Our analysis suggests that, instead of following the familiar model of ‘‘unique malady–unique (disjoint with others) set of broken genes’’ applicable to most Mendelian disorders (Fig. 2D), most complex phenotypes are probably rooted in genetic variation that is significantly shared (in either a competitive or cooperative manner) by multiple disease phenotypes (Fig. 2E).
Phenotypes of non-Mendelian disorders are often defined with a considerable degree of fuzziness, especially those that are neurological: it is not uncommon to define a neuropsychiatric disease phenotype as comprising, for example, at least five of a list of 10 symptoms (4). This fuzziness arises because, in many cases, the observed disease is a heterogeneous collection of multiple maladies that have partially similar symptoms and potentially different genetic causes. However, these genetically heterogeneous maladies are combined because of the history of disease identification and the incompleteness of our knowledge about the disease causes.
Our interpretation of genetic overlap among pairs of disorders does not exclude the possibility that one disorder can cause the other. For example, it is possible that comorbidity of autism (or schizophrenia, or bipolar disorder)with infectious and autoimmune maladies indicates that the neurodevelopmental disorder can be triggered by different developmental insults, including viral or bacterial infection, or an autoimmune disease launched by a benign allergen. Another possibility is that the same molecular features that make a child more susceptible to infection or to autoimmune attack have a pleiotropic effect on brain development and function.
For the full list of overlaps, see the appendices. Appendix 1 has more pretty pictures, while 4 and 5 have some easier-to-read tables. Here's the pretty picture for plague (click to enlarge):
No comments:
Post a Comment