Output options

PRINTQHAT (Boolean) When this is turned on, the point estimate for $Q$ is not only printed into the main results file, but also into a separate file with suffix ``q''. This file is required in order to run the companion program STRAT.

UPDATEFREQ (int) Number of MCMC iterations between printing each update on the screen. Value of 0 will cause this to be set automatically.

PRINTLIKES (Boolean) Print the current value of the likelihood to the screen in every iteration.

INTERMEDSAVE (int) If you're impatient to see preliminary results before the end of the run, you can have results printed to file at intervals during the MCMC run. A total of INTERMEDSAVE such files are printed, at equal intervals following the completion of the BURNIN. Turn this off by setting to 0. Names of these files created using OUTFILE name.

PRINTKLD (Boolean) Print the ``Kullback-Leibler'' divergence ``D'' between populations. The output to the screen gets a bit wide for moderate $K$, so the user may want to switch this off to clarify the output. The distance between populations $A$ and $B$ is computed as

$${\rm D}_{A,B} = \sum_{l=1}^L \sum_{j=1}^{J_l} \hat p_{A,j}^{(l)} \ln({{\hat p_{A,j}^{(l)}}\over{\hat p_{B,j}^{(l)}}}),$$ (5)

where $\hat p_{A,j}^{(l)}$ is the estimated allele frequency of allele $j$ at locus $l$ in population $A$, and where $L$ is the number of loci, and $J_l$ the number of alleles at locus $l$. Notice that ${\rm D}_{A,B}$ and ${\rm D}_{B,A}$ are not, in general the same; however to save space in printing the running update, we print the average of these. This measure of divergence can be motivated as follows. Suppose that we want to infer whether an individual that is actually from population $A$ is from $A$ or $B$. Then it is natural to compute the ratio of the likelihood of their genotype in each population. ${\rm D}_{A,B}$ gives the average contribution of a single genotyped allele to the log-likelihood ratio. Suppose for example that ${\rm D}_{A,B} = 0.1$. Then by genotyping 10 (diploid) loci, the expected log-likelihood ratio is 2.0. ${\rm D}_{A,B}$ provides a natural measure for the strength of evidence for assignments of individuals.

ECHODATA (Boolean) Print a brief summary of the data set to the screen and output file. (Prints the beginnings and ends of the top and bottom lines of the input file to allow the user to check that it has been read correctly.)

ANCESTDIST (Boolean) Collect information about the distribution of $Q$ for each individual, as well as just estimating the mean. When this is turned on, the output file includes the left- and right-hand ends of the probability intervals for each $q{(i)}$. (A probability interval is the Bayesian analog of a confidence interval.) The values printed show the middle $100p\%$ of the probability interval, where $p$ is a number in the range 0.0 to 1.0 and is set using ANCESTPINT. The distribution of $Q$ is estimated by recording the number of hits in each of a number of boxes between 0 and 1, to form a sort of histogram. The width of these boxes, which are of equal size, is set using NUMBOXES.