The analysis of loss of heterozygosity (LOH) is perhaps the most widely use
d technique in cancer genetics. In primary tumors, however, the analysis of
LOH is fraught with technical problems that have limited its reproducibili
ty and interpretation. In particular, tumors are mixtures of neoplastic and
nonneoplastic cells, and the DNA from the nonneoplastic cells can mask LOH
. We here describe a new experimental approach, involving two components, t
o overcome these problems. First, a form of digital PCR1 was employed to di
rectly count, one by one, the number of each of the two alleles in tumor sa
mples. Second, Bayesian-type likelihood methods were used to measure the st
rength of the evidence for the allele distribution being different from nor
mal. This approach imparts a rigorous statistical basis to LOH analyses, an
d should be able to provide more reliable information than heretofore possi
ble in LOH studies of diverse tumor types.