Phenotypes influencing each other

[durrantmm]

Say you hypothetically had three phenotypes - cough, headache, and congestion. These phenotypes will all highly correlate with each other among individuals. Wouldn't this correlation inflate the Bayes factor? How can I account for correlation among tested phenotypes?

[durrantmm]

Or say hypothetically that 75% of the time, when an individual receives diagnosis A, then they'll also be diagnosed with B. Doesnt't having both A and B in the phenotype tree improperly inflate the Bayes factor?