The files in this directory compare the power to detect a QTL that
accounts for 10% of the phenotypic variance using a) sib pairs and b)
sibships of size four.

The guts of the output is at the end, where the power at the .0002
significance level (equivalent to a lod score of about 3) is computed.

Additive QTL 10% of variance, residual genetic variance 30%
Sib pairs
 Power of this test, at the 0.0002 significance level
 with   1. df is 0.555025
 Based on your combined observed sample size of        4000.
  The following sample sizes would be required to reject the hypothesis:
  
  Power         Total N
  .25            6558.
  .50            9788.
  .75           13663.
  .80           14722.
  .90           17701.
  .95           20368.
  .99           25875.

Sib quads
 Power of this test, at the 0.0002 significance level
 with   1. df is 0.862171
 Based on your combined observed sample size of        2560.
  The following sample sizes would be required to reject the hypothesis:
  
  Power         Total N  
  .25            1061.   
  .50            1584.   
  .75            2211.   
  .80            2382.   
  .90            2864.   
  .95            3296.   
  .99            4187.   

For these parameter values, with sib quads there seems to be little effect
of making qtl variance non-additive as the extra chisq gets
counterbalanced by the additional df.