Test of "Independent Birth" theory

Arlin Stoltzfus arlin at ac.dal.ca
Fri Apr 7 19:39:35 EST 1995


The "Independent Birth of Organisms" theory proposes that the genomes of
all organisms are "immutable" and arose "independently". On  p. 311 of
Senapathy's book, _The Independent Birth of Organisms_, there is a diagram
that nicely illustrates the meaning of "independent" origins.  I have
attempted to replicate the diagram below.  The original has a circle around
the central text, with  8 separate arrows pointing outwards to the 8
separate organisms:

                         dragonfly

              rat                           millipede
          
          
                         common pool of
 jellyfish           the genetic code and the         frog
                     genetic machineries in the
                       primordial pond
 
          dolphin                           crab
      
      
                        sea anemone


This idea can be distinguished experimentally from the
hypothesis of common ancestry + descent with modification (i.e.,
rats+dolphins arose from a common mammal stock; jellyfish+anemone arose
from a common coelenterate stock), as follows.  One may scan the sequence
databases for a gene of interest that has been characterized in all or most
of the above species, e.g., suppose it is a gene for 16S rRNA.  One may
make a bifurcating dendrogram from this sequence information using a
phylogenetic inference package (e.g., PAUP, Phylip).  Under the common
ancestry theory, the dendrogram for 16S rRNA is interpreted as an estimate
of the organismal phylogeny, representing the course of evolution from a
common animal ancestor, through descent with modification.  

A single dendrogram will not allow us to distinguish common ancestry vs.
independent birth.  Under the independent birth theory, the 16S rRNA
sequences for crab, anemone, rat, dolphin, frog, etc. are proposed to have
been drawn from random sequences in the primordial pond. A dendrogram *can*
be constructed from these randomly generated sequences-- but the pattern of
branching that results represents nothing in particular, only the
stochastic process of drawing from a pool of random sequences.

However, the two theories have different predictions about what will happen
if we *construct dendrograms from additional genes and compare them with
each other*.  The common ancestry theory predicts a relationship of
congruence-- the trees inferred from 16S rRNA from rat, frog, anemone, etc.
should tend to match the trees inferred from tubulin (or whatever) genes
from rat, frog, anemone, etc..  In this view, the trees are *not
independent*, but arise from a common set of processes of descent with
modification.  By contrast, in the independent birth theory, the trees are
*independent* in the exact sense of "independent": they are derived from
drawing random sequences from the primordial pond, and no particular
relationship is expected.  

This creates the conditions for an objective and falsifiable test of the
independent birth theory.  One may quantify the expectations for random
congruence of tree topologies, as predicted by the independent birth
theory, and compare this to observations.  A significant tendency for the
trees inferred from different genes to match would exclude a major tenet of
the independent birth theory, which would then have to be either abandoned
or revised extensively.  At the very least, the name would have to be
changed from "Independent Birth of Organisms" to "Non-independent Birth of
Organisms".  

By the way, gene sequences are readily available from rat and frog, but 
not (yet) from millipedes, dragonflies, crabs, anemones and dolphins, 
so I would suggest that we substitute fruitfly, mouse, human, C.elegans 
and chicken.  Before actually making the trees, though, perhaps Dr. 
Senapathy will comment on whether the test is fair.

Arlin



More information about the Mol-evol mailing list