Two Problems with Multiple Genetic Studies of the Same Disease

Abstract

Different teams of investigators often conduct genetic studies of the same disease; this suggests the use of ``meta-analysis'' to determine what the combined evidence shows. In particular, it is important to determine whether the same family is participating in multiple studies, thus potentially improperly magnifying the influence of the data that it contributes. Basic reasons that the raw data cannot be combined straightforwardly include (1) lack of permission to do so in human-subject protocols, (2) rivalries among teams of investigators, and (3) different experiments performed by different teams. Privacy-preserving data mining and other PORTIA research themes should be brought to bear on this problem.

Back to publications
Back to previous page