[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

re: Bootstrap question; & a v? for Dan Varner

David Peters (davidrpeters@earthlink.net) wrote:

<The problem I've been having is replicate hangup, where a single replicate
takes several hours with 99 to go. Very discouraging. Is there any sin to
divide and conquer? i.e. splitting up the big cladogram into smaller parts for
analysis? That appeared to give the right answer.>

  If one splits the cladogram up, then I think finding a bootstrap value for
the whole becomes less useful. However, Mishler has advocated splitting a large
tree up for a variety of reasons, including

  1. reducing rampant and bootstrap-defying homoplasy
  2. reducing run time of analysis
  3. allowing more easy manipulation of the dataset so that emmendations are
more obvious.

  Sereno has done this for his large analysis in 1999, and Carpenter did this
in 2001 for Ankylosauria. For example, we can split certain groups off and
develop a set of codings in a smaller portion of the matrix to reflect them as
an outgroup test, whereas the ingroups would be such taxa being examined in
some detail, such as the drepanosaurs, or the protorosaurs, or
crocodylomorphans (or mesoeucrocodylians) with characters that are not relevant
in larger analyses, even if they appear in maybe two or more taxa. These
characters can be used in a broader, catch-all matrix comprise of major groups
and "problem" taxa. Like in Coelurosaurs, this is most non-maniraptorans, whose
positions tend to shift in LARGE matrices. Seeing the apparently non sequituur
positions some taxa can make with the appearance of only one taxon more or less
implies massive homoplasy and long-branch attraction affecting how taxa are
treated. Such variability may be noted is drastically different and likely
untrustworthy positions seemingly rampant in some of the Holtz, Mortimer, and
even my own tests. Just such a problem may be causing the Dromaeosauridae to
collapse in virtually every matrix published since 1999, where the increasing
variation and homoplasy in the taxa is rendering each clade pointless. Removing
non-maniraptorans and focusing on Maniraptora and one or two outgroups may help
resolve the tree (I'm just guessing).

  Remember the mantra: There is no "one true tree." No current phylogeny
machine is likely going to be able recover a phylogeny of life while either
rejecting reticulation and population-level groups of individuals, and without
a good ability to interpret with genomics and genetic drift (i.e., not every
mutation is an adaptation).


Jaime A. Headden

  Little steps are often the hardest to take.  We are too used to making leaps 
in the face of adversity, that a simple skip is so hard to do.  We should all 
learn to walk soft, walk small, see the world around us rather than zoom by it.

"Innocent, unbiased observation is a myth." --- P.B. Medawar (1969)

Do you Yahoo!? 
Yahoo! Small Business - Try our new Resources site