[Taxacom] The difference

[Karl Magnacca]

On 26 Oct 2007 at 15:58, Richard Zander wrote:
> After 10 years of mulling over the difference between morphological and
> molecular phylogenetic analysis [...]

I think you left out one of the important differences, which is the role 
of the scientist in the analysis.  A major issue in morphological 
phylogenetics (both advantage and disadvantage) is that the person doing 
it leaves out characters that are invariant, uninformative, highly 
variable within taxa, or obviously widely convergent.  While this 
results in a better data set, it also allows for subjectivity that can 
influence the outcome of the analysis.  Molecular analysis is less 
subject to this (one can pick which genes to use, but not which bases; 
though that comes back to your comments on size equivalence), but 
results in the problem that tons of garbage is included.  Part of the 
thing about comparisons between numbers of characters in morphological 
vs. molecular data sets is that typically >70% of molecular characters 
are uninformative, and of the remainder a much higher proportion are 
homoplasious.

Karl
=====================
Karl Magnacca
UC-Berkeley, ESPM Dept.
137 Mulford Hall #3114
Berkeley, CA 94720
http://nature.berkeley.edu/~magnacca/