Re: To Bob re: Macroevol. and N.S.

From: Paul Nelson (pnelson2@ix.netcom.com)
Date: Mon Jul 24 2000 - 21:24:34 EDT

  • Next message: James W Stark: "Spirit of truth"

    Tim Ikeda wrote:

    > Then again, one could take common ancestry to mean that
    > organisms share many parts from an earlier pool of components
    > (possibly not shared by all progenitors, but largely in common).

    One could say that, sure, but it's unclear what biological
    meaning it has. A "pool of components" is not a living
    thing. Functional cells may possibly exchange genes
    laterally -- a "pool of components" cannot.
     
    >> In the concluding section of his paper, Doolittle grapples
    >> with the consequences of his view for the standard definition
    >> of homology (i.e., similarity due to common ancestry).
    >
    > Paul, would you care to recount Doolittle's discussion in this
    > area and perhaps extrapolate to how it may pertain to things
    > such as mitochondrial phylogeny, deuterosomes or vertebrates?

    Doolittle argues that we may have to cash in the standard
    view of homology. "[H]omology is still a funny word," he
    writes (2000, p. 357). "In the context of proteins and
    genes, it makes sense only if we don't think about it too
    deeply." If there never was a common ancestor, Doolittle
    continues, then perhaps "*all genes are homologous*" (p. 357,
    emphasis in original).

    If all genes are homologous, however, "we might still be
    able to distinguish between orthologs and paralogs, as a
    matter of logical principle but, in practice, this will
    often be impossible....It is ironic that the words we seem
    to need in order to think productively about biology, words
    such as 'homology', 'individual', 'organism' and 'species',
    have no precise meaning" (pp. 357-358).

    I leave it to your imagination, Tim, to sort out the implications
    of that position for phylogenetic reconstruction. Of anything.

    >> These are minority viewpoints, of course. Most evolutionary
    >> biologists accept the monophyly of life as a background theory
    >> in their day-to-day research.
    >>
    >> But that's not the point at issue. It is increasingly possible
    >> to cast doubt on monophyly at various levels in the taxonomic
    >> hierarchy without stepping outside the arena of reasonable
    >> scientific discourse.
    >
    > Which levels for instance? Eukaryotes? Plastid phylogeny?
    > Higher primates?

    Doolittle and Woese doubt the common ancestry of the three
    primary domains. Gordon doubts that, too, and throws in
    skepticism about the common ancestry of the tetrapods for
    good measure: "Based on all these considerations the case
    seems strong that the tetrapods were polyphyletic in origin"
    (1999, p. 342). Fryer (1996) thinks the arthropods are
    polyphyletic. Conway Morris speculates that the animals
    may be polyphyletic, and favors monophyly only very weakly
    over polyphyly:

         For what it is worth my own belief is that metazoans
         are indeed monophyletic, but to my mind the argument
         is not yet won. More importantly, the question of
         biological constraint, the prevalence of convergence
         and the inevitability of polyphyly are not only
         interconnected topics, but also unjustly neglected.
         In part I suggest this is because of the atomistic
         emphasis now given to biology, as well as an
         obsession with cladistic methodology, which although
         freely acknowledges homoplasy regards it as an
         irritating diversion rather than a profoundly
         interesting problem in its own right. (1998, p. 13)

    At the Baylor naturalism conference in April, Conway Morris
    told me in conversation that the molecular evidence on the
    polyphyly of the animals was equivocal, but that in the near
    future, it may favor polyphyly. His mind, he said, is more
    than open to that possibility.

    > In actuality, what is being discussed by Doolittle et al.
    > is whether large bouts of horizontal transfer preceded the
    > split between the eubacteria, archeabacteria and eukaryotes,
    > which would obscure the past evolutionary trajectories of
    > these groups. There are certainly instances of horizontal
    > transfer after the split, so there is no reason to suspect
    > that genomes were necessarily as fixed as many are today.
    > I personally would not be surprised if this was the case and
    > I would like to note further that most of proposed episodes
    > of horizontal transfers discussed by Doolittle are confined
    > to single-celled
    > organisms.

    Well, that's simplifying the situation a tad, wouldn't you say? ;-)

    What really happened is the advent of whole genome sequencing
    blew a massive hole in the received view of the tree of life,
    and each new genome that comes in makes that hole wider.
    As Nierman et al. summarize the situation:

         As a result of the completion of genome sequences from
         representatives of all three domains of life, it is
         now possible to examine evolutionary relationships
         among living things in a more comprehensive way.
         However, this task has turned out to be anything but
         straightforward. Incongruities can be seen everywhere
         in the phylogenetic tree, from its root to the major
         branchings, when single protein phylogenies are
         examined. (2000, p. 345)

    The hypothesis of promiscuous early lateral gene transfer
    (LGT) has come to the rescue, so to speak, but at a rather
    high cost. One must invoke entities very unlike modern
    cells, which carefully regulate their LGT. There is little
    specificity to these hypotheses of early widespread LGT,
    perhaps because, as Woese observes (2000, p. 8395),
    LGT today doesn't resemble what (putatively) happened
    early in the history of life:
         
         Modern cells are fully evolved entities. They are
         sufficiently complex, integrated, and "indivdualized"
         that further major change in their designs does not
         appear possible...there will come a point at which
         certain of the cell's componentry becomes sufficiently
         complex (idiosyncratic) and sufficiently integrated into
         the emerging cellular fabric that horizontal gene
         displacement (especially of the phylogenetically
         distant variety) will not strongly influence them.

    One day soon some investigator is going to wake up in a
    huffy mood and write a paper questioning the extent of
    early LGT currently being invoked, on the grounds that
    one simply do not know what one talking about when one
    postulates "simple" cells with "primitive, error-prone"
    systems freely exchanging basic functional components
    (e.g., aminoacyl tRNA synthetases). Maybe that paper
    is being drafted even now.

    >> But based on current evidence, can one truely question
    >> monophyly at "various levels" much higher than that
    >> delimited by, say, higher eukaryotes?

    See the papers cited above. I should also mention that
    phylogenetic reconstruction is not the molecules-confirm-
    morphology story often depicted in popular presentations.
    For a strikingly different perspective, see Lynch (1999, p.
    323):

         Clarification of the phylogenetic relationships of the
         major animal phyla has been an elusive problem, with
         analyses based on different genes and even analyses
         based on the same genes yielding a diversity of
         phylogenetic trees. For example, considerable effort
         has gone into resolving the relationships of the three
         major protostome lineages (annelids, arthropods, and
         molluscs). Analyses with 18S rRNA consistently identify
         annelids and molluscs as sister taxa, as does a
         morphology-based analysis and the analysis of Guigo
         et al. (1996). However, results based on partial 28S
         rRNA sequences are not even consistent with the
         monophyly of protostomes, suggesting instead a
         mollusc-deuterostome affiliation. Two studies employing
         elongation factor 1-alpha have also yielded different
         topological relationships involving annelids, arthropods,
         and molluscs. Similarly, the phylogenetic position of
         nematodes has been controversial. Of the two studies
         that are based on 18S rDNA sequences, one positions
         nematodes basally with respect to the coelomate lineages,
         while the other joins them with arthropods. Aguinaldo
         et al. (1997) have argued that this inconsistency is an
         artifact of using nematode sequences that have evolved
         at unusually high rates. In still another example of
         phylogenetic inconsistency, in a study involving two
         protein-coding genes thought to be ideally suited to
         phylogenetic analysis, Nikoh et al. (1997) found amphioxus
         to cluster in its expected position with the chordates in
         one case, but to branch off prior to the deuterostome-
         protostome divergence in the other case.

         Given the substantial evolutionary time separating the
         animal phyla, it is not surprising that single-gene analyses
         yield such discordant results.

    Tim continues:

    >> I realize Paul, that you question human/ape monophyly,
    >> but do you suppose that Ford's comments could be used
    >> to reasonably cast doubt on many vertebrate phylogenies?

    I think so, once one follows out the logic of his position
    ("all genes are homologous"). Phylogenetic reconstruction
    is entering a revolutionary period; where the debris will
    land is anyone's guess.

    >> Does another Fellow at your institute, Mike Behe, question
    >> common descent at such levels?

    No, but he and Jonathan Wells and I have a long-standing
    friendly discussion going about the strengths and weaknesses
    of theories of common descent. Mike accepts common descent
    provisionally. As he's said on more than one occasion, however,
    if the theory turns out to be false, OK. I think he's waiting
    to see what Wells and I have to say (in the refereed
    literature) about the theory.

    >> BTW, Paul - Here's a question I've always wondered about,
    >> but for which 've never seen a clear answer: What's Phil
    >> Johnson's position on common descent? Largely in agreement,
    >> partially in agreement, not in agreement, strongly in
    >> disagreement?

    I'd guess that Phil doubts common descent, but he holds his
    skepticism lightly. He's at the hub of a growing research
    community (the ID group) where views on common descent
    differ. The real issue for Phil is naturalism.

    Paul Nelson
    Senior Fellow
    The Discovery Institute
    www.discovery.org/crsc

    References

    Conway Morris, S. 1998. The Question of Metazoan Monophyly and
       the Fossil Record. _Progress in Molecular and Subcellular
       Biology_ 21:1-19.

    Doolittle, W. Ford. 2000. The nature of the universal ancestor
       and the evolution of the proteome. _Current Opinion in
       Structural Biology_ 10:355-358.

    Fryer, Geoffrey. 1996. Reflection on arthropod evolution.
       _Biological Journal of the Linnean Society_ 58 (1996):1-55.

    Gordon, Malcolm S. 1999. The Concept of Monophyly: A Speculative
       Essay. _Biology and Philosophy_ 14:331-348.

    Lynch, Michael. 1999. The Age and Relationships of the
       Major Animal Phyla. _Evolution_ 53:319-325.

    Nierman, William C. 2000. Genome data: what do we learn?
       _Current Opinion in Structural Biology_ 10:343-348.

    Woese, Carl R. 2000. Interpreting the universal phylogenetic
       tree. _PNAS_ 97:8392-8396.



    This archive was generated by hypermail 2b29 : Tue Jul 25 2000 - 09:26:05 EDT