Custom selected reference genes outperform pre-defined reference genes in transcriptomic analysis

摘要: RNA sequencing allows the measuring of gene expression at a resolution unmet by arrays or RT-qPCR. It is however necessary to normalize data library size, transcript size and composition, among other factors, before comparing levels. The use internal control genes spike-ins advocated in literature for scaling read counts, but methods choosing reference are mostly targeted RT-qPCR studies require set pre-selected candidate controls target genes. Here, we report an R-based pipeline select based solely on counts sizes. This novel method first normalizes Transcripts per Million (TPM) then excludes weakly expressed using DAFS script calculate cut-off. selects as references with lowest TPM covariance. We used this pick custom differential analysis three transcriptome sets from transgenic Arabidopsis plants expressing heterologous fungal effector proteins tagged GFP (using alone control). showed lower covariance fold change well broader range levels than commonly When analyzed NormFinder, both typical were considered suitable controls, selected more stably expressed. geNorm produced similar result which most ranked higher (i.e. expressed) proposed innovative, rapid simple. Since it does not depend genome annotation, can be any organism, candidates that always available.