Many people are currently searching for genes which contribute to the susceptibility of individuals to complex diseases such as schizophrenia, diabetes, coronary heart disease and asthma. Researchers typically have no solid knowledge concerning the nature or mode of action of such genes, or their frequencies in study populations. This has some relevance to the statistical methods used in searching for such genes. For example, traditional linkage analyses require a fully specified probability model, including gene frequencies and penetrances. In recent years there has been an increase in the use of Identity-By-Descent data, and statistical methods using only affected family members in this context, perhaps arising from a perception that these methods are to some extent more model-robust. Very recently a debate has arisen concerning the extent to which this last suggestion is true, and on the merits of affected only methods using IBD data, rather than traditional linkage anlayses. This is the background to the talk.

Back to Workshop Schedule