With gene based counts, our results agree with what Leonard reported using SNP based counts.  In addition to his report,  we have checked ratio of maternally mapped reads of F1 samples for autosome genes.  For each sample, the WIKI file named SummaryAutoRatio.csv summarizes the # of genes (with total reads>49) on each chrosomose.  For the genes with total read counts > 49, we calulated the average, median, and variance of the M/(M+P) ratios. The Last column lists P-values for testing if overall M/(M+P) ratio is 0.5.  Clearly GH_0081_M and  GH_0087_F  are problematic.  Some samples have average ratios of ~0.52 and  ~0.48 for some chromosomes, especially for chrs 10 and 11.  But I guess they should not be a big conern.  

Fei