The Addiction project has three categories of case status: (a) alcohol and possibly other illicit drug dependence (‘case’), (b) controls exposed to alcohol, but never addicted to alcohol or illicit drugs (‘control’) and (c) addicted to illicit drug(s) but not to alcohol (‘other’). There is a significant association between case status and genotyping batch, which could lead to bias in case-control allelic frequencies, although the occurrence of 213 batches with median size of 21 samples reduces the magnitude of any potential problem. Using analysis of variance, there is a significant effect of case status on the logarithm of missing call rate (p<0.01), which is due to the ‘other’ category having a higher rate than the other two (0.12% versus 0.10%, Figure S9). This effect appears to be due to confounding with the DNA source, which was either blood or cell line. Among the DNA samples in the ‘other’ category, 77% are from cell lines, whereas the values for alcohol cases and controls are 34% and 25%, respectively. The missing call rate for cell lines is very significantly higher than for
