Properly reconstructing molecular phylogenies requires a good
understanding of the way proteins and nucleotidic sequences
evolve. Because any inadequation of the sequence evolution model can
seriously flaw the reconstructed phylogeny [1], there is a strong need
for one that would best fit the observed data, and also that would not
be too heavy to implement in an algorithm. Up to now, reconstruction
methods have relied on increasingly sophisticated models, from
"independent and equivalent" sites to gamma-law distributions of
substitution rate. However, all these models assume that the
substitution rate of a position remains constant throughout
time. Besides being biologically unsound, this hypothesis has now been
demonstrated wrong [2,3], meaning that a position can undergo drastic
changes in its substitution rate among lineages. The pioneering
analyses of Fitch [4,5] led him to postulate the existence of
covarions, i.e. concomitant sets of variable codons. These
covarions have now been formalized by assuming a position can switch
between "on" and "off" states, which defines if the position can vary
or not [6]. The open question is now to implement this feature, even
in a simplistic way, into a reconstruction algorithm. Especially, in a
maximum likelihood perspective, this sums up to adding a few new
parameters to a gamma-law model, namely the frequency of this
switching and the initial state of the parameter. However, this is not
a simple task and the best way to accomplish it shall be discussed
here. Even though we still don't know much about the covarion model
properties, such a model would at least be much closer to the observed
data, and certainly would open the way to more reliable phylogenies.
1. Lockhart P.J. et al. (1996) Evolution of chlorophyll and
bacteriochlorophyll: the problem of invariant sites in sequence
analysis. Proc. Natl. Acad. Sci. USA 93:1930-1934
2. Lopez P., Forterre P. and Philippe H. (1999) The root of the
tree of life in the light of the covarion model. J. Mol. Evol.
in press
3. Lockhart P.J. et al. (1998) A covariotide model explains apparent
phylogenetic structure of oxygenic photosynthetic
lineages. Mol. Biol. Evol. 15:1183-1188
4. Fitch W.M. (1971) The nonidentity of invariable positions in the
cytochromes c of different species. Biochem. Genet. 5:231-41
5. Fitch W.M. (1971) Rate of change of concomitantly variable
codons. J. Mol. Evol. 1, 1:84-96
6. Tuffley C. and Steel M. (1998) Modeling the covarion hypothesis
of nucleotide substitution. Math Biosci. 147:63-91
