Missing data with *BEAST

[Huateng]

Hi,

I'm trying to *Beast for estimating species-tree with multi-locus
data. There are lots of missing data in my dataset-- the number of
individuals in each species are not equal for different loci, and some
loci have zero individual from some species. I could not get the
program to run if I include those loci with zero individual. So my
question is: do I have to throw these loci away? or is there a way to
code these loci to make the *Beast run?

Best,
Huateng

[pepster]

Hi,

You can add a single empty sequence (all NNN) for each missing species
in loci missing even one individual. This will be fine as long as for
each species, there are some loci containing more than 1 individuals
for that species. *BEAST will work even if some species are missing
even that, but then it is impossible to estimate the population size
for those extant species, and the estimation of their divergence times
may suffer.

-Joseph