Dear dartR community,
I could use some clarification/help running EMIBD9 inside dartR.captive. I have a situation where I probably have high background relatedness so I
need to adjust for this by selecting argument Inbreeding=TRUE in gl.run.EMIBD9() command. Have I understood correctly that default value for option
EM_Method is always 1 meaning that expectation maximisation (EM) algorithm estimates both Δ and allele frequencies jointly? This is not included
into arguments for gl.run.EMIBD9() nor simulations for kinship. Is there any way to add or change it?
Then related to this, I need to also run simulations with gl.sim.relatedness() to determine thresholds for kinship with the same setting, EM enabled (EM_Method = 1)
and Inbreeding = TRUE. Now I can see from the produced output that EM is enabled (it says: "Performming EM for Delta and P: Replicate=1" while running). However, there is no
argument to select Inbreeding=TRUE in arguments. Again, is it possible to get it added (manually) for the function to change the default which I think is FALSE?
Below the options I have selected currently to run this. I don't think they match now, one is using inbred model as I wish while the other one is not.
I assume both enable EM_Method but it would be nice to test own results against random background assumption as well.
I used dartR.captive version 1.0.2
To run EMIBD9:
EMIBD9 <- gl.run.EMIBD9(mydata_gl,
emibd9.path = "[path_toEMIBD9_here]",
Inbreed = TRUE,
ISeed = [randomnumber])
To run simulations for full sibs:
full_sibs_simulated <- gl.sim.relatedness(
gl.run.EMIBD9(mydata_gl,
rel = "full.sib",
nboots = 100,
emibd9.path = "[path_toEMIBD9_here]",
conf = 0.95,
iseed = [different_randomnumber],
plot.out = TRUE,
plot.dir = NULL,
plot.file = NULL,
verbose = NULL
)
Thanks a ton for any help!