Naturally I will list first the portal that I'm currently working on. This resource attempts to use the principles of population genetics to investigate type 2 diabetes.
This large, collaborative project represents an attempt to identify characteristic attributes of human DNA around the globe. It was a relatively early effort (starting in 2002) with enormous, ongoing impact in the field.
Sequence data from the 1000 genomes project. This was a project intended to capture the majority of the genetic variability present in down to 1% of the human population.
The website describing Daniel MacArthur's Exome Aggregation project. Examine the data here or else download everything and process it yourself.
The Genotype-Tissue Expression project takes tissues from deceased contributors and subjects them to expression analysis. The resulting information offers great insight into tissue-specific expression, with implications for many other fields.
An effort to organize the data coming out of the Cancer Genome Atlas project, this site offers access to a huge quantity of data for cancer researchers.
What is the relationship between particular genes and cancer lines?
The human genome, the fly genome, many other genomes along with abundant supporting annotations -- they are all here and freely available.
The Source Read Archive is an excellent resource if you want to download DNA sequencing data. This time without the genome browser, but instead a better source from which to download raw FASTA files to perform your own analysis
So you have sequencing data and you'd like to align it to one of the referenced genomes. This NCBI site is a great place to get started.
A useful tool for estimating the linkage disequilibrium between any two nearby snips. A combination of great functionality with a rather unwieldy user interface.