"Most importantly, there is a need for systematic benchmarks where abundances are known a priori, yet where experiments mimic the complexities of in vivo transcriptomes." LPachter - Models for transcript quantification from RNA-Seq, 2011
NIST is considering the production of transcript isoform controls to be used as spike-in controls for isoform quantification experiments and as testing for various bioinformatics tools. These controls are envisioned as a group of RNAs at known concentrations, with sequences that are orthogonal to model genomes yet based on well-studied transcript isoforms (such as the set of isoforms of tP53).
What problems would you want to solve with control transcripts? I am seeking input on any possible uses, experiments, pitfalls, or other considerations.
As we aim to include a wider variety of models for controls, what concerns should guide these choices? We can not realistically sample much of the transcriptome for our controls, but careful planning might allow us to at least sample from exemplary regions in a way that allows you to comfortably claim that your sequencing run and pipeline can properly detect (some) differential transcript abundances.
NIST is considering the production of transcript isoform controls to be used as spike-in controls for isoform quantification experiments and as testing for various bioinformatics tools. These controls are envisioned as a group of RNAs at known concentrations, with sequences that are orthogonal to model genomes yet based on well-studied transcript isoforms (such as the set of isoforms of tP53).
What problems would you want to solve with control transcripts? I am seeking input on any possible uses, experiments, pitfalls, or other considerations.
As we aim to include a wider variety of models for controls, what concerns should guide these choices? We can not realistically sample much of the transcriptome for our controls, but careful planning might allow us to at least sample from exemplary regions in a way that allows you to comfortably claim that your sequencing run and pipeline can properly detect (some) differential transcript abundances.