Type 2 diabetes (T2D) is a metabolic disorder characterized by insulin resistance and impaired insulin secretion that affects more than 20 million Americans, although the genetic component of the disorder is largely unknown. Individual genetic susceptibility to type 2 diabetes and other complex traits is the result of variation that is both common in human populations and rare, de novo and inherited mutations. We adopted a diverse set of genetics, genomics and informatics approaches to prioritize candidate genomic regions and variants and perform in-depth, targeted analysis of their contributions to type 2 diabetes susceptibility and related trait variability. Our initial efforts focused on the selection of candidate genes relevant to a complex trait by developing a metric to weight the relevance of functional gene annotations to the known biology of a trait. We used this method to select candidate genes for type 2 diabetes and performed a T2D case-control and quantitative trait association study in 2,335 Finnish individuals from the FUSION study. After follow-up in additional samples, we identified several variants that might contribute to T2D susceptibility. Genomic regions associated with plasma levels of HDL cholesterol and triglycerides were re-sequenced in individuals with trait-extreme values. Our analysis revealed a denser set of common and rare functional target variants including several non-synonymous, 3' UTR, and non-coding SNPs and indels. Finally, we utilized two approaches to identify candidate functional non-coding variants that may directly contribute to trait susceptibility. First, we used Formaldehyde-assisted isolation of regulatory elements (FAIRE) coupled with high-throughput sequencing to identify nucleosome-depleted regions in pancreatic islets. We used islet FAIRE-seq data to identify SNPs associated with T2D that potentially alter islet transcriptional regulation. A SNP in TCF7L2, rs7903146, was located in a FAIRE-seq site and demonstrated allelic differences in islet chromatin openness and enhancer activity, suggesting that it may contribute functionally to T2D susceptibility. Second, we used transcription factor binding site motifs to computationally predict variants that have allelic differences in regulatory activity. Taken together, these results suggest that identifying candidate genomic regions can successfully enrich for variation important for type 2 diabetes and other complex traits.