The cost of RNA-Seqing multiplexed samples is approaching the cost of running a panel of QPCRs on each sample. For a lot of experiments that we do looking at marker gene expression the RNA-Seq approach seems preferable. However, if we multiplex 20-40 samples in one lane I worry that we would only see the most abundant transcripts, and maybe not the ones we are looking for. I'd be interested in hearing if anyone has experience comparing the two - for example threshold value in qPCR versus abundance by RNA-Seq. Also, if anyone has experience of doing multiplexed RNA-Seq regularly, how much time does it take to process the data from a multiplexed run to gene expression rankings for each sample? Consistency of technical replicates? Numbers of biological replicates you use? Thanks, Leif
I can provide a graph I made while trying to decide if genes I was interested in were expressed in a cell line. y-axis is ct, x-axis is expression values from FlyBase (from modENCODE sequencing data), which I presume is FPKM.
It looks like qPCR is more sensitive from this, however, in my own sequencing experiments (on tissue) I saw all these genes expressed, although some were at low levels (FPKM
Hi Leif! How are you?
I also think that low read RNA-seq problem is that it might be difficult to detect low abundant transcripts, however, to me the main difference is between a biased and an un-biased approach...
I really feel like they address different questions. If you are not interested in the whole story and just want to focus on certain genes, RT-qPCR will be more sensitive even if transcripts are low, and it will give you relatively accurate quantitative information. RNA-seq will provide more data overall, but quantitative information is trickier to get (though not impossible) and you do risk missing certain genes if they are not expressed in quantities that are high enough for the threshold set by sequencing depth, which I don't believe you can predict. You can find some papers that address the question of minimum depth in the literature.
Also, getting good quality RNA-seq libraries can be difficult, expensive and time-consuming.
Eventually, it really depends on what you want to do!
I hope that helps,
Marine
There is also tag-Seq, $50 per sample and it is actually more accurate than standard RNA-seq in inferring abundances of polyadenilated transcripts (true! although we did not publish that study yet). Recent paper showcasing itag-seq, plus all the updated protocols and references in the supplementary materials, here:
https://www.dropbox.com/s/480rb8tq9kzf211/Science-2015-Dixon-1460-2.pdf?dl=0
I would always choose tag-seq over qPCR unless I really want to look at 3-4 specific genes.
you can consider the targeted amplicon kits (AmpliSeq from LifeTech) for RNA - or Fluidigm, as an alternative to qPCR instead of normal RNAseq
20-40 samples per lane may be just a waste of money and time.
- Badges
- Science method