RNA integrity check
A nanolitre of RNA is separated in a microfluidic channel or a prepared tape against a sizing ladder and detected by an intercalating dye, giving an electropherogram, a 28S/18S ratio, a DV200 and an integrity score from one or two microlitres of sample in about half an hour for a dozen samples.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Bioanalyzer / TapeStation
A nanolitre of RNA is separated in a microfluidic channel or a prepared tape against a sizing ladder and detected by an intercalating dye, giving an electropherogram, a 28S/18S ratio, a DV200 and an integrity score from one or two microlitres of sample in about half an hour for a dozen samples.
How it differs from the standard method
- The chip chemistry sets the working concentration range, roughly 5–500 ng/µL for a standard RNA chip and 50–5000 pg/µL for the high-sensitivity version, and a sample outside that range gives a number that describes the loading rather than the RNA.
- Needs a sizing ladder and an intact lower marker in every well: without them there is a trace, but no sizes, no ratio and no score.
- Heat-denature at 70 °C for two minutes and snap-cool before loading, even though the gel-dye matrix contains a denaturant, or secondary structure moves part of each ribosomal peak out of its integration window.
- Returns DV200 as well as a score, which is what makes it the instrument of choice for FFPE and other fragmented material.
- Consumes very little sample, which matters when the whole prep is a few hundred nanograms.
1The readout is a size distribution, not a number
Every other measurement you make on a tube of RNA returns one value: so many nanograms per microlitre, such a ratio of absorbances. An integrity check returns a shape.
2Why the big band goes first
RNases and chemical hydrolysis cut more or less at random along the chain, so the chance that a molecule survives intact falls exponentially with its length. The 28S is two and a half times the length of the 18S, so it takes two and a half times as many hits, and it is gone long before the 18S is.
3What an integrity score adds
The published integrity score (the RNA Integrity Number on a Bioanalyzer, and the equivalent on other platforms) is computed from the whole electropherogram by an algorithm trained on a large set of samples that experienced users had ranked. It takes the ribosomal peak heights and areas, the total area, the height of the lower marker, and crucially the area of the region between the small-RNA cluster and the 18S peak: the fast region, where the first degradation products land.
28S and 18S
About half and a quarter of the mass of total RNA. Their ratio starts near 2 and falls as the longer one is cut, which makes it a late and coarse indicator.
The fast region
The stretch between the small-RNA cluster and the 18S peak. Degradation products appear here first, which is why a score computed from the whole trace is more sensitive than a ratio.
One number, with caveats
Comparable only between runs done the same way, on samples of the same kind, loaded inside the instrument's concentration range.
4A low score can be the measurement rather than the sample
Before you believe a bad number, rule out the three ways the measurement itself produces one. Overloading is the commonest: push a sample well past the top of the chip's concentration range and the tall 28S peak saturates the detector, so its area is clipped, the ratio flattens and the score falls on RNA that is perfectly intact.
5What a bad score does downstream
- Poly-A selection for a sequencing library pulls on the 3′ tail, so a degraded sample contributes only its 3′-most fragments. The library reads 3′-biased, coverage along each transcript slopes, and gene-level counts shift in a way that correlates with integrity across your samples, which is how integrity becomes a batch effect rather than a random loss.
- Ribosomal depletion is more forgiving than poly-A selection, because it does not depend on an intact 3′ end, which is why it is the route taken for FFPE and other degraded material.
- A qPCR amplicon of 70–120 bases survives a lot of degradation; one of 400 bases does not, so a degraded sample reports lower expression for long amplicons than for short ones. If the reference gene's amplicon is short and the target's is long, degradation looks exactly like regulation.
- Reverse transcription primed with oligo(dT) is most sensitive of all, since it starts from the 3′ end and has to run the length of the transcript; random priming is more robust on fragmented RNA.
- If every sample in a comparison has the same, moderate score, a short-amplicon assay can still be perfectly sound. What ruins an experiment is a spread of integrity across groups.
6Judge the score against the sample
There is no universal threshold, because the best achievable score depends on where the RNA came from. Cultured cells lysed straight into a chaotrope routinely score above 9.
- Fast region
- The part of the trace between the small-RNA cluster and the 18S peak. In an intact sample it is nearly empty; degradation products land there before the ribosomal peaks visibly shrink, which is why a score computed from the whole trace is more sensitive than the 28S/18S ratio.
- DV200
- The percentage of RNA mass longer than 200 nucleotides. Used instead of an integrity score for FFPE and other heavily fragmented material, where the score saturates at the bottom of its range but DV200 still ranks samples and predicts how a library will behave.
- Denaturing conditions
- RNA is single-stranded and folds, and a folded molecule migrates as a compact shape rather than as a rod. Chip and capillary systems carry a denaturant in the gel-dye matrix and still expect a 70 °C heat step; a slab gel needs formaldehyde or glyoxal, or the apparent sizes and the ratio are both wrong.
Key terms
Practise these as flashcards →The trace of fluorescence against migration time from a capillary or chip separation, plotted against size once the sizing ladder has calibrated the time axis. It is the primary data; the ratio and the score are summaries of it.
The area of the large ribosomal peak over the small one. Near 2 in intact mammalian total RNA, falling as degradation cuts the longer molecule first. A late, coarse indicator that can also be wrong for non-biological reasons.
RNA Integrity Number: a 1–10 score computed by an algorithm from the whole electropherogram, including the region between the peaks, trained against expert rankings of real traces. The number sequencing facilities specify.
The calibration run alongside the samples that converts migration time into size. Without it the instrument cannot place the integration windows or assign a score, so the trace becomes a picture rather than a measurement.
A short fragment spiked into every well at a known concentration, used to align traces and to normalise signal between samples. Its height is one of the features the integrity algorithm uses, and a missing marker invalidates the run.
The inter-peak stretch where early degradation products appear. Watching it is how you catch degradation before the bands look wrong.
Clipping of the tallest peak when too much sample is loaded. It lowers the apparent ratio and the score without anything being wrong with the RNA, and is the first thing to rule out when a good prep suddenly scores badly.
An agarose gel run with formaldehyde, or with the sample denatured by glyoxal, so RNA migrates by length rather than by shape. A native gel can make intact RNA look degraded and cannot be compared between runs.
When to use it
- Before committing RNA to a sequencing library, where the facility specifies an integrity threshold and the library chemistry depends on it.
- Before an RT-qPCR experiment with long amplicons or oligo(dT) priming, and whenever samples from different collection batches are to be compared.
- As the acceptance check on a new extraction protocol, a new tissue or a new person doing the prep, since degradation is the failure a concentration reading cannot see.
- When triaging archival or FFPE material, where the question is not the score but whether DV200 is high enough for a degraded-input library.
Compared with alternatives
The step that decides what this measurement will find. Almost every bad integrity trace is an account of how the sample was collected and handled, not of the instrument.
The same separation on a slab gel. Prefer it when you want a quick yes/no on gross degradation without an instrument, remembering that RNA needs a denaturing gel.
The companion measurement: concentration and purity. It cannot see integrity, and this cannot see concentration accurately, so both are needed.
The other place a trace like this is read, for library size distribution and adapter dimers rather than for RNA integrity, on the same instrument.
The main consumer of the number: integrity sets the library chemistry, the input mass and whether the results can be compared across batches.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
A nanolitre of RNA is separated in a microfluidic channel or a prepared tape against a sizing ladder and detected by an intercalating dye, giving an electropherogram, a 28S/18S ratio, a DV200 and an integrity score from one or two microlitres of sample in about half an hour for a dozen samples.
- The chip chemistry sets the working concentration range, roughly 5–500 ng/µL for a standard RNA chip and 50–5000 pg/µL for the high-sensitivity version, and a sample outside that range gives a number that describes the loading rather than the RNA.
- Needs a sizing ladder and an intact lower marker in every well: without them there is a trace, but no sizes, no ratio and no score.
- Heat-denature at 70 °C for two minutes and snap-cool before loading, even though the gel-dye matrix contains a denaturant, or secondary structure moves part of each ribosomal peak out of its integration window.
- Returns DV200 as well as a score, which is what makes it the instrument of choice for FFPE and other fragmented material.
- Consumes very little sample, which matters when the whole prep is a few hundred nanograms.