Recent advances in DNA sequencing technologies have made it possible to sequence the entire genome of several individuals, catalog the full spectrum of genetic variation in human populations and identify genes underlying rare human diseases. High-throughput sequencing technologies also offer unprecedented opportunities for assessing the contribution of rare genetic variation to complex diseases. For the success of sequencing-based association studies, in addition to large sample sizes, computational methods for the accurate detection of rare variants from sequence reads and association analysis of rare variants are needed. In this talk, I will describe an approach that combines DNA pooling with high-throughput target enrichment methods to enable the cost-efficient sequencing of selected genomic regions in thousands of individuals. I will describe a novel probabilistic method for the accurate detection of rare (and common) variants from sequence data derived from pooled DNA samples. I will briefly discuss results from a pooled sequencing study of 136 genes associated with type 2 diabetes in 3900 cases and controls. Although humans are diploid, virtually all sequencing studies of human genetic variation ignore phase or haplotype information. I will describe combinatorial and stochastic algorithms for haplotype assembly, i.e. reconstructing the two haplotypes for an individual using sequence reads generated from whole-genome sequencing. These algorithms are based on computing cuts in read-haplotype graphs and have been utilized to assemble haplotypes for several individual genomes. Finally, I will present results on the design of sequencing experiments to enable haplotype assembly.