[Taxacom] Character weighting

[Richard Zander]

When phylogeneticists weight all traits alike, is this objective? It's commonly accepted as "wrong" but we kick clods sheepishly and glance about furtively when we admit this. I say "we" since the Laplacian (I think it was who said when nothing is known about weighting, equal weighting is better in some way) fault is everyones. 

I have a paragraph I wrote in an early, early paper:
"Weighting highly those characters with best fit as determined after the parsimony analysis, may be done in various ways: Ladiges et al. 1989 used the consistency index (c) over the patristic unit character length (Farris 1969) and also the product of character consistency and character retention index as was determined in Hennig86 (Farris 1989). Inasmuch as all characters theoretically have different phylogenetic importance (as per discussion by Farris 1983: 11), equal weighting prior to the exact algorithm is equivalent to arbitrary weighting (Swofford & Olsen 1990: 464). Thus, the more the characters are (reasonably) weighted differently, the less arbitrary is their weighting; Kluge and Farris (1969) recommended weighting by degree of variability of a character within OTU's, this being an estimate of the rate of evolution of that character. No method now exists to recommend the best weighting by character fit (Farris' 1969 example works for an artificial data set with a known random element)."

I can give references if asked. So there are objective (as opposed to certain or true) methods of weighting on the basis of evolutionary theory. Convergence is an evolutionary theory. So: In one paper I increased weighting of morphological traits NOT associated with convergence of a particular plant organ to force convergence-related traits high in the tree. All attempts at recovering a one-time historical event are heuristic, and one should not confuse "objectivity" in this context with objectivity in analyzing a universal in physics.

We all get funny results, and those call for considered and reasoned judgment, which can be called "subjective" by those promoting automatic classification or automatic evolutionary relationships, but is the basis for true advances in the field.

Note that when you add comments about 3rd codon positions and problems with that, you are correct but off subject. 

*****************************
Richard H. Zander 
Voice: 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
Non-post deliveries to:
Missouri Botanical Garden, 4344 Shaw Blvd., St. Louis, MO 63110
*****************************


-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Jason Mate
Sent: Wednesday, August 19, 2009 4:13 PM
To: Taxacom
Subject: [Taxacom] Character weighting


I´ll ask you if you tell me how you arrive at it. Thanks to Pierre for succintly putting an end to "molecular data is phenetic" mantra. I will get into hot water with many but how do you weigh your characters objectively? I can see objectiveness in weighing based on codon position or transitions vs transversions or according to the aminoacid encoded. I can also see the objectiveness of character reweighting according to some measure of internal congruence (the idea being that you encourage the signal; of course doing this between datasets encourages the selection of an "average" phylogeny which might not be the species tree). But even these can arrive at "funny" results. In the case of extreme weighting (i.e. effectively 0 weight to 3rd codons) you can loose all resolution.