Hi everyone, I am thinking on running some RNA-seq on MiSeq, it is currently the only machine I have available, but I am afraid of not getting enough reads. Have anyone here tried it?
Thanks.
Well a typical bacteria is around the order of mega bases, millions of bases (Mb). The Miseq produces data on the order of giga bases, billions of bases (Gb) .
That allows for a thousand fold coverage. That should be plenty to run many samples and replicates in parallel.
The OP is asking about RNA-Seq. Coverage, per se, is not the driving consideration for RNA-Seq, read counts are. The MiSeq can generate ~15x10^6 reads (or read pairs) in one run. For a complex eukaryote that number is _barely_ enough for one replicate. For a bacterium with a simpler transcriptome my gut tells me 4-5 million reads per replicate is where you want to be. This means 3-4 replicates per MiSeq run. If you have a complex experiment with several conditions & replicates this will add up to a number of runs. In that case I would recommend finding a service provider to do one multiplexed lane on a HiSeq.
I can confirm these numbers form our experiences. There are early bacterial transcriptome studies based on 454 reads with only some 100 k of reads that brought at least some helpful results. Today we aim for 3 - 5 M reads for a single bacterial transcriptome depending on the genome size and the aim of the study. Sure you can sequence deeper but the gain of additionally detected transcripts will decrease asymptotically with further generated reads. There is a publication which discusses this topic: *How deep is deep enough for RNA-Seq profiling of bacterial transcriptomes?*, Haas et al. 2012.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Maybe you can specify what you want to examine. The required depth depends on your aim.
If you think the MiSeq is underperforming, it could be that your sample is AT-rich or GC-rich. Is your bacterial genome known to have an AT-rich or GC-rich genome?