Median-Joining Network

Description

This function computes the median-joining network (MJN) as described by Bandelt et al. (1999).

Usage

mjn(x, epsilon = 0, max.n.cost = 10000, prefix = "median.vector_")

Arguments

Details

MJN is a network method where intermediate (unobserved) sequences (the median vectors) are reconstructed and included in the final network. Unlike mst, rmst, and msn, mjn works with the original sequences, the distances being calculated internally using a Hamming distance method (with dist(x, "manhattan") for binary data or dist.dna(x, "N") for DNA sequences).

The parameter epsilon controls how the search for new median vectors is performed: the larger this parameter, the wider the search (see the example with binary data).

If the sequences are very divergent, the search for new median vectors can take a very long time. The argument max.n.cost controls how many such vectors are added to the network (the default value should avoid the function to run endlessly).

Value

an object of class "haploNet" with an extra attribute (data) containing the original data together with the median vectors.

Note

This version still needs to be tested with large data sets. Bandelt et al. (1999) reported long computing times because of the need to compute a lot of median vectors.

Author(s)

Emmanuel Paradis

References

Bandelt, H. J., Forster, P. and Rohl, A. (1999) Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution, 16, 37–48.

See Also

haploNet, mst

Examples

## data in Table 1 of Bandelt et al. (1999):
x <- c(0, 0, 0, 0, 0, 0, 0, 0, 0,
       1, 1, 1, 1, 0, 0, 0, 0, 0,
       1, 0, 0, 0, 1, 1, 1, 0, 0,
       0, 1, 1, 1, 1, 1, 0, 1, 1)
x <- matrix(x, 4, 9, byrow = TRUE)
rownames(x) <- LETTERS[1:4]
(nt0 <- mjn(x))
(nt1 <- mjn(x, 1))
(nt2 <- mjn(x, 2))
plot(nt0)

## Not run: 
## same like in Fig. 4 of Bandelt et al. (1999):
plotNetMDS(nt2, dist(attr(nt2, "data"), "manhattan"), 3)

## End(Not run)

## data in Table 2 of Bandelt et al. (1999):
z <- list(c("g", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a"),
          c("a", "g", "g", "a", "a", "a", "a", "a", "a", "a", "a", "a"),
          c("a", "a", "a", "g", "a", "a", "a", "a", "a", "a", "g", "g"),
          c("a", "a", "a", "a", "g", "g", "a", "a", "a", "a", "g", "g"),
          c("a", "a", "a", "a", "a", "a", "a", "a", "g", "g", "c", "c"),
          c("a", "a", "a", "a", "a", "a", "g", "g", "g", "g", "a", "a"))
names(z) <- c("A1", "A2", "B1", "B2", "C", "D")
z <- as.matrix(as.DNAbin(z))
(ntz <- mjn(z, 2))

## Not run: 
## same like in Fig. 5 of Bandelt et al. (1999):
plotNetMDS(ntz, dist.dna(attr(ntz, "data"), "N"), 3)

## End(Not run)