Kimura alignment

Ronald DeBry histone at acpub.duke.edu
Wed Feb 1 13:30:42 EST 1995


In article <D39Dv5.Ax8 at zoo.toronto.edu> mes at zoo.toronto.edu (Mark Siddall) writes:
>
>Given a multiple hit site wherein roman numerals represent taxa and letters
>are bases and the following distribution of states:
>
>Taxon   I   II   III   IV   V  VI
>        A   A     G    G    T   T
>
>And the order above implies nothing about their relationship, that is a
>priori, we know nothing of relationships.
>There are two possible explanations:
>a) one transition and one transversion.
>    for example, transformation series
>      A-G-T,  or G-A-T
>b) two transversions:
>    for example, transformation series
>      A-T-G
>
>The decision as to which, requires reference to some tree.  But we do not
>yet have a tree, in fact, the tree is the end point of what we are 
>attempting to do.

You've hit on the real problem with this last sentance.  Any multiple
alignment should really be conditional on the evolutionary relationships
(whether you are using a Kimura substitution model or not).  Two kinds
of solutions have been proposed.  One can use an iterative reciprocal
process, where a trial alignment is used to give a trial phylogeny,
which is used to generate a new alignment then a new phylogeny and so on
until it converges on a stable alignment/phylogeny.  This approach has
been implemented as a computer program by Jotun Hein.  The on;y
alternative that I know of is a series of papers by Rich Thorne.  In his
method (essentially) all possible alignments are generated.  Each
alignment is given a likelihood based on a model incorporating both
insertion/deletion and substitution, and a distance matrix is calculated
by weighting each alignment based on its likelihood.  The phylogeny is
then inferred from that weighted distance matrix.

As someone who subscribes to the idea that phylogenetic inference is a
statistical problem and that maximum likelihood in some form is the best
approach, I am very interested in Thorne's papers.  However, it is
obviously very computer-intensive, and I am not aware of any practical
applications of his methods yet.

Ron DeBry
Department of Medicine
Duke Univ Medical Center

histone at acpub.duke.edu





More information about the Mol-evol mailing list