TOWARDS INTEGRATIVE GENE PRIORITIZATION IN ALZHEIMER'S DISEASE

摘要: Many methods have been proposed for facilitating the uncovering of genes that underlie pathology different diseases. Some are purely statistical, resulting in a (mostly) undifferentiated set differentially expressed (or co-expressed), while others seek to prioritize through comparison against specific known targets. Most recent approaches use either single data or knowledge sources, combine independent predictions from each source. However, given multiple kinds heterogeneous sources potentially relevant gene prioritization, subject levels noise and varying reliability, source bearing information not carried by another, we claim an ideal prioritization method should provide ways discern amongst them true integrative fashion captures subtleties each, rather than using simple combination sources. Integration is thus more challenging its type counterpart. What propose novel, general, flexible formulation enables multi-source integration maximizes complementary nature order make most content aggregate data. Protein-protein interactions Gene Ontology annotations were used as together with assay-specific expression genome-wide association Leave-one-out testing was performed Alzheimer's Disease validate our method. We show performs better best systems currently published.