rRNA depletion strategies

Ribosomal RNA (rRNA) can get in the way, so here are some methods to take it away! Ribosomes (the protein-making machinery in cells) are made up of proteins and rRNA. Cells have a lot of ribosomes (because they need to make a lot of proteins). So cells have a lot of rRNA. Like, a lot a lot (~90% of cellular RNA). And, when it comes to creating sequencing libraries, want that rRNA you do not!

There are a couple of main methods to remove it that rely on using DNA probes complementary to rRNA sequences. These probes bind to the rRNA and allow you to selectively degrade it (e.g. with RNaseH that cuts RNA/DNA duplexes) or physically remove it (e.g. with biotinylated probes you can get to stick to magnetic beads) - this is sometimes referred to as a form of subtractive hybridization.

Kits using kits using RNaseH include NEBNext rRNA depletion, Kapa RiboErase, and Takara/Clontech’s RiboGone. And kits using biotinylated probes & magnetic beads include Illumina’s RiboZero, Qiagen GeneRead rRNA depletion, and Lexogen RiboCop.

When doing it, there are some special considerations that make it so the nuclease-mediated depletion methods are less than ideal (they can cause bias and inaccuracy when it comes to exact ribosome positions). More on that here:

So it’s recommended you use the subtractive hybridization method instead. I’m using depletion probes based on this paper:

And this paper talks about a software program you can use to design optimal custom probes for your experiment to maximize the rRNA minimizing!

There’s also, for example, a strategy that uses probes to block Reverse Transcriptase & prevent it from making cDNA copies of rRNA.

Other strategies enrich for the thing of interest (such as mRNA) instead of depleting things of non-interest. For example, NEBNext® Poly(A) mRNA Magnetic Isolation Module uses oligoDT-labeled beads to capture mRNAs by their polyA tails. Of course, that strategy only works if all you care about is mRNA. And, even then, you will miss parts of mRNAs that aren’t intact.

With any method, there’s likely still going to be some rRNA contamination. But that’s okay because we can filter them out of the sequencing data. We just don’t want to waste a ton of reads reading stuff we filter out! Instead, we’d rather get more reads of the stuff we care about.

[Kveldrunari]

Thank you so much for making these videos! A little smarter every day