You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The relatedness() function could be broadened to support providing the kinship coefficient between individuals, in addition to the current "coefficient of relationship". On the Eidos side this is easy; just add a [s$ type="Wright"] parameter that provides the default as it is now, and add type=="kinship". On the math side, I have no idea how to do this, so I'm assigning it to you @petrelharp, but marking it as long-term; up to you whether/when this gets done. :->
Incidental to this: I can imagine adding a new level of pedigree tracking that says "keep info up to great-grandparents". Tree-seq would only turn on the basic level, but users that wanted to expend more time and memory usage on extended pedigree tracking could choose to do so, which would make these relatedness metrics that much more accurate. Not sure whether that's worth doing; maybe if people want that degree of precision they ought to be calculating it from the tree sequence anyway, but of course that can't be done as the simulation is running.
The text was updated successfully, but these errors were encountered:
The
relatedness()
function could be broadened to support providing the kinship coefficient between individuals, in addition to the current "coefficient of relationship". On the Eidos side this is easy; just add a[s$ type="Wright"]
parameter that provides the default as it is now, and addtype=="kinship"
. On the math side, I have no idea how to do this, so I'm assigning it to you @petrelharp, but marking it as long-term; up to you whether/when this gets done. :->Incidental to this: I can imagine adding a new level of pedigree tracking that says "keep info up to great-grandparents". Tree-seq would only turn on the basic level, but users that wanted to expend more time and memory usage on extended pedigree tracking could choose to do so, which would make these relatedness metrics that much more accurate. Not sure whether that's worth doing; maybe if people want that degree of precision they ought to be calculating it from the tree sequence anyway, but of course that can't be done as the simulation is running.
The text was updated successfully, but these errors were encountered: