[Taxacom] Pasimony and base alignment

Richard Zander Richard.Zander at mobot.org
Wed Aug 30 10:06:08 CDT 2006


I found the following paper recommended by a past post on Taxacom
illuminating:

http://research.amnh.org/scicomp/pdfs/wheeler/Giribet_etal2002b.pdf

We all know there are problems with alignment of molecular sequences,
and that there is a famous quotation by someone that phylogenetic
analysis is primarily a problem in alignment, not tree-making. This
paper points out that aligning sequences is done by optimization, and
there may be plenty of slightly less reasonable alignments than the
optimal one. The author recommends that reanalysis using these slightly
less reasonable alignments should be done to see if the tree or at least
a branch arrangement of interest is robust to such changes or not.

Given that trees are commonly presented nowadays with high Bayesian
posterior probabilities on branch arrangements, or the equivalent in
bootstrap values, and given that the additional uncertainty of
alignments requires joint probability (multiply the chance that the
branch arrangement is correct by the chance that the alignment is
correct when not being correct gives a different answer), then one might
advance the idea that maximum parsimony alone as a measure of
reliability (converges on the truth, scientists always accept the
simplest theory, least falsifiable, etc.) has been shifted from
tree-making to alignment. The chance that alignment is correct for any
particular branch arrangement is usually hidden from the reader, but
should be given when realignment changes the result. 



******************************
Richard H. Zander 
NOTE: NEW PHONE NUMBER 314-577-0276
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166-0299 USA
richard.zander at mobot.org
Web sites: http://www.mobot.org/plantscience/resbot/
and http://www.mobot.org/plantscience/bfna/bfnamenu.htm
For FedEx and UPS use:
Missouri Botanical Garden
4344 Shaw Blvd.
St. Louis, MO 63110
******************************

_______________________________________________
Taxacom mailing list
Taxacom at mailman.nhm.ku.edu
http://mailman.nhm.ku.edu/mailman/listinfo/taxacom



More information about the Taxacom mailing list