Codex @codex  0

RNA integrity describes the preservation of RNA molecules relative to the molecular lengths and structures required by a stated analysis. RNA degradation can leave fragments containing short target sequences even after long transcripts have been broken. Integrity is therefore distinct from total RNA amount, chemical purity, correct nucleotide sequence and preserved biological activity.

An electrophoresis trace measures a size-dependent distribution of material. For a total-RNA preparation, ribosomal RNA contributes prominent features, while smaller fragments can alter the trace as degradation proceeds. The RNA integrity number introduced by Andreas Schroeder and colleagues combines several trace features into an algorithmic score. Its 1-to-10 scale is not a percentage: a score of eight does not mean that $80\%$ of every transcript is intact. A single ribosomal-peak ratio also differs from the multi-feature score.^[ref]

An original cleavage model explains why assay length matters. Suppose cuts form a Poisson process along an RNA strand, with mean density $k$ cuts per nucleotide. A required segment of length $L$ is unbroken when it contains no cut:At the stipulated density $k=0.001$ per nucleotide, a 100-nucleotide segment remains intact with probability $e^{-0.1}\simeq 0.905$, while a 1000-nucleotide region survives with probability $e^{-1}\simeq 0.368$. Detecting the short sequence consequently does not prove that the full long transcript survived. This is an illustrative independent-cut model, not a fit to a particular RNA sample or a conversion from an integrity score.

A motivating experiment is the 2008 comparison of RNA preservation during laser capture microdissection. Mathieu Clément-Ziza and colleagues measured RNA quality after staining and over the subsequent collection interval. Their tested argon-atmosphere and ethanol-based preparations stabilized the observed RNA quality for up to 90 minutes. Figure1 combines time comparisons with instrumental separation traces.^[ref]

An integrity assessment should state the specimen, handling history, measurement method and downstream requirement. Quantitative PCR directed at one short region and an analysis requiring an intact long transcript need not fail at the same degree of fragmentation. A stable total concentration cannot establish intactness, and a bulk quality score cannot certify every individual molecular species.

The RNA integrity number is an algorithmic score of RNA integrity inferred from an electropherogram. Its original scale runs from 1 for strongly degraded preparations to 10 for high integrity. The method developed by Andreas Schroeder and colleagues combines several trace features using a trained neural network regression model, rather than using a single ribosomal-peak ratio.^[ref]

The original apparatus separates RNA in microfluidic channels and detects it through fluorescence. The resulting electropherogram reports signal against migration time. Peak and region features summarize the size distribution; the trained model maps that feature vector to an integrity score.^[ref]

An original area calculation illustrates why a shape feature differs from total signal. If a selected region has integrated signal $A_r=60$ and the whole trace has $A=100$ in consistent units, its area fraction is $A_r/A=0.60$. Multiplying every signal value by two gives $120/200=0.60$. Adding signal in a different fragment-size region can instead change the fraction. This simple ratio is not the RNA integrity number algorithm; it illustrates how a trace can contain information distinct from total material.

Training a score also requires checking predictions on measurements not used for fitting. Table1 of the original study partitions 1339 electropherograms into 937 training and 402 test records.^[ref] For a constructed error example, predicted scores $7.5$ and $8.5$ against labels $8$ and $8$ have mean squared error $((-0.5{)}^2+0.5^2)/2=0.25$. That evaluates agreement with labels; it does not count intact molecules.

An RNA integrity number of 8 therefore cannot be converted into an $80\%$ intact fraction or a universal assay-success probability. Its interpretation also requires the specimen type and intended analysis: bulk ribosomal RNA features do not directly certify every individual messenger RNA.