9.5 Running lm_pval example and sample output

Under the subdirectory `TraitTests/', run the lm_pval example by typing:

./lm_pval ped73_pval.par > pval.out

A portion of the output giving latent (fuzzy) p-values is below. See `pval.out' for the entire output file.

Combined distribution of fuzzy p-values, by locus: pval maxim marker-1 marker-2 marker-3 marker-4 marker-5 marker-6 marker-7 marker-8 marker-9 marker-10 0.00 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.000 0.000 0.000 0.01 0.004 0.000 0.000 0.000 0.004 0.000 0.009 0.011 0.000 0.000 0.000 0.02 0.008 0.000 0.000 0.000 0.009 0.005 0.020 0.024 0.005 0.005 0.005 0.03 0.011 0.000 0.000 0.000 0.016 0.019 0.033 0.036 0.019 0.019 0.019 0.04 0.015 0.000 0.000 0.000 0.023 0.032 0.045 0.049 0.032 0.032 0.032 0.05 0.019 0.000 0.000 0.000 0.029 0.046 0.058 0.062 0.046 0.046 0.046

The output table shows the cumulative distribution of the latent (fuzzy) p-values generated at each marker position, as well as the cumultative distribution of the maximum latent p-value over the markers. These distributions are over the latent inheritance patterns sampled, given the marker data. That is, for each value of `pval' in the left column, the table gives the proportion of sampled inheritance vectors at each marker that yeild a p-value less than `pval'. In the last row of the example output, when pval = 0.05, 4.6% of the realizations have a p-value less than 0.05 at marker-5; at marker-7 this value is 6.2%. Overall, 1.9% of the realizations have a maximum p-value over the markers that is less than pval = 0.05 (shown in the second column labeled 'maxim').

Recall again, that exact values in the output will depend on the random seed. In the case of a relatively short run of lm_pval there may be substantial differences in the estimated latent p-value distributions.

For more information regarding the MCMC parameters and diagnostic output, See MCMC parameters and options.