Output 


The primary output file is called 'ATRIUMtest.out'. It has

   Summary of phenotype file information: total number of individuals in
   phenotype file, number of independent families, number of individuals 
   in each phenotype class (affected/unaffected/unknown) 

   Number of individuals in the genotype file and number of individuals
   who appear in both the genotype and phenotype files.

   Prevalence value used in the ATRIUM calculations.

   For each untyped SNP, 
     - rs number, position, M_D value based on its tag SNPs, largest
       pairwise r^2 of untyped SNP with any of its tag SNPs
     - tag SNPs and their positions
     - among those individuals with at least one tag SNP genotyped, 
       the numbers who are affected, unaffected, and of unknown phenotype, 
       respectively,
     - haplotype frequencies of tag SNPs in the full 
       (affected+unaffected+unknown) sample, estimated using the 
       IQLS method of Wang and McPeek (2009) JASA 104: 1251-1260, where 
       haplotype information for each individual is based on the 
       individual's genotype data plus parental genotypes when available.
     - value of the ATRIUM statistic and corresponding p-value 
       based on the chi-squared null distribution.
     - estimated allele frequencies for the untyped SNP in the full, 
       affected, unaffected, and unknown samples (Note: if any sample 
       size is below 15, the corresponding frequency will not be 
       estimated)
     

A second output file called 'ATRIUMtest.top' lists the top 20 untyped 
   SNPs, i.e. the 20 SNPs with the smallest p-values for association.  
   For each of these SNPs, only the rs number and p-value are given in
   ATRIUMtest.top.  The number of SNPs output to this file can be 
   decreased or increased by the user by changing MAXTOP (currently set 
   to 20) in the peddata.cc source file.