Dear all,
I took the liberty to summarize the responses and re-post it to the discuss=
ion group for the aid of all. Hereby a short summary of the respons I got. =
Thanks for the respons !
See also Mindell & Thacker (1996). Rates of Molecular Evolution: phylogenet=
ic issues and applications. Ann. Rev. Ecol. Syst. 27:279-313.
Bromham et al. (2000) The power of RRT depends on the data. J Mol Evol 20:2=
96-301
> Dear all,
>> We are wrestling to explain the finding that recently derived
> nematodal clades have a significant higher rate of evolution that
> older clades (Relative rate tests). In fact, we witness an increase in
> evolutionary rate from primitive clades towards more recently derived
> clades.
>> Materials & Methods:
> The program Phyltest {Kumar, 1995} was used to conduct relative rate
> tests to compare the evolutionary rates between different groups of
> nematodes. This program conducts the relative rate tests according to
> Takezaki et al. {Takezaki, 1995}. Kimura's distance {Kimura, 1980}
> with the gamma correction calculated by Modeltest was used to
> calculate pairwise distances. All groups were tested pairwise and the
> sistergroup to the two groups compared (not necessarily the outgroups
> to the phylum) was used as outgroup when calculating the distances to
> get the best result {Kumar, 1995}.
>> 1. Can anyone help us in finding an explanation for this phenomenon ?
> (e.g. relevant articles/opinions/vicariance/life hostory traits ?) 2.
> Do we have to correct for this in a phylogenetic analysis (in
> MP/Bayesian/NJ trees) ?
>> Best regards,
>> Andre van der Wurff
1. Kerstin Hoef-Emden wrote:
Dr. Kerstin Hoef-Emden=09=09=09Gyrhofstr. 15
Universit=E4t zu K=F6ln=09=09=0950931 K=F6ln
Botanisches Institut=09=09=09Germany
Is it also obvious from the phylogenetic trees? Which gene(s) or DNA fragme=
nts did you use? May the acceleration of evolutionary rates indicate specia=
tion processes?
Often an acceleration of substitution rates was accompanied by a change in =
life style, e.g. a change from free-living to endosymbiontic life style, a =
change from photosynthesis to heterotrophy etc. In endosymbionts and organe=
lles also a bias toward increased A+T contents was found.
Are these younger nematodal lineages endosymbiotic, did they change habitat=
=3D , did they become asexual ....?
a few refs:
Moran NA (1996). Accelerated evolution and Muller's ratchet in endosymbioti=
=3D c bacteria. PNAS 93: 2873-2878
Nedelcu A (2001). Complex patterns of plastid 16S rRNA gene evolution in no=
nphotosynthetic algae. J. Mol. Evol. 53: 670-679
dePamphilis CW, Young ND, Wolfe AD (1997). Evolution of plastid gene rps2 i=
=3D n a lineage of hemiparasitic and holoparasitic plants: many losses of p=
hotosynthesis and complex patterns of rate variation. PNAS 94: 7367-7372
Lynch M (1997). Mutation accumulation in nuclear, organellar, and prokaryot=
ic transfer RNA genes. Mol. Biol. Evol. 14: 914-925
Charlesworth D, Morgan MT, Charlesworth B (1993). Mutation accumulation in =
finite outbreeding and inbreeding populations. Genet. Res. 61: 39-56
If not corrected, the long branches may cause LBA artefacts. In ML and NJ: =
run analyses always with a gamma distribution. It does not really account f=
=3D or taxon-specific accelerated rates, but by including a site-specific r=
ate variation, the problem will be mildered. If you use Modeltest to choose=
the evolutionary model fitting best the data, it will presumably anyway co=
me up with an evolutionary model with a gamma distribution.
In MrBayes: settings to invgamma+covarion
In MP: presumably difficult
2. Byron Adams wrote:
Microbiology and Molecular Biology Dept.
775 WIDB
Brigham Young University
Provo, UT 84602-5253
One thing I think you'll want to look at is phylogenetic autocorrelation am=
ong the taxa. Kumar and Takezaki's tests don't take this into account, and=
so you usually end up with a pretty weak test. I've found what I thought =
(hoped) would be significant differences in rates only to see them disappea=
r when I took phylogenetic constraints into account.
Robinson M, Gouy M, Gautier C, Mouchiroud D (1998) "Sensitivity of the rela=
tive-rate test to taxonomic sampling", Mol Biol Evol 15(9):1091-1098.
Marc Robinson-Rechavi and Doroth=C8e Huchon (2000) "RRTree: Relative-Rate T=
ests between groups of sequences on a phylogenetic tree", Bioinformatics
16(3): 296-297.
Also, by what criterion are you determining primitive vs derived clades? Wh=
at you are probably seeing is a rate effect, like what has been seen in pla=
nts - taxa that are undergoing the most rapid rates of evolution also appea=
r to be the most species rich.
3.Mark Dowton wrote:
Institutes of Biomolecular Sciences, and Conservation Biology Dept Biology,
Wollongong University, Wollongong, NSW, 2522 AUSTRALIA
under opinion: it may be that you are seeing less hidden change in the rec=
ently derived clades, more in the more divergent. This would give the appe=
arance of higher rates in recently derived clades. I would consider trying=
to correct for hidden substitutions, then redoing the test.
---
