I sent a codon sequence to codeml and set seqtype = 2, here is my control file:
seqfile = reformatted_alignment.phy
treefile = reformatted_tree.tre
outfile = out.paml
noisy = 3
verbose = 9
runmode = 0
seqtype = 2
aaRatefile = jones.dat
model = 3
fix_alpha = 0
alpha = 1.0
ncatG = 4
RateAncestor = 2
Small_Diff = 1.0e-6
cleandata = 0
method = 1
fix_blength = 2
its still working. I'm so confused. I need an ancestor reconstruction result, and I found there are some different reconstruction sites between sent aa type squence and codon type squence with no parameter change. I want to know if it can work, does it treat a nucletide just as an aa to calculate substituion model and any calculate program or treat it as a codon translate to aa? And can the result of the codon type sequence using the parameter "seqtype=2" provide a correct result?