A nucleotide sequencing and computational technology used to identify and quantify RNA expressed in biological samples. A variety of sequencing and computational strategies may be used, all based on the general strategy of purifying RNA from samples, generating cDNA fragments from the purified RNA, adding adaptors to one or both ends of the fragments, and obtaining short sequence reads from one or both ends of the the cDNA fragments. In some circumstances the cDNA fragments are selected by an exome capture probe-set in order to limit sequencing to exomic regions. Computational techniques are used to align the sequences to a reference genome or exome to identify the transcripts’ 5’ and 3’ ends, splicing sites, and post-transcriptional modifications, and to quantify expression levels. The methods are also used to analyze for expression of mutated alleles, and determine if structural alterations are present (small insertions or deletions [indels]) or alternative splicing. Throughput is high, resolution is to a single nucleotide, dynamic range of expression quantification is high, the required amount of RNA is low, and relative cost is low. Techniques for sequencing long fragments of RNA, either directly or as cDNA, are rapidly becoming available, allowing more direct detection of RNA structural variants such as alternative splicing or mis-splicing.