You can use the function nmissing() to determine the number of missing genotypes at each marker.
data(fake.f2)
pmis <- nmissing(fake.f2, "mar") / nind(fake.f2)
You can then use drop.markers() to drop a set of markers, for example those with > 20% missing data.
fake.f2.sub <- drop.markers(fake.f2, names(pmis)[pmis > 0.2])
karl