This function implements the SDM method of Criscuolo et al. (2006) for a set of n distance matrices.
SDM(...)2n elements (with n > 1), the first n elements are the
distance matrices: these can be (symmetric) matrices, objects of
class "dist", or a mix of both. The next n elements are the
sequence length from which the matrices have been estimated (can be
seen as a degree of confidence in matrices).
a 2-element list containing a distance matrix labelled by the union of the set of taxa of the input distance matrices, and a variance matrix associated to the returned distance matrix.
Reconstructs a consensus distance matrix from a set of input distance
matrices on overlapping sets of taxa. Potentially missing values in
the supermatrix are represented by NA. An error is returned if
the input distance matrices can not resolve to a consensus matrix.
Criscuolo, A., Berry, V., Douzery, E. J. P. , and Gascuel, O. (2006) SDM: A fast distance-based approach for (super)tree building in phylogenomics. Systematic Biology, 55, 740--755.