Generating A Tree From Nucleotide Frequencies

Question

[Oliver]

Hello, I am trying to generate a distance tree in PAUP from base frequencies. The 'Basefreq' command gives the base frequencies for each taxon but i need this to be converted to distances for tree building. Does anybody know how to convert these frequencies to distances? (either within PAUP or via another program)

Thankyou very much.

[Fucitol]

If you're familiar with R, you can easily calculate distances between sequences using the dist.dna function from the ape package.

From there on it's easy to create a graphical tree using other functions from that package. Please let us know if you need help.

[Oliver]

Thank you for your reply Fucitol,

My problem is the need to calculate distances based on a frequency table rather than directly from sequences,

for example turning something like this...

##Taxon          A       C         G      T
A           0.345     0.270   0.094   0.284      
B           0.304     0.331   0.107   0.256       
C           0.309     0.324   0.101   0.264

into...

            A     B     C
A           0     x1    x2  
B           x3    0     x4
C           x5    x6    0

Thank you again

[Oliver]

Thank you for your reply Fucitol,

My problem is the need to calculate distances based on a frequency table rather than directly from sequences, for example turning something like this...

Taxon   A     C     G     T
 A    0.345 0.270 0.094 0.284 
 B    0.304 0.331 0.107 0.256 
 C    0.309 0.324 0.101 0.264

into...

       A   B   C 
    A  0   x   x 
    B  x   0   x
    C  x   x   0

Thank you again

[Oliver]

Thank you for your reply Fucitol,

My problem is the need to calculate distances based on a frequency table rather than directly from sequences. The frequency table will contain for each taxa; the frequencies of each nucleotide and I need to figure out how to convert this to a distance table which contains the pairwise distances between each taxa pairing.

Thank you again

Answer 1

[Joseph Hughes]

If I understand correctly, you want to determine the similarities between the base frequencies of your taxa. Here is a bit of code that should help you to do that in R:

#using the data you provided above
freq<-c( 0.345, 0.270 ,0.094 ,0.284, 0.304, 0.331 ,0.107 ,0.256, 0.309, 0.324, 0.101, 0.264 )
freq<-matrix(freq, nrow = 3, byrow=TRUE)
freq<-as.data.frame(freq)
row.names(freq)<-c("A","B","C")
# using dist will calculate the pairwise distances from the frequencies
hc = hclust(dist(freq))
plot(hc)

This gives you a dendrogram as follows:

I hope this might help even if it is 16 months ago you posted.