sim.arg error when rho is high

[WongDWai]

Had this warning when increasing rho above 3:
sim.arg ERROR: missing value where TRUE/FALSE needed
Warning messages: 1: In rhyper(1, count, p.size - count, b.size) : NAs produced

Checking the values:
Browse[1]> count
[1] 202
Browse[1]> p.size
[1] 100
Browse[1]> b.size
[1] 1
Browse[1]> n.transfer
[1] NA
Browse[1]> recipient$get.name()
[1] "I_27"
Browse[1]> rhyper(1, count, p.size-count, b.size)
[1] NA

[ArtPoon]

The problem is that we have exponential growth in the number of active lineages within a given host, when there is a non-zero recombination rate (because every active lineage can potentially "recombine"). In other words, we are always adding a new active lineage to the host de novo. This is incorrect! What we should be doing instead is sampling a recombination parent from a fixed number of N lineages (where N is the effective population size), and a proportion of N are currently active. We activate an inactive lineage only if one of those gets chosen at random. The total number of lineages always stays the same.