I am quite new in the topic of next generation sequencing. I am trying to find the best extraction method of small RNAs. The idea is to compare different extraction methods by sRNAseq. I wonder what would be the best criteria then. I am not sure if the number of mapping reads is enough. I should look at the coverage on the exon intron and intergenic regions. Any other suggestions or link to some papers where I can find more info?
What about a TapeStation, the Agilent Bioanalyzer. A bit up from running a gel, gives you some better estimates on what you captured and the quality of what you captured. You can run it pre and post size selection. However, I'm not sure how well this will work once you do size selection for small RNAs, it might think your sample is degraded.
It might be better to find some people who already do this and ask them, find people who do sRNA-Seq in your model of interest and email them. See if there are any NGS core facilities near you and ask them what they use. Or, just ask the company you're getting the kit/machine from.
Outside of that, your only option is to spend a ton of money and benchmark each method by sequencing it. As for metrics, I'm guessing you could look at n-mer counts and see if any particular n-mer is in greater/lesser abundance than you'd expect.
Here's a paper that's done some comparison: http://nar.oxfordjournals.org/content/early/2013/11/05/nar.gkt1021.full
Keep in mind that most small RNAs that you'll be getting are miRNAs, so they won't be spliced.
Why are you talking about splicing here? I don't get it.
You mentioned introns, so...
Edit: Oops, you're not the OP! Mea culpa!
Oh ok, I thought schelarina was talking about looking for reads mapping to introns/exons/intergenic regions to evaluate the RNA extraction. Meaning: If a lot of reads map to not-smallRNA regions, the extraction was not good. And then your comment would make no sense, but now I see what you mean.
this is what I meant
How well this will work depends on the species a bit. If you're working on human or mouse then that'll probably work. If you're working on a different organism, then it's hard to know if a non-small RNA alignment is background contamination or just a something missing from the annotation.
I just noticed that you weren't the OP, sorry about the confusion there!
Yeah, OP could have meant what you wrote or have thought that small RNAs are also spliced and then have wanted to see how much of the pre-mRNA equivalent is present. This is one of the methods for evaluating mRNA extraction methods, since you only normally care about the mature mRNAs.