When Would You Use Sage Over Rna-Seq

Question

[Ying W]

Is there ever a situation where you would use SAGE (or related protocols) over RNA-seq in organism where all transcripts are known?

The only situation I can think of is if you are assaying expression for multiple samples and constrained by amount of sequencing you can run (say you can only run one lane for cost reasons). Then you would get more information from SAGE than RNA-seq since SAGE uses less sequence per gene transcript. However, you would lose all information on SNPs/alternative splicing/differential exon usage.

Other than this specialized circumstance, is there any reason to use SAGE? Does SAGE require less starting material / less PCR / more accurate quantification?

[Sean Davis]

There may be some fine points, but the number of sequenced reads, not their location in the transcript, is the important factor for gene expression and differential gene expression, so RNA-seq and SAGE will behave similarly with regard to sequencing depth. In other words, I would not count reduced sequencing costs as a reason to prefer SAGE over RNA-seq.