Today, with the Leek group and other collaborators at JHU, we released a new version of the recount resource.  recount provides processed and summarized expression data for nearly 60,000 human RNA-seq samples from the Sequence Read Archive (SRA). The associated Bioconductor package provides a convenient API for querying, downloading, and analyzing the data. Each processed study consists of meta- and phenotype data, the expression levels of genes and their underlying exons and splice junctions, and corresponding genomic annotation.

Take a look at our preprint describing the resource as well as workflows illustrating how recount can be leveraged to perform differential expression analysis including meta-analysis, annotation-free base-level analysis, and replication of smaller studies using data from larger studies. These analyses are fully reproducible, as seen here.

The previous version of the recount was released in 2011 and summarized gene-level expression for about 475 samples of RNA-seq data.  (That was a lot back in 2011.) That resource was used, for example, to develop new methods for differential expression and normalization, to compile co-expression networks, and to study the effect of ribosomal DNA dosage on gene expression.