Prediction and Interaction in Complex Disease Genetics Experience in Type 1 Diabetes 英文参考文献.docVIP

Viewpoints PredictionandInteractioninComplexDiseaseGenetics: ExperienceinType1Diabetes DavidG.Clayton* Juvenile Diabetes Research Foundation/Wellcome Trust Diabetes and Inflammation Laboratory, Department of Medical Genetics, Cambridge Institute for Medical Research,CambridgeUniversity,Cambridge,UnitedKingdom Introduction eventsuchasseriousdiseaseincidence,the ratio of true positives to false positives, PPV/(12PPV),isgivenbymultiplyingthe populationriskbytheratiooftheordinate totheabcissaoftheROCcurve.Although there have been many attempts to repre- sentpredictiveefficacyintermsofasingle number [4], such indices are often mis- leading,anditwillgenerallybenecessary toconsiderthewholecurvewhenassessing the usefulness of prediction for clinical or publicheathpurposes. menthasrecentlybeenreawakenedinthe context of advances in genetics. For example, Sir George Radda, then chief executiveoftheMedicalResearchCoun- cil,stated:‘‘In20years’time,wemaysee individualized approaches to disease pre- vention and treatment’’ [6]. Ironically, such public pronouncements came at a Muchofthepublicmediadiscussionof geneticsofcommondiseaseshascentered on opportunities for targeted preventive actions.Atthesametime,inthespecialist literature,therehasbeenextensivediscus- sionof‘‘interaction,’’bothbetweengenes and between genes and environment. These two topics concern the use and interpretation ofstatisticalmodels forrisk of diseases with several, perhaps many, etiological risk factors, and were both the subjectoflivelydebatesome30to40years ago when such models first came into widespread use in epidemiology. Here these debates are revisited and illustrated with results from an analysis of the genetics of type 1 diabetes (T1D). Details ofthisanalysisareprovidedinsection1of TextS1. time when complex disease genetics seemed to be making little headway [7]. Recent successesofgenome-wideassocia- tion studies have established a more Initialhopesthatmultivariateriskscores could form the