parent), the recurrent use of some early hybrids (or special clones) which must have been outstanding for some reason (I still have to find pictures of them) and it seems it takes this route to fix the characteristics that distinguish a Yamamoto Dendrobium, stout and strong growth, lots of long lasting flowers with heavy substance, big, colourful, upright, round - genetic and lost of feeding.
    What is lost - as in all flowering plant that have undergone a breeding program like this - is the dainty beauty, the frail peduncules, the nearly transparent appearance of the petals, the 'wild look' of course.
     
  10. Dec 13, 2015 #10

    PaphMadMan

    PaphMadMan

    PaphMadMan

    phytomanic

    Joined:
    Feb 12, 2008
    Messages:
    2,066
    Likes Received:
    2
    Location:
    Madison, Wisconsin USA
    From OrchidWiz I can tell you the completed tree will cover 13 generations, 8 species, and 51 different hybrids involved along the way, going back to Den. Ainsworthii in 1874 as you noted. Den. nobile accounts for a theoretical 45.7% of the ancestry, and I think I'm interpreting this correctly, appearing in the complete family tree 215 times. You've still got some work to do.

    The display you show is using a definite row and column grid structure. For me that would suggest using Excel or another spreadsheet program, but my mind tends to go to that because I use it constantly at work. That would allow you copy and paste functionality for repeating branches of the tree, no real limit on rows and columns in the display, and plenty of formatting and drawing tool choices. It might be interesting to color code each species when it appears and see if that highlights any patterns or trends in the breeding.
     
  11. Dec 16, 2015 #11

    Ozpaph

    Ozpaph

    Ozpaph

    Well-Known Member

    Joined:
    Jan 15, 2011
    Messages:
    14,736
    Likes Received:
    61
    Location:
    Brisbane, Australia
    That is a VERY good idea - it would make it visual more readable by adding a 'quantitative' element to the tree.
     

Share This Page



arrow_white