We've done RNA-seq on polysome fractionation samples across three different genotypes. We collected Rnp (free mRNA's) and Polysome (actively translating mRNA's) fractions as well as input mRNA from the samples before fractionation. The idea of collecting input is not only so that we can look at transcriptional changes but to normalize to something that would be free of the technical variation and RNA loss that polysome fractionation can produce. This is my current idea for a workflow: Tophat/Bowtie => htseq-count => divide poly/rnp counts by input counts per gene/feature => edgeR differences between genotypes across polysome/rnp/input categories My question is thus: Is there a way to tell edgeR to normalize to the input? Am I mucking things up my manually normalizing the read counts before giving them to edgeR? Thanks. Best, Scott Daniel Zarnescu Lab MCB Ph.D. Program University of Arizona [[alternative HTML version deleted]]