|
|
MethodsThe alignments were constructed using the program CLUSTALW: THOMPSON et al. (1994). CLUSTALW: improving the sensitivity of progressive multiple sequence alignments through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673-4680 The "hypervariable" portions, of variable lengths in different subunits, were removed, in order to avoid bias, particularly with distance-methods. The phylogenetic inferences were computed using the package PHYLIP: FELSENSTEIN, J (1993). PHYLIP, phylogenetic inference package. Distributed by the author, department of Genetics, University of Washington, Seattle. according to the methods presented in: Nicolas Le Novère and Jean-Pierre Changeux (1995). Molecular evolution of the nicotinic acetylcholine receptor: an example of multigene family in excitable cells. J Mol Evol 40: 155-172. Above the branches are the bootstrap scores (each alignment is bootstrapped 1000 times) of the Maximum Parsimony (rounded to the nearest integer). Under the branches are the bootstrap scores of the Neighbor-Joining. The triplications denote uncertainties, i.e. total opposition of the methods, or the two methods providing bootstrap scores below 50 %. When the topology obtained contradicts the species systematic, the trees are corrected, the new branches being annotated "syst". The position of subunits linked by dotted line have been determined with an independant alignment (because the sequence is very short and could distort the general result). Those trees were experimentally designed. They do not fully represent the personal opinion of the author, and they are surely not fully identical to the actual evolutionary history of the subunits. Contents
General ViewClick on the denomination of a subset to get a more detailed tree. Alignment28 sequences, 303 positions. Outgroups are human ACHα7 and α9.Results
β, δ, π, ρ SubunitsAlignment25 sequences, 425 positions. Outgroups are Lymnea "GABA" ζ and human glycine α1 subunits.Results
RDL-like SubunitsAlignment8 sequences, 335 positions. Outgroups are Human GABA β1 and Human GABA α1.Results
α, γ SubunitsAlignment37 sequences, 348 positions. Outgroups are Human GABA β1 and Drosophila RDL subunits.Results
Comments
Glycine Receptor SubunitsAlignment15 sequences, 336 positions. Outgroups are human GABA β1 and human GABA α1.Results
CommentsBecause of α4, M4 region is not taken into account,hence the small length of the alignment. New Subfamily of Subunits(Only in C. elegans so far.) Alignment11 sequences, 218 positions. Outgroups are human Lymnea GABA α1 and β1 subunits.Results
Anionic Glutamate Receptor SubunitsAlignment9 sequences, 332 positions. Outgroup is Lymnea GABA ζ subunit.Results
CommentsThe subunit "GAB"gbr2cael is probably the true ortholog of GLUclxonvo, rather than the so-called GLUclxcael, much more divergent (not shown here). Related BibliographyOrtells, M. O. and Lunt, G. G. (1995). Evolutionary history of the ligand-gated ion-channel superfamily of receptors. Trends Neurosci, 18(3): 121-126 In the article quoted above, there was certainly a confusion between the delta and dzeta subunits. Last modification: Wed Nov 24 10:59:05 GMT 2004 | Nicolas Le Novère
|
|