DESCRIPTION (provided by applicant): Sickle cell anemia is a paradigmatic single gene disorder caused by homozygosity for a unique mutation on the beta-globin locus producing the abnormal sickle hemoglobin (HbS). Phenotypically, sickle cell anemia is a complex disease with different clinical courses ranging from early childhood mortality to virtually unrecognized conditions. Damaged red cells initiate hemolysis, vaso-occlusion and the vascular pathology of sickle cell disease. Vaso-occlusion injures vital tissues causing pain and impairing function. Death is premature and life can be oppressive. Supported by the NIH/NHLBI R01 HL68970 "Genetic modulation of sickle cell anemia", in 2001 Dr Steinberg initiated a genome scan study to understand the genetic basis of the major sickle cell anemia phenotypes. This study has led so far to the discovery of several genes that are associated with individual phenotypes of sickle cell anemia, and more than one phenotype appears to be associated with the same genetic variants. These findings support the hypotheses that clinical heterogeneity in sickle cell disease, as in other "single gene" Mendelian disorders, must be caused by the genetic variability in genes that influence the occurrence of defined phenotypes. This variability may be also modulated by other clinical conditions, and some of the sub-phenotypes of sickle cell anemia may have common genetics bases. To model these relationships and to allow ultimately the use of these discoveries as prognostic and therapeutic models, we are developing new computational methods for learning about simultaneous gene-phenotypes associations based on multivariate dependency models. In this project, we propose to use these new modeling techniques for the simultaneous discovery of the genetic basis of several sickle cell anemia phenotypes, and to use the discovered associations for prognosis of the risk of complications in sickle cell anemia patients.