Paphs and DNA analysis...

Discussion in 'Taxonomy' started by bench72, Apr 20, 2009.

Slippertalk Orchid Forum

Help Support Slippertalk Orchid Forum:

  1. Apr 20, 2009 #1
    is anyone doing the same type of DNA analysis on Paphs as they did with the Laeliinae? might not be the bible of classification but might be nice to see what this shows as far as relationships between various closely related species and esp those which are still pretty contentious e.g. kolo vs topperi; phil vs roebelenii, etc etc...
     
  2. Apr 20, 2009 #2

    emydura

    emydura

    emydura

    Well-Known Member

    Joined:
    Oct 4, 2008
    Messages:
    6,763
    Likes Received:
    74
    Location:
    Canberra, Australia
    They have also carried out a genetic analyses of Dendrobiums which has resulted in a fairly large restructure.

    You wonder why it hasn't been done on Paphs. Someone did post a comment a while back that there were some problems in carrying out a DNA analyses on Paphs. I'm not sure why and if it is true. I look forward to reading the responses to your query.

    David
     
  3. Apr 20, 2009 #3
    When my son was in college, he did a research project with Dr. Ken Cameron. He analysed some genes on multifloral paphs...he proved to me that my sanderianum was really P of E...and that my roth wasn't a real roth either....
     
  4. Apr 20, 2009 #4

    SlipperKing

    SlipperKing

    SlipperKing

    Madd Virologist

    Joined:
    Sep 16, 2007
    Messages:
    18,704
    Likes Received:
    97
    Location:
    Pearland TX
    Double bumper!
     
  5. Apr 20, 2009 #5

    TyroneGenade

    TyroneGenade

    TyroneGenade

    mad scientist

    Joined:
    Sep 15, 2008
    Messages:
    2,217
    Likes Received:
    10
    Location:
    Johnson City, TN
  6. Apr 20, 2009 #6

    emydura

    emydura

    emydura

    Well-Known Member

    Joined:
    Oct 4, 2008
    Messages:
    6,763
    Likes Received:
    74
    Location:
    Canberra, Australia
  7. Apr 20, 2009 #7

    TyroneGenade

    TyroneGenade

    TyroneGenade

    mad scientist

    Joined:
    Sep 15, 2008
    Messages:
    2,217
    Likes Received:
    10
    Location:
    Johnson City, TN
    Hello,

    This study was at the level of the species available at the time. Such a study would reveal the sectional leveling as well. As many as 100 species were used for this study. Seems that only one individual from each species was used in the study.

    The type of detailed work Bench72 hints to does not seem to have been done, but the basic phylogeny is there. It would be straightforward for enterprising people to use the primer sequences published in the article, and use them to generate sequences for other populations. Of course, this wouldn't really tell you whether philippinense and roebelenii are different species. It would only tell you that the "philippinense" types cluster together (share a common ancestor) and the roebelenii types cluster together (share a common ancestor). You would get the same result if you compared the human poplutions of Terra del Fuego and St Helena to each other. This doesn't mean they seperate species...

    On the up side, if one performed such a study and found that the philippinense, roebelenii and laevigatum do not cluster but that supposed laevigatum types pop up clustered with philippinense and roebelenii then we can say that they probably don't represent species, but just examples of normal philippinense variation.

    A genetic study by itself is nice to disprove a species, but useless in itself in proving a species. You need to back up genetic data with other features which can be used to differentiate it. Of course, if you all the philippinense types from all the Philippines cluster together and among them is a island of roebelenii keeps clustering together as a defined group (that is, lying basal or derived to/from the philippinense group) then you have a strong reason to say "well hang one here, look, these roebelenii a distinct reproductive entity, ipso facto, a species sensu Ernst Mayr's biological species concept" But given how orchids seem to freely hybridize naturally the odds of this are nil.

    This would make for a very interesting time for some enterprising graduate student with nothing better to do with his time that run about a hot steam jungle collecting leaf samples and then sitting around idly while the DNA sequencer does it work and then reading said sequences... Boring...

    tt4n
     
  8. Apr 20, 2009 #8

    PaphMadMan

    PaphMadMan

    PaphMadMan

    phytomanic

    Joined:
    Feb 12, 2008
    Messages:
    2,037
    Likes Received:
    2
    Location:
    Madison, Wisconsin USA
    A cladogram (think of a family tree) showing the placement of species involved in that study is available here:

    http://www.slipperorchids.info/taxonomy/index.html

    It is a start, but there are other ways to approach DNA or RNA analysis as well, and more data is needed to consider distinctions at the level of species vs. variety.
     
  9. Apr 20, 2009 #9

    Ernie

    Ernie

    Ernie

    Guest

    Very nice. Thanks for linking. The Paph portion of the concensus tree seems to collapse a little in the Barbata, but who cares about those. :) Interesting that all of the Pardalos (lowii etc) are sister to adductum- as opposed to being sister to all of the Coryos. The spaghetti-petaled Coryos group together separate from the less curly, fatter-petaled ones- neat. I guess I always thought of the Parvis as derived, but I guess not. They are pretty similar in appearance to the cyps, so that figures I suppose.

    Yes, one has to interpret this cautiously from the species standpoint. Only one exemplar was used per species, so there's no telling how it would shake up if several of each were used as was mentioned with the phil/laev/roeb example given above. It is a decent show of support for the sub-generic, supra-specific groups (ie sections, subgenera) though. I could see a grad student tackling each section/subgenus doing several exemplars fo each species. Could even use the same ITS I'd guess.

    -Ernie
     
  10. Apr 20, 2009 #10

    Leo Schordje

    Leo Schordje

    Leo Schordje

    wilted blossom

    Joined:
    Aug 22, 2006
    Messages:
    2,462
    Likes Received:
    0
    Location:
    NE Illinois
    There is a serious problem with DNA based cladistics, and that is the DNA people think of DNA as being the "ultimate authority", and tend to be weak in their own use of clasical taxonomy. Unfortunately, if their taxonomy is weak, they can end up with completly false cladistics simply by the inclusion of a single mis-identified species being used to do the study. One of the valid cristicisms of the current studies in cattleya alliance is that it is likely that more than one of the reference speicies used were misidentified. In particular one taxonomist I know is very skeptical of the recent Oncidium / Odontoglossum alliance study because several specimens used were incorrectly identified.

    Without proper classical taxonomy all these DNA studies are nothing but junk science. The look authorative science without meaning a damn thing.

    Think about it. If one of the specimens labelled ODM was really an ONC, is it any surprise that the computer would say all ODM are really ONC?

    Before one accepts any of these DNA cladistics proposals one really needs to look at the taxononomy of the exact specimens used.
     
  11. Apr 20, 2009 #11

    Ernie

    Ernie

    Ernie

    Guest

    Yep, Leo. For my catfish studies, I had to key out each specimen myself before I could use it. Corydoras collections in the USNM, U Mich, American Museum NH in NY are all terribly misid'ed. I can certianly see the same happening with ANY taxa.

    -Ernie
     
  12. Apr 20, 2009 #12

    NYEric

    NYEric

    NYEric

    Well-Known Member

    Joined:
    Sep 14, 2006
    Messages:
    47,168
    Likes Received:
    83
    Location:
    New York City Apartment
    Then why wouldn't a DNA study be combined w/ a taxonomy study!? Or does that make too much sense! :mad:
     
  13. Apr 20, 2009 #13

    Lance Birk

    Lance Birk

    Lance Birk

    Well-Known Member

    Joined:
    Jul 29, 2006
    Messages:
    182
    Likes Received:
    0
    It would make sense if it made sense.

    Trouble is, it doesn't make sense.

    Question is........... who do we elect supreme being who will determine just exactly Which plant or animal is THE true species?


    When you decide that, then you have a basis from which to work.


    DNA sampling for species will, maybe, yield a roadmap from where it came. In a perfect world.


    Trouble is.......... we seem to want to know how to separate one species from another so we can tell them apart.

    We haven't gotten that far yet....... especially when latest tests show humans and chimps are within a hair's breadth of being the same.

    Go figure.
     
  14. Apr 20, 2009 #14
    ok, so if i can figure this out right.. and i will use phil/roeb/laevi/palawanense as an example..

    1. the process of genetic study within the species might be boring... ok, so the fact that i'm liking the idea means I really need to get a life! :D

    2. the plants have to be classified correctly first to work out whether the DNA sequence that arises matches the expected result... so if people are using length of petals as a guide to differentiate, then if the results show that plants with short and long petals intermingle, then that would disprove that theory. (yes? no?) (check out the glanduliferum A vs B.. is that a process of speciation? could this be happening in the phil/roeb group?)

    sorry for all the dumb ass questions....
     
  15. Apr 21, 2009 #15

    SlipperFan

    SlipperFan

    SlipperFan

    Addicted

    Joined:
    Jun 10, 2006
    Messages:
    43,301
    Likes Received:
    11
    Location:
    Michigan, USA
    Not dumb at all. Keep probing.
     
  16. Apr 21, 2009 #16

    PaphMadMan

    PaphMadMan

    PaphMadMan

    phytomanic

    Joined:
    Feb 12, 2008
    Messages:
    2,037
    Likes Received:
    2
    Location:
    Madison, Wisconsin USA
    What this kind of DNA testing can tell you with almost absolute certainty is, of the plants you actually test which ones are most closely related, and in relative terms how recently they diverged from common ancestors and in what order.

    One thing it can not tell you is where to draw the line between the most closely related species. If your 4 plants labeled phil, roeb, laevi and pala are all more closely related to each other than they are to any other plants tested, then they could be considered the same species, but wouldn't have to be. Species are not defined that way. If one of them turned out to be more closely related to the plant labeled glanduliferum than to the other 3 then that one could not be considered the same species as the other 3 unless you wanted to consider all 5 the same species, including glanduliferum.

    Another thing it can not tell you is if any of your plants are labeled correctly, or are truly representative of the type. That's where the classical taxonomy has to come in, and it is essential to know where each plant came from.

    And among all of this, it is worth remembering that species is an artifical concept we try to impose on the plants, not one that nature defines. In many cases there are several different ways the lines could be drawn without being wrong.
     
  17. Apr 21, 2009 #17

    labskaus

    labskaus

    labskaus

    Well-Known Member

    Joined:
    Jul 25, 2006
    Messages:
    1,830
    Likes Received:
    1
    Location:
    Germany
    I found the work by Cox et al. very interesting, but always thought of it as a preliminary or initial study. They used only on gene in their analysis, and the Laelia/Sophronitis story told us that`'s not good. For slippers, I'd like to see a study with 4 or six genes of different resolution power involved.
    Including more than one sample of éach species also is a good idea. Actually, would be essential since how would you know inter-species borders if you have no ideas about intra-specific variation?
    In Bacteriology, people use the (living) type strain of the taxa in question in their analyses. Well, here the situation is different, dried herbarium material is difficult to work with and of limited quantity, and does not even exist for all species. Use of the bacterial type strain does not even safe people from misidentification (contaminated cultures), but the use of proper type strains in molecular studies provides a pretty nice frame for taxonomic work. Nowadays, it is mandatory to provide at least one (16SrRNA) sequence with the description of a new bacterial species and I believe that's a good idea for plants as well.
     
  18. Apr 21, 2009 #18

    Roy

    Roy

    Roy

    Well-Known Member

    Joined:
    Jun 9, 2006
    Messages:
    3,260
    Likes Received:
    1
    Location:
    Halls Gap,Western Victoria, Australia
    I had a conversation with a notable Taxonomist a while ago regarding DNA testing on orchids. He suggested that the testing of Paphs was totally inconclusive as various clones of the apparent same species tested differently and a lot more work was required to refine the process greatly before and positive conclusions can be drawn. As to Lances discussion, a colleague of this Taxonomists did a DNA study of a large number of Phalaenopsis species in their country of origin, for their Phd. The plants selected were checked as per taxon' specs against recorded info and deemed correct. They all proved to be hybrids, even the ones collected from the wild. Because of evolution & natural hybridization potential between similar clones growing in a given area maybe we will have to consider the potential of "natural hybrids" being given species statis.
    One orchid that would be worth testing, both wild collected and man-made outcrossings is Paph. rothschildianum. I believe its the only Paph of the type in the area where natural hybridisation could occur. Paph dayanum grows there as well but I don't believe has produced seedlings with roth' in nature ( yes / no ?? )
     
  19. Apr 21, 2009 #19

    NYEric

    NYEric

    NYEric

    Well-Known Member

    Joined:
    Sep 14, 2006
    Messages:
    47,168
    Likes Received:
    83
    Location:
    New York City Apartment
    Well OK, you're saying it's better to be indiscriminate and not make a choice than to have someone/group who is very knowlegable in the taxonomy of species, [since they have been named and exist in library and live collections someone can use these for a reference], assist in the dna classifications!? :rolleyes: BTW I don't think that would make anyone a "supreme being"!
     
  20. Apr 21, 2009 #20

    Leo Schordje

    Leo Schordje

    Leo Schordje

    wilted blossom

    Joined:
    Aug 22, 2006
    Messages:
    2,462
    Likes Received:
    0
    Location:
    NE Illinois
    A species is defined by one thing only, the type specimen deposited in a herbarium. That ONE specimen and its description ARE the ONLY reference that counts. "Full Stop." to quote Dr Guido Braem. Not to argue the definitions of species, but in taxonomy the process starts with a single specimen being given a name. Then you compare all subsequent specimens to see if they fit the "same species" as the named type. If the type speciemen was a typical example, about average for the population - this is easy. If the type specimen was unique, well that creates problems. But the end result is ONE specimen serves to define the species name.

    The only reliable way to build a DNA reference database is to take samples from the actual type specimens and extract the DNA from the plant actually used to do the original description.

    For example if this were possible, we could figure out things like which plant Mr Low really wanted the name chamberlainianum to apply to and which he wanted victoria-regina to apply to. They were published in the same bulletin, at the same time without drawings or watercolors. I speculate that if you could get the DNA from the 125 year old type specimens that are flattenned, colorless, fragmented messes in their dried up condition you might be able to sort the mess out.

    Problems; Extracting DNA from a type specimen usually wouldn't be allowed because the sample taken is destroyed in the process, not many curators would be happy about that. Also Many of the type specimens were pressed and dried over 100 years ago. The DNA may no longer be intact. Type specimens are scattered all over the world, the travel budget would be huge. And a fair number of type sheets have been lost over the years due to fire, flood, war & other disasters. These obstacles may make it impossible to actually do more than a partial survey for any group of plants.

    So be suspicious of any certainty credited to DNA studies. It is not like the fanatsy crime shows like CSI Miami, it just doesn't work that way.
     
    Last edited: Apr 21, 2009

Share This Page



arrow_white