The obesity epidemic in the U.S. continues unabated, and is accompanied by substantial complications, including diabetes, cancer, cardiovascular disease, and death. Although the recent rise in obesity is due to lifestyle changes, susceptibilityor resistance to obesity is strongly influenced by genetic factors. Genome-wide association (GWA) studies, led by the investigators, have identified dozens of places (loci) in the human genome where common DNA sequence variation is associated with body mass index (BMI), and other measures of obesity. Although successful, these GWA studies have only been able to assess common variation, and most of these variants are noncoding, making it difficult to be confident about which of the multiple genes in each of the associated loci are relevant. Although loci identified from GWA studies can still highlight relevant biological pathways, the uncertainty about the relevant genes represents a key obstacle to completing the functional and computational follow-up studies that would reveal in detail the biology that has been hinted at by the GWA studies. Accordingly, we will focus our efforts on studying rare variation, and particularly rare coding variation, which can more precisely pinpoint genes that influence obesity. We will also reanalyze existing large (N~200,000) GWA data sets, using new "reference panels" containing rarer genetic variants to be able to test rarer genetic variation forassociation with BMI and other measures of obesity. We will test and follow-up in replication samples both coding and noncoding variation; we will correlate noncoding variation with genomic annotations that could provide further insight into mechanism (Aim 1). In addition, we will select 600 genes from within the loci identified by GWA studies and perform focused, targeted sequencing of the coding regions of these genes in well-phenotyped samples and samples drawn from the tails of the BMI distribution (Aim 2). To test coding variants that are of even lower frequency and therefore beyond the reach of GWA studies, we will perform more comprehensive genotyping studies in samples drawn from the tails of the BMI distribution, using an "exome chip" that comprehensively surveys nonsynonymous variation with frequency >0.5%. We will combine these data with additional available whole exome genotype and sequence data to identify individual variants (and hence genes) that show association with BMI and other measures of obesity (Aim 3). Finally, we will integrate the genotyping and sequencing data with other data sets (gene expression, protein-protein interaction) to identify an additional 600 genes for further targeted sequencing to identify genes with collections of rare variants that influence BMI and other measures of obesity, and extend these studies into samples of non-European ancestries (Aim 4). Successfully pinpointing genes that are associated with obesity would be a critical next step in uncovering key pathways that influence obesity in humans, which in turn could guide efforts at therapy and prevention.