RNA-seq data analysis for differential expression of a single gene

Question

[nancy]

I have some RNA-seq data for a large number of samples (treated vs control), to the order of ~50M PE reads per sample, from mouse. These data have not been aligned yet to the reference genome.

I am, however interested in the expression level of one single target gene across all these samples, and a quantification of whether it is differentially expressed between treated and control.

Is there a quick way of doing this, where I can bypass alignment to the whole genome, but only align to the target gene of interest? (An alignment-free method like kallisto/sleuth may be another option, it just occurred to me, but I was wondering more about a brute-force alignment to target method)

Thanks

Answer 1

[JC]

Your major problem will be uncertainness of reads mapped to more than one location in the genome/transcriptome, definitively you need to align to the genome or transcriptome to remove non-unique reads or apply some strategy to decide origin. Kallisto, Sleuth or Salmon are fast enough to create your expression tables and then filter your gene.