Publishing on human origins

[Richard.Zander at MOBOT.ORG]

-----Original Message-----
From: John Grehan [mailto:jgrehan at tpbmail.net]
Sent: Sunday, May 30, 2004 7:57 AM
To: Richard.Zander at mobot.org; TAXACOM at LISTSERV.NHM.KU.EDU
Subject: RE: [TAXACOM] Publishing on human origins


At 01:48 PM 5/28/04 -0500, Richard.Zander at mobot.org wrote:


No, we are comparing apples and apples. Because the analytical results with
morphological and molecular data sets generally support each other,
morphological traits and molecular traits have much the same probabilistic
utility in statistical evaluations.


Would I be correct to interpret this as saying that because the two data
sets 'generally' agree, one can use either as a substitute for the other? If
that is the case does the best tree comes only from combining data where
each set does not agree?

[[R.Z.:] ] One can use one as a substitute for the other where they agree.
Sure. Where they don't agree, you either need an explanation that allows
both to be reasonable or be able to explain away (delete) one of them.

The best tree may be no tree. One should not act upon something merely
because it is "best" of several likely explanations, especially if the risk
is large. Combining data is only acceptable if you have evidence for or at
least no evidence against the chance that the data is derived from two
different processes, e.g. lineage sorting, or some kind of genetic
combinatorial confusion.




You may not have as high confidence intervals in morphological results, but
confidence intervals can be combined. E.g., rolling a hexagonal die four
times increases the chance of getting a "one" to 0.52, and a single flip of
a coin may be substituted. Two contiguous internodes at 0.78 CI can be
replaced by one at 0.952. (Got a paper in review with this.) So with
morphological data, a poorly supported ((A,B)C(D,E)) can be replaced with a
well supported ((A,B)(D,E)). Then test it with molecular data.



I guess the notion that a morphological arrangement is to be 'tested' by
molecular data is problematic for me.

[[R.Z.:] ] We test all data against all data, and throw out only that data
that we can explain away. If you map molecular data on a morphological tree
and map morphological data on a molecular tree, perhaps, if they disagree,
one or the other might be rejected as due to chance alone given the risk of
accepting it. Or maybe not, in which case we are stuck in lieu of more data.


John Grehan




______________________
Richard H. Zander
Bryology Group
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166-0299
richard.zander at mobot.org <mailto:richard.zander at mobot.org>
Voice: 314-577-5180
Fax: 314-577-9595
Websites
Bryophyte Volumes of Flora of North America:
http://ridgwaydb.mobot.org/bfna/bfnamenu.htm
<http://ridgwaydb.mobot.org/bfna/bfnamenu.htm>
Res Botanica:
http://ridgwaydb.mobot.org/resbot/index.htm
<http://ridgwaydb.mobot.org/resbot/index.htm>
Shipping address for UPS, etc.:
Missouri Botanical Garden
4344 Shaw Blvd.
St. Louis, MO 63110





______________________
Richard H. Zander
Bryology Group
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166-0299
richard.zander at mobot.org < mailto:richard.zander at mobot.org
<mailto:richard.zander at mobot.org> >
Voice: 314-577-5180
Fax: 314-577-9595
Websites
Bryophyte Volumes of Flora of North America:
http://ridgwaydb.mobot.org/bfna/bfnamenu.htm
<http://ridgwaydb.mobot.org/bfna/bfnamenu.htm>
Res Botanica:
http://ridgwaydb.mobot.org/resbot/index.htm
<http://ridgwaydb.mobot.org/resbot/index.htm>
Shipping address for UPS, etc.:
Missouri Botanical Garden
4344 Shaw Blvd.
St. Louis, MO 63110


-----Original Message-----
From: John Grehan [ mailto:jgrehan at TPBMAIL.NET <mailto:jgrehan at TPBMAIL.NET>
]
Sent: Friday, May 28, 2004 7:03 PM
To: TAXACOM at LISTSERV.NHM.KU.EDU
Subject: Re: [TAXACOM] Publishing on human origins


A probability test might be suitable if one is comparing apples with
apples. In the case of morphological characters vs DNA sequences I don't
thing that would be the case.

snip

John